`, `

`, ``, etc.)\n- **Automatic styling** - Beautiful design out of the box\n- **CDN delivery** - Single `` tag, no build step\n\n### Rules\n\n❌ **NEVER use CSS class names** (no `className` attributes)\n❌ **NEVER use TailwindCSS, Bootstrap, or other class-based frameworks**\n✅ **ALWAYS use semantic HTML elements**\n✅ **Use inline styles sparingly** - Only for dynamic/conditional styling\n\n### Semantic HTML Elements\n\nUse these semantic elements - Pico styles them automatically:\n\n**Layout:**\n- `` - Main content container\n- `` - Self-contained content (cards, sections)\n- `` - Thematic grouping of content\n- ``, `` - Page/section headers and footers\n- `` - Sidebar content\n\n**Typography:**\n- `` through ` ` - Headings\n- `` - Paragraphs\n- `` - Bold/important text\n- `` - Italic/emphasized text\n- `` - Fine print\n- `` - Highlighted text\n- `` - Inline code\n- ` ` - Code blocks\n- `` - Keyboard input / tags\n\n**Lists:**\n- ``, ``, `` - Lists\n- ``, ``, ` ` - Definition lists\n\n**Forms:**\n- ``, ``, `<select>` - Form inputs\n- `<button>` - Buttons\n- `<label>` - Form labels\n- Use `aria-busy` attribute for loading states on buttons\n\n**Links & Media:**\n- `<a>` - Links\n- `<time>` - Dates and times\n- `<figure>`, `<figcaption>` - Images with captions\n\n### Example Patterns\n\n**Card layout:**\n```tsx\n<article>\n <h2>Card Title</h2>\n <p>Card content goes here.</p>\n <button>Action</button>\n</article>\n```\n\n**Form with loading state:**\n```tsx\n<div style={{ display: 'flex', gap: '1rem' }}>\n <input type=\"text\" placeholder=\"Enter value\" style={{ flex: 1 }} />\n <button disabled={loading} aria-busy={loading}>\n Submit\n </button>\n</div>\n```\n\n**Error message:**\n```tsx\n{error && <mark>{error}</mark>}\n```\n\n**Tags/badges:**\n```tsx\n<kbd>Tag 1</kbd> <kbd>Tag 2</kbd>\n```\n\n**Conditional styling:**\n```tsx\n<span style={checked ? { textDecoration: 'line-through', opacity: 0.6 } : undefined}>\n Task text\n</span>\n```\n\n### Loading Pico CSS\n\nAlready included in `frontend/index.html`:\n\n```html\n<link href=\"https://cdn.jsdelivr.net/npm/@picocss/pico@2.1.1/css/pico.classless.min.css\" rel=\"stylesheet\" />\n```\n\n## Shared Code (`shared/`)\n\n**Types** (`shared/types.ts`):\n- All TypeScript interfaces shared between frontend/backend\n- Notion data structures\n- API request/response types\n- Common data models\n\n**Utils** (`shared/utils.ts`):\n- Pure functions that work in both Deno (backend) and browser (frontend)\n- ❌ NEVER use Deno APIs (won't work in browser)\n- ✅ Only use standard JavaScript/TypeScript\n\n## Environment Variables\n\nRequired environment variables for this val:\n\n- `NOTION_API_KEY` - Notion integration token\n- `FIRECRAWLER_API_KEY` - Firecrawler service (if used)\n- Additional service API keys as needed\n\n**Access via**: `Deno.env.get('VAR_NAME')`\n\n**NEVER hardcode secrets** - always use environment variables\n\n### Logging Configuration\n\nControl log verbosity via `LOG_LEVEL` environment variable:\n\n| Level | Description |\n|-------|-------------|\n| `error` | Only errors |\n| `warn` | Errors and warnings |\n| `info` | Errors, warnings, and summary info (default, recommended for production) |\n| `debug` | All logs including verbose per-item details (useful for development) |\n\n**Example**:\n```bash\n# Production (quiet, only summaries)\nLOG_LEVEL=info\n\n# Development (verbose, all details)\nLOG_LEVEL=debug\n```\n\nThe logger utility is in `backend/utils/logger.ts` and provides: `logger.error()`, `logger.warn()`, `logger.info()`, `logger.debug()`.\n\n### Notion Property Overrides\n\nOverride default property names via env vars. Defaults are in `/backend/utils/notionConfig.ts`.\n\n**Pattern**: `TODOS_PROP_{SCREAMING_SNAKE_KEY}`\n\n**Logic**:\n- Not set → use default\n- Empty string → disable (null)\n- Set to value → use that value\n\n**Todos Database Properties** (`TODOS_PROP_*`):\n| Env Var | Default | Key |\n|---------|---------|-----|\n| TODOS_PROP_NAME | Name | name |\n| TODOS_PROP_SOURCE_BLOCK_ID | Source block ID | sourceBlockId |\n| TODOS_PROP_SOURCE_BLOCK_URL | Source block URL | sourceBlockUrl |\n| TODOS_PROP_SOURCE_PAGE_URL | Source page URL | sourcePageUrl |\n| TODOS_PROP_SOURCE_LAST_EDITED_TIME | Source last edited time | sourceLastEditedTime |\n| TODOS_PROP_SOURCE_AUTHOR | Source author | sourceAuthor |\n| TODOS_PROP_SOURCE_TEXT | Source text | sourceText |\n| TODOS_PROP_OWNER | Owner | owner |\n| TODOS_PROP_OTHER_PEOPLE | Other people | otherPeople |\n| TODOS_PROP_DUE_DATE | Due date | dueDate |\n| TODOS_PROP_LINKS | Links | links |\n| TODOS_PROP_STATUS | (disabled) | status |\n\n**Examples**:\n```bash\n# Override property names\nTODOS_PROP_DUE_DATE=Deadline\n\n# Enable checkbox status sync\n# - Checked todos → \"Done\" (or first matching done option)\n# - Unchecked todos → first non-done option (e.g., \"Not started\")\n# Auto-detects property type (status, select, checkbox) and finds options\nTODOS_PROP_STATUS=Status\n\n# Disable a property\nTODOS_PROP_OWNER=\n```\n\n## Data Validation\n\nValidate request data in **controllers**, not routes:\n\n```typescript\nif (!pageId || typeof pageId !== 'string') {\n return {\n success: false,\n data: null,\n error: \"Invalid pageId\",\n details: \"pageId must be a non-empty string\"\n };\n}\n```\n\n**Check:**\n- Required fields exist\n- Types are correct\n- Formats are valid\n- Values are within expected ranges\n\n**Return helpful error messages** with `details` field\n\n## Response Filtering\n\n**Filter sensitive data in controllers** before returning:\n\n- Remove button properties from Notion data\n- Don't expose internal IDs or tokens\n- Clean up API responses before sending to frontend\n- Strip unnecessary metadata\n\n**Example:**\n```typescript\n// Remove sensitive properties\nconst filtered = {\n ...notionData,\n properties: Object.entries(notionData.properties)\n .filter(([key, value]) => value.type !== 'button')\n .reduce((acc, [key, value]) => ({ ...acc, [key]: value }), {})\n};\n```\n\n## Questions to Ask User\n\nWhen the user requests a new feature, clarify:\n\n1. **Is this a webhook handler or API endpoint?**\n - Webhook → `routes/tasks/`\n - API → `routes/api/`\n\n2. **What external service needs integration?**\n - New service → create `backend/services/[name].ts`\n\n3. **What data needs to be saved back to Notion?**\n - Update Notion → use/extend `notionService.ts`\n\n4. **Does this need authentication?**\n - Protected → apply `authCheck` middleware\n\n5. **Is this frontend-facing?**\n - Yes → create React component in `frontend/components/`\n\n## Common Patterns Checklist\n\nBefore implementing, verify:\n\n- ✅ Following 3-layer architecture (Route → Controller → Service)\n- ✅ Routes only handle HTTP, no business logic\n- ✅ Controllers return standardized `{success, data, error}` structure\n- ✅ Services handle external API calls\n- ✅ Never skipping layers\n- ✅ TypeScript types defined in `shared/types.ts`\n- ✅ Environment variables used for secrets\n- ✅ Error handling at each layer\n- ✅ Validation in controllers\n- ✅ Sensitive data filtered before returning\n\n## Integration with AGENTS.md\n\n**Use AGENTS.md for:**\n- Val Town platform requirements\n- Deno-specific patterns\n- Standard library usage\n- React configuration\n- Import patterns (esm.sh)\n- Platform limitations\n\n**Use CLAUDE.md (this file) for:**\n- This val's architecture\n- MVC layer separation\n- Notion integration patterns\n- External service integration\n- File organization\n- Response formats\n\n## Entry Point\n\n`main.http.tsx` is the HTTP entry point:\n- Exports `app.fetch` for Val Town\n- Uses Hono for routing\n- Serves static frontend assets\n- Mounts API routes at `/api`\n- Includes error handler\n\n**Always mount new routes in main.http.tsx:**\n```typescript\nimport { taskRoutes } from './backend/routes/tasks/index.ts';\n\napp.route('/tasks', taskRoutes);\n```\n\n## Hono Framework Patterns\n\nThis val uses **Hono** as the web framework. Follow these patterns:\n\n### Importing Hono\n\n```typescript\nimport { Hono } from \"npm:hono@4\";\nimport { Context, Next } from \"npm:hono@4\";\n```\n\n**Note**: Use `npm:` specifier for backend-only packages (better performance, modern Deno way).\n\n### Creating Route Modules\n\nWhen creating route modules (e.g., `backend/routes/api/pages.ts`):\n\n```typescript\nimport { Hono } from \"npm:hono@4\";\nimport * as pageController from '../../controllers/pageController.ts';\n\nconst pages = new Hono();\n\n// GET /api/pages/:id\npages.get('/:id', async (c) => {\n const pageId = c.req.param('id');\n const result = await pageController.getPage(pageId);\n\n if (!result.success) {\n return c.json({ error: result.error, details: result.details }, 400);\n }\n\n return c.json(result.data);\n});\n\n// POST /api/pages/:id\npages.post('/:id', async (c) => {\n const pageId = c.req.param('id');\n const body = await c.req.json();\n\n const result = await pageController.updatePage(pageId, body.properties);\n\n if (!result.success) {\n return c.json({ error: result.error, details: result.details }, 400);\n }\n\n return c.json(result.data);\n});\n\nexport default pages;\n```\n\n### Mounting Routes\n\nCreate an index file for each route directory (e.g., `backend/routes/api/index.ts`):\n\n```typescript\nimport { Hono } from \"npm:hono@4\";\nimport pages from './pages.ts';\n\nconst api = new Hono();\n\napi.route('/pages', pages);\n// Add more routes as needed\n\nexport default api;\n```\n\nThen mount in `main.http.tsx`:\n\n```typescript\nimport api from './backend/routes/api/index.ts';\napp.route('/api', api);\n```\n\n### Accessing Request Data\n\n```typescript\n// URL params\nconst id = c.req.param('id');\n\n// Query params\nconst query = c.req.query('q');\n\n// Headers\nconst apiKey = c.req.header('x-api-key');\n\n// JSON body\nconst body = await c.req.json();\n\n// Form data\nconst formData = await c.req.formData();\n```\n\n### Returning Responses\n\n```typescript\n// JSON response\nreturn c.json({ data: 'value' });\n\n// JSON with status code\nreturn c.json({ error: 'Not found' }, 404);\n\n// HTML response\nreturn c.html('<h1>Hello</h1>');\n\n// Plain text\nreturn c.text('Hello');\n\n// Redirect (use this pattern, not Response.redirect)\nreturn new Response(null, { status: 302, headers: { Location: '/new-url' }});\n```\n\n### Middleware\n\nApply middleware to routes:\n\n```typescript\nimport { authCheck } from '../authCheck.ts';\n\n// Apply to single route\npages.post('/:id', authCheck, async (c) => {\n // Protected route logic\n});\n\n// Apply to all routes in this module\npages.use('*', authCheck);\n```\n\n### Error Handler\n\nAlready configured in `main.http.tsx`:\n\n```typescript\napp.onError((err, c) => {\n throw err; // Re-throw to see full stack trace\n});\n```\n\n### Static File Serving\n\nAlready configured in `main.http.tsx`:\n\n```typescript\nimport { serveFile } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n\napp.get(\"/frontend/*\", c => serveFile(c.req.path, import.meta.url));\napp.get(\"/shared/*\", c => serveFile(c.req.path, import.meta.url));\n```\n\n### Route Organization Example\n\n```\nbackend/routes/\n├── api/\n│ ├── index.ts # Mounts all API routes\n│ ├── pages.ts # /api/pages/* endpoints\n│ └── databases.ts # /api/databases/* endpoints\n├── tasks/\n│ ├── index.ts # Mounts all webhook routes\n│ └── notion.ts # /tasks/notion webhook handler\n├── views/\n│ └── dashboard.ts # /views/dashboard page\n└── authCheck.ts # Auth middleware\n```\n",[],[288],"0ac45fe0-0427-11f1-998d-42dde27851f2","CHANGELOG.md","ffe13986-02dc-11f1-8dcf-42dde27851f2",432,["D",1770470115581],"# Changelog\n\n## 2026-02-07\n\n### README Restructure\n\nRewrote README from 1686 lines to ~420 lines with clearer organization:\n- **Quick Start**: Required env vars and properties upfront\n- **Property Mapping**: New dedicated section explaining defaults and overrides\n- **Owner Resolution**: Consolidated priority flow with heading proximity rules\n- **Other People Resolution**: New dedicated section for AI-extracted names\n- **Due Date Resolution**: Simplified explanation\n- **Relation Mapping**: Clear strategy breakdown with Projects-only whitelist note\n- **Configuration Reference**: All env vars in organized tables\n- **Advanced Topics**: Technical details moved to end\n\n### Fix Owner Property Lookup\n\nFixed bug where owner could be added to wrong property when multiple relations target the same database. Previously used `cache.relations.find(r => r.targetDatabaseId === cache.ownerTargetDbId)` which could return the wrong relation if Steels DB appeared before DRI in the array. Now uses `TODOS_PROPERTIES.owner` directly.\n\n## 2026-02-06\n\n### Heading Proximity Scoping for Owner Resolution\n\nHeadings now only apply to uninterrupted stacks of todo blocks directly beneath them. Any non-heading, non-todo block (empty line, paragraph, list item, etc.) resets the heading context to null. This prevents distant, unrelated headings from being incorrectly matched to todos.\n\nAlso added stopword filtering to heading matching to prevent common words like \"the\", \"in\", \"see\", \"what\", \"database\" from matching against contact names.\n\nTightened owner name matching to only allow exact matches or prefix matches (entry starts with search word). Removed reverse prefix matching that allowed \"brad\" to match any contact with a short word like \"b\" or \"br\".\n\n### AI-Extracted Names for Other People (Relation Type Only)\n\nAI now extracts names of people mentioned in todo text (not @mentions) and adds them to the Other People relation when configured as a relation type. For example, \"Taylor to call Jamie R about the project\" will add Jamie Rodriguez to Other People if they exist in the Contacts database.\n\nMatching rules:\n\n- Exact name match first (case-insensitive)\n- Falls back to prefix matching: \"Jamie R\" matches \"Jamie Rodriguez\" (all words must be prefixes)\n- Skipped if ambiguous (multiple matches)\n- Skipped if already in Owner or @mentions\n- Task targets ARE included (e.g., \"call Jamie\" will match Jamie)\n\n## 2025-02-05\n\n### Skip Updates When User Edited Todo Page\n\nWhen a user (not the integration bot) last edited a todo page in the Todos database, sync now skips updating that page to preserve the user's changes. Logs show \"Skipping update for block X: page was edited by user (not bot)\".\n\n### Whitelist Contextual Matching for Projects Only\n\nContextual matching (Strategy 2a/2b/2c) now only applies to the Projects relation. All other relations (Owner, otherPeople, etc.) require explicit @mentions. This prevents source page relations like DRI or Additional Members from polluting the Contacts field.\n\n### Opt-In Relation Loading\n\nOnly relations with explicit `TODOS_PROP_xxx` mappings are loaded from Notion. Unmapped relations are discovered but not loaded, reducing API calls and preventing false positive matches.\n\n### Exclude Owner from Contextual Matching\n\nOwner relation is excluded from contextual matching strategies. Owner resolution uses its dedicated flow: heading match, @mention, or AI extraction.\n\n### Simplify Owner Resolution\n\nRemoved \"contact\" as an owner source. Owner now comes from:\n\n1. Heading match (with AI disambiguation if multiple)\n2. @mention (exact match priority)\n3. AI extraction (validated against owner DB)\n\n### Remove Fuzzy Matching\n\nFuzzy matching code was already removed in a prior session. Relation matching now uses only @mentions (Strategy 1a/1b) and contextual matching (Strategy 2a/2b/2c, Projects only).\n",[],[296],"1cb98422-b360-11f0-bc6f-0224a6c84d84","AGENTS.md","019a26dc-99bf-7269-a8f6-692ef14e0fe6",["D",1761588582800],"You are an advanced assistant specialized in generating Val Town code.\n\n## Core Guidelines\n\n- Ask clarifying questions when requirements are ambiguous\n- Provide complete, functional solutions rather than skeleton implementations\n- Test your logic against edge cases before presenting the final solution\n- Ensure all code follows Val Town's specific platform requirements\n- If a section of code that you're working on is getting too complex, consider refactoring it into subcomponents\n\n## Code Standards\n\n- Generate code in TypeScript or TSX\n- Add appropriate TypeScript types and interfaces for all data structures\n- Prefer official SDKs or libraries than writing API calls directly\n- Ask the user to supply API or library documentation if you are at all unsure about it\n- **Never bake in secrets into the code** - always use environment variables\n- Include comments explaining complex logic (avoid commenting obvious operations)\n- Follow modern ES6+ conventions and functional programming practices if possible\n\n## Types of triggers\n\n### 1. HTTP Trigger\n\n- Create web APIs and endpoints\n- Handle HTTP requests and responses\n- Example structure:\n\n```ts\nexport default async function (req: Request) {\n return new Response(\"Hello World\");\n}\n```\n\nFiles that are HTTP triggers have http in their name like `foobar.http.tsx`\n\n### 2. Cron Triggers\n\n- Run on a schedule\n- Use cron expressions for timing\n- Example structure:\n\n```ts\nexport default async function () {\n // Scheduled task code\n}\n```\n\nFiles that are Cron triggers have cron in their name like `foobar.cron.tsx`\n\n### 3. Email Triggers\n\n- Process incoming emails\n- Handle email-based workflows\n- Example structure:\n\n```ts\nexport default async function (email: Email) {\n // Process email\n}\n```\n\nFiles that are Email triggers have email in their name like `foobar.email.tsx`\n\n\n## Val Town Standard Libraries\n\nVal Town provides several hosted services and utility functions.\n\n### Blob Storage\n\n```ts\nimport { blob } from \"https://esm.town/v/std/blob\";\nawait blob.setJSON(\"myKey\", { hello: \"world\" });\nlet blobDemo = await blob.getJSON(\"myKey\");\nlet appKeys = await blob.list(\"app_\");\nawait blob.delete(\"myKey\");\n```\n\n### SQLite\n\n```ts\nimport { sqlite } from \"https://esm.town/v/stevekrouse/sqlite\";\nconst TABLE_NAME = 'todo_app_users_2';\n// Create table - do this before usage and change table name when modifying schema\nawait sqlite.execute(`CREATE TABLE IF NOT EXISTS ${TABLE_NAME} (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n name TEXT NOT NULL\n)`);\n// Query data\nconst result = await sqlite.execute(`SELECT * FROM ${TABLE_NAME} WHERE id = ?`, [1]);\n```\n\nNote: When changing a SQLite table's schema, change the table's name (e.g., add _2 or _3) to create a fresh table.\n\n### OpenAI\n\n```ts\nimport { OpenAI } from \"https://esm.town/v/std/openai\";\nconst openai = new OpenAI();\nconst completion = await openai.chat.completions.create({\n messages: [\n { role: \"user\", content: \"Say hello in a creative way\" },\n ],\n model: \"gpt-4o-mini\",\n max_tokens: 30,\n});\n```\n\n### Email\n\n```ts\nimport { email } from \"https://esm.town/v/std/email\";\n// By default emails the owner of the val\nawait email({ \n subject: \"Hi\", \n text: \"Hi\", \n html: \"<h1>Hi</h1>\"\n});\n```\n\n## Val Town Utility Functions\n\nVal Town provides several utility functions to help with common project tasks.\n\n### Importing Utilities\n\nAlways import utilities with version pins to avoid breaking changes:\n\n```ts\nimport { parseProject, readFile, serveFile } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n```\n\n### Available Utilities\n\n\n#### **serveFile** - Serve project files with proper content types\n\nFor example, in Hono:\n\n```ts\n// serve all files in frontend/ and shared/\napp.get(\"/frontend/*\", c => serveFile(c.req.path, import.meta.url));\napp.get(\"/shared/*\", c => serveFile(c.req.path, import.meta.url));\n```\n\n#### **readFile** - Read files from within the project:\n\n```ts\n// Read a file from the project\nconst fileContent = await readFile(\"/frontend/index.html\", import.meta.url);\n```\n\n#### **listFiles** - List all files in the project\n\n```ts\nconst files = await listFiles(import.meta.url);\n```\n\n#### **parseProject** - Extract information about the current project from import.meta.url\n\nThis is useful for including info for linking back to a val, ie in \"view source\" urls:\n\n```ts\nconst projectVal = parseProject(import.meta.url);\nconsole.log(projectVal.username); // Owner of the project\nconsole.log(projectVal.name); // Project name\nconsole.log(projectVal.version); // Version number\nconsole.log(projectVal.branch); // Branch name\nconsole.log(projectVal.links.self.project); // URL to the project page\n```\n\nHowever, it's *extremely importing* to note that `parseProject` and other Standard Library utilities ONLY RUN ON THE SERVER.\nIf you need access to this data on the client, run it in the server and pass it to the client by splicing it into the HTML page\nor by making an API request for it.\n\n## Val Town Platform Specifics\n\n- **Redirects:** Use `return new Response(null, { status: 302, headers: { Location: \"/place/to/redirect\" }})` instead of `Response.redirect` which is broken\n- **Images:** Avoid external images or base64 images. Use emojis, unicode symbols, or icon fonts/libraries instead\n- **AI Image:** To inline generate an AI image use: `<img src=\"https://maxm-imggenurl.web.val.run/the-description-of-your-image\" />`\n- **Storage:** DO NOT use the Deno KV module for storage\n- **Browser APIs:** DO NOT use the `alert()`, `prompt()`, or `confirm()` methods\n- **Weather Data:** Use open-meteo for weather data (doesn't require API keys) unless otherwise specified\n- **View Source:** Add a view source link by importing & using `import.meta.url.replace(\"ems.sh\", \"val.town)\"` (or passing this data to the client) and include `target=\"_top\"` attribute\n- **Error Debugging:** Add `<script src=\"https://esm.town/v/std/catch\"></script>` to HTML to capture client-side errors\n- **Error Handling:** Only use try...catch when there's a clear local resolution; Avoid catches that merely log or return 500s. Let errors bubble up with full context\n- **Environment Variables:** Use `Deno.env.get('keyname')` when you need to, but generally prefer APIs that don't require keys\n- **Imports:** Use `https://esm.sh` for npm and Deno dependencies to ensure compatibility on server and browser\n- **Storage Strategy:** Only use backend storage if explicitly required; prefer simple static client-side sites\n- **React Configuration:** When using React libraries, pin versions with `?deps=react@18.2.0,react-dom@18.2.0` and start the file with `/** @jsxImportSource https://esm.sh/react@18.2.0 */`\n- Ensure all React dependencies and sub-dependencies are pinned to the same version\n- **Styling:** Default to using TailwindCSS via `<script src=\"https://cdn.twind.style\" crossorigin></script>` unless otherwise specified\n\n## Project Structure and Design Patterns\n\n### Recommended Directory Structure\n```\n├── backend/\n│ ├── database/\n│ │ ├── migrations.ts # Schema definitions\n│ │ ├── queries.ts # DB query functions\n│ │ └── README.md\n│ └── routes/ # Route modules\n│ ├── [route].ts\n│ └── static.ts # Static file serving\n│ ├── index.ts # Main entry point\n│ └── README.md\n├── frontend/\n│ ├── components/\n│ │ ├── App.tsx\n│ │ └── [Component].tsx\n│ ├── favicon.svg\n│ ├── index.html # Main HTML template\n│ ├── index.tsx # Frontend JS entry point\n│ ├── README.md\n│ └── style.css\n├── README.md\n└── shared/\n ├── README.md\n └── utils.ts # Shared types and functions\n```\n\n### Backend (Hono) Best Practices\n\n- Hono is the recommended API framework\n- Main entry point should be `backend/index.ts`\n- **Static asset serving:** Use the utility functions to read and serve project files:\n ```ts\n import { readFile, serveFile } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n \n // serve all files in frontend/ and shared/\n app.get(\"/frontend/*\", c => serveFile(c.req.path, import.meta.url));\n app.get(\"/shared/*\", c => serveFile(c.req.path, import.meta.url));\n \n // For index.html, often you'll want to bootstrap with initial data\n app.get(\"/\", async c => {\n let html = await readFile(\"/frontend/index.html\", import.meta.url);\n \n // Inject data to avoid extra round-trips\n const initialData = await fetchInitialData();\n const dataScript = `<script>\n window.__INITIAL_DATA__ = ${JSON.stringify(initialData)};\n </script>`;\n \n html = html.replace(\"</head>\", `${dataScript}</head>`);\n return c.html(html);\n });\n ```\n- Create RESTful API routes for CRUD operations\n- Always include this snippet at the top-level Hono app to re-throwing errors to see full stack traces:\n ```ts\n // Unwrap Hono errors to see original error details\n app.onError((err, c) => {\n throw err;\n });\n ```\n\n### Database Patterns\n- Run migrations on startup or comment out for performance\n- Change table names when modifying schemas rather than altering\n- Export clear query functions with proper TypeScript typing\n\n## Common Gotchas and Solutions\n\n1. **Environment Limitations:** \n - Val Town runs on Deno in a serverless context, not Node.js\n - Code in `shared/` must work in both frontend and backend environments\n - Cannot use `Deno` keyword in shared code\n - Use `https://esm.sh` for imports that work in both environments\n\n2. **SQLite Peculiarities:**\n - Limited support for ALTER TABLE operations\n - Create new tables with updated schemas and copy data when needed\n - Always run table creation before querying\n\n3. **React Configuration:**\n - All React dependencies must be pinned to 18.2.0\n - Always include `@jsxImportSource https://esm.sh/react@18.2.0` at the top of React files\n - Rendering issues often come from mismatched React versions\n\n4. **File Handling:**\n - Val Town only supports text files, not binary\n - Use the provided utilities to read files across branches and forks\n - For files in the project, use `readFile` helpers\n\n5. **API Design:**\n - `fetch` handler is the entry point for HTTP vals\n - Run the Hono app with `export default app.fetch // This is the entry point for HTTP vals`\n\n",[],[304],"1cb9830a-b360-11f0-bc6f-0224a6c84d84",".vtignore","019a26dc-99bf-7269-a8f6-6555ae856e94",["D",1761588582800],".git\n.vscode\n.cursorrules\n.DS_Store\nnode_modules\nvendor\nvars.env",[],[311],"0a91ab24-0033-11f1-8a6c-42dde27851f2","utils.ts","019a26dc-99bf-7269-a8f6-5e2af0c9fdd2","019a26dc-99bf-7269-a8f6-63c6fd1d84b1",406,["D",1770035464562],"script","/**\n * Shared utility functions\n * IMPORTANT: These must work in both browser AND Deno (no Deno APIs allowed)\n */\n\n/**\n * Check if a string is a valid Notion page ID format\n */\nexport function isValidPageId(pageId: string): boolean {\n if (!pageId || typeof pageId !== 'string') {\n return false;\n }\n\n // Notion IDs are 32 characters (with or without hyphens)\n const cleanId = pageId.replace(/-/g, '');\n return cleanId.length === 32 && /^[a-f0-9]+$/i.test(cleanId);\n}\n\n/**\n * Normalize a Notion page ID (remove hyphens)\n */\nexport function normalizePageId(pageId: string): string {\n return pageId.replace(/-/g, '');\n}\n\n/**\n * Format a Notion page ID with hyphens in standard format\n * 8-4-4-4-12\n */\nexport function formatPageId(pageId: string): string {\n const clean = normalizePageId(pageId);\n return `${clean.slice(0, 8)}-${clean.slice(8, 12)}-${clean.slice(12, 16)}-${clean.slice(16, 20)}-${clean.slice(20)}`;\n}\n\n/**\n * Extract plain text from Notion rich text array\n */\nexport function getPlainText(richText: any[]): string {\n if (!Array.isArray(richText)) {\n return '';\n }\n return richText.map(text => text.plain_text || '').join('');\n}\n\n/**\n * Safely get a property value from Notion page properties\n */\nexport function getPropertyValue(properties: any, propertyName: string): any {\n const property = properties[propertyName];\n if (!property) {\n return null;\n }\n\n // Extract value based on property type\n switch (property.type) {\n case 'title':\n return getPlainText(property.title);\n case 'rich_text':\n return getPlainText(property.rich_text);\n case 'number':\n return property.number;\n case 'select':\n return property.select?.name || null;\n case 'multi_select':\n return property.multi_select?.map((s: any) => s.name) || [];\n case 'date':\n return property.date?.start || null;\n case 'checkbox':\n return property.checkbox;\n case 'url':\n return property.url;\n case 'email':\n return property.email;\n case 'phone_number':\n return property.phone_number;\n case 'status':\n return property.status?.name || null;\n default:\n return property;\n }\n}\n\n/**\n * Format an ISO timestamp to relative time\n * Examples: \"2 hours ago\", \"Yesterday\", \"3 days ago\"\n *\n * @param isoString - ISO 8601 timestamp string\n * @returns Human-readable relative time string\n */\nexport function formatRelativeTime(isoString: string): string {\n try {\n const date = new Date(isoString);\n const now = new Date();\n const diffMs = now.getTime() - date.getTime();\n const diffMinutes = Math.floor(diffMs / (1000 * 60));\n const diffHours = Math.floor(diffMs / (1000 * 60 * 60));\n const diffDays = Math.floor(diffMs / (1000 * 60 * 60 * 24));\n\n // Less than 1 minute\n if (diffMinutes < 1) {\n return \"Just now\";\n }\n\n // Less than 1 hour\n if (diffMinutes < 60) {\n return `${diffMinutes} minute${diffMinutes === 1 ? '' : 's'} ago`;\n }\n\n // Less than 24 hours\n if (diffHours < 24) {\n return `${diffHours} hour${diffHours === 1 ? '' : 's'} ago`;\n }\n\n // Less than 2 days (show \"Yesterday\")\n if (diffDays === 1) {\n return \"Yesterday\";\n }\n\n // Less than 7 days\n if (diffDays < 7) {\n return `${diffDays} days ago`;\n }\n\n // More than 7 days - show date\n return date.toLocaleDateString('en-US', {\n month: 'short',\n day: 'numeric',\n year: date.getFullYear() !== now.getFullYear() ? 'numeric' : undefined\n });\n\n } catch (err) {\n return isoString; // Fallback to original string\n }\n}\n\n/**\n * Compose a display string from todo components\n * Format: \"[Owner] to [task] by [due date]\"\n *\n * This function takes pre-extracted string values. The caller is responsible\n * for extracting owner name from the appropriate property type (people or relation)\n * using the active property mappings from TODOS_PROPERTIES.\n *\n * @param ownerName - Extracted owner name (optional)\n * @param task - Task description (required)\n * @param dueDate - Due date as Date or ISO string (optional)\n * @returns Formatted display string\n */\nexport function composeDisplayString(\n ownerName: string | null,\n task: string,\n dueDate: Date | string | null\n): string {\n let result = '';\n\n // Owner prefix\n if (ownerName) {\n result += ownerName + ' to ';\n }\n\n // Task (lowercase first letter if owner prefix exists)\n if (task) {\n if (ownerName) {\n result += task.charAt(0).toLowerCase() + task.slice(1);\n } else {\n result += task;\n }\n }\n\n // Due date suffix\n if (dueDate) {\n const date = typeof dueDate === 'string' ? new Date(dueDate) : dueDate;\n const formatted = date.toLocaleDateString('en-US', {\n month: 'long',\n day: 'numeric',\n year: 'numeric'\n });\n result += ' by ' + formatted;\n }\n\n return result;\n}\n",[],[327,318],"shared","da95474c-0388-11f1-a576-42dde27851f2","types.ts","019a26dc-99bf-7269-a8f6-5917ad49537e",424,["D",1770402174256],28,"/**\n * Shared TypeScript types and interfaces\n * These work in both frontend (browser) and backend (Deno)\n */\n\n/**\n * Standard response format for all controller functions\n */\nexport interface ControllerResponse<T = any> {\n success: boolean;\n data: T | null;\n error: string | null;\n details?: string;\n}\n\n/**\n * Notion page response (simplified)\n * For full types, use @notionhq/client types in backend\n */\nexport interface NotionPage {\n id: string;\n object: 'page';\n created_time: string;\n last_edited_time: string;\n properties: Record<string, any>;\n parent: {\n type: string;\n database_id?: string;\n page_id?: string;\n };\n url: string;\n}\n\n/**\n * Notion database response (simplified)\n */\nexport interface NotionDatabase {\n id: string;\n object: 'database';\n created_time: string;\n last_edited_time: string;\n title: Array<{\n type: string;\n plain_text: string;\n }>;\n properties: Record<string, any>;\n}\n\n/**\n * Notion webhook payload\n */\nexport interface NotionWebhookPayload {\n type: 'page' | 'database';\n page?: NotionPage;\n database?: NotionDatabase;\n}\n\n/**\n * Common page properties (extend as needed)\n */\nexport interface PageProperties {\n [key: string]: any;\n}\n\n/**\n * API request/response types\n */\nexport interface UpdatePageRequest {\n properties: PageProperties;\n}\n\nexport interface CreatePageRequest {\n databaseId: string;\n properties: PageProperties;\n}\n\nexport interface PageResponse {\n page: NotionPage;\n}\n\nexport interface PagesListResponse {\n pages: NotionPage[];\n}\n\n/**\n * Faviconizer API response\n */\nexport interface FaviconizerResponse {\n domain: string;\n faviconUrl: string;\n success: boolean;\n source: string;\n contentType: string;\n}\n\n/**\n * Favicon update result\n */\nexport interface FaviconUpdateResult {\n success: boolean;\n iconUpdated?: boolean;\n errorNoteAdded?: boolean;\n successNoteAdded?: boolean;\n error?: string;\n details?: string;\n}\n\n/**\n * Todo search result\n */\nexport interface TodoSearchResult {\n success: boolean;\n blocksFound: number;\n blocksNew?: number;\n blocksUpdated?: number;\n blocksSkipped?: number;\n blockIds: string[];\n error?: string;\n details?: string;\n}\n\n/**\n * Sync metadata for blob storage optimization\n */\nexport interface SyncMetadata {\n synced: boolean; // true = already synced, false = needs sync\n target_page_id?: string; // Cached ID of page in todos database (avoids query on update)\n archived?: boolean; // true = Notion page was archived by user\n}\n\n/**\n * Link mention with optional parent database context\n * Used for dynamic relation mapping\n */\nexport interface LinkMention {\n text: string;\n url: string;\n pageId?: string | null; // Extracted Notion page ID (normalized, no hyphens)\n parentDatabaseId?: string | null; // Parent database ID if page is in a database\n}\n\n/**\n * Reduced/transformed Notion block structure\n */\nexport interface ReducedNotionBlock {\n todo_string: string;\n block_id: string;\n block_url: string;\n page_url?: string; // Source page URL where the todo block was found\n parent_id?: string | null; // Parent block ID for traversal (null if parent is page)\n has_children?: boolean; // true = block has nested children\n preceding_heading?: string | null; // Closest heading before this block (for context matching)\n last_edited_time: string;\n author?: { // Block creator (from created_by)\n id: string;\n name: string | null;\n email: string | null;\n } | null;\n people_mentions: Array<{ // @mentions in block text (all go to Other people)\n id: string;\n name: string;\n email: string | null;\n }>;\n date_mentions: string[];\n link_mentions: LinkMention[];\n source_page_database_id?: string | null; // Parent database of the source page (for contextual mapping, only set when parent.type === 'database_id')\n source_page_relations?: Record<string, string[]> | null; // Relation values from source page (propName -> pageIds[]) for contextual matching\n sync_metadata?: SyncMetadata; // Optional sync tracking metadata\n checked?: boolean; // to_do block checkbox state (true = checked, undefined = unchecked or not a to_do)\n}\n\n/**\n * Blob storage save result\n */\nexport interface BlobSaveResult {\n success: boolean;\n key: string | null;\n error: string | null;\n}\n\n/**\n * Todo save to Notion result\n */\nexport interface TodoSaveResult {\n success: boolean;\n totalBlobs: number;\n pagesCreated: number;\n pagesUpdated: number;\n pagesSkipped: number;\n pagesFailed: number;\n failures: Array<{\n blockId: string;\n error: string;\n }>;\n error?: string;\n details?: string;\n}\n\n/**\n * Property category for organizational grouping\n * - system: Infrastructure properties (blockId, urls, timestamps)\n * - core: Essential todo properties (name, dueDate, status, owner)\n * - enrichment: Additional context (author, links, summary, projects)\n */\nexport type PropertyCategory = 'system' | 'core' | 'enrichment';\n\n/**\n * Property metadata for declarative configuration\n * Defines what each property does and how it can be discovered\n */\nexport interface PropertyMetadata {\n displayName: string; // Human-readable name for UI (e.g., \"Due date\", \"Block ID\")\n purpose: string; // Human-readable description\n category: PropertyCategory; // Organizational grouping (system/core/enrichment)\n required: boolean; // Is this property required for basic functionality?\n autodiscoverable: boolean; // Can this be found automatically by type or name pattern?\n expectedTypes: string[]; // Notion property types this can be (e.g., [\"title\"], [\"status\", \"select\", \"checkbox\"])\n namePatterns?: string[]; // Optional name patterns for autodiscovery (e.g., [\"due\", \"deadline\"])\n}\n\n/**\n * Property status in health check response (property-centric)\n */\nexport interface PropertyStatus {\n key: string; // Internal property key (e.g., \"name\", \"status\")\n displayName: string; // Human-readable name for UI (e.g., \"Due date\")\n purpose: string; // Human-readable description\n category: PropertyCategory; // Organizational grouping (system/core/enrichment)\n required: boolean; // Is this property required?\n envVar: string; // Env var name (e.g., \"TODOS_PROP_DUE_DATE\")\n envValue: string | null; // Current env var value (if set)\n defaultName: string | null; // System default property name\n foundName: string | null; // Property name found in database (if any)\n foundType: string | null; // Notion property type (if found)\n candidates?: string[]; // Suggested property names (if missing/disabled)\n // Relation metadata (only populated when foundType === 'relation')\n relationTargetDbId?: string; // Target database ID\n relationTargetDbName?: string | null; // Target database name\n relationPageCount?: number; // Number of pages in target database\n}\n\n/**\n * Database schema property (from Notion API)\n */\nexport interface DatabaseSchemaProperty {\n name: string;\n type: string;\n id: string;\n}\n\n/**\n * Project info from Val Town\n */\nexport interface ProjectInfo {\n name: string;\n username: string;\n version: number;\n branch: string;\n url: string | null;\n}\n\n/**\n * Connection status type\n */\nexport type ConnectionStatus = 'connected' | 'error' | 'not_configured';\n\n/**\n * Notion API connection status\n */\nexport interface NotionConnection {\n envVar: string;\n status: ConnectionStatus;\n integrationName?: string;\n error?: string;\n}\n\n/**\n * Database connection status\n */\nexport interface DatabaseConnection {\n envVar: string;\n status: ConnectionStatus;\n id: string | null;\n name: string | null;\n url: string | null;\n error?: string;\n}\n\n/**\n * OpenAI connection status\n */\nexport interface OpenAIConnection {\n envVar: string;\n status: ConnectionStatus;\n error?: string | null;\n}\n\n/**\n * All service connections\n */\nexport interface Connections {\n notion: NotionConnection;\n database: DatabaseConnection;\n openai: OpenAIConnection;\n}\n\n/**\n * Category display info\n */\nexport interface CategoryInfo {\n title: string;\n description: string;\n}\n\n/**\n * Health check response (property-centric)\n */\nexport interface HealthCheckResponse {\n project: ProjectInfo;\n timestamp: string;\n connections: Connections;\n properties: PropertyStatus[];\n categories: Record<PropertyCategory, CategoryInfo>;\n relations: Array<{\n propertyName: string;\n targetDatabaseId: string;\n pageCount: number;\n }>;\n settings: Array<{\n key: string;\n envVar: string;\n value: string | null;\n defaultValue: string | null;\n description: string;\n }>;\n displayFormula: string | null; // Notion formula for display string (copy-paste helper)\n}\n\n/**\n * Legacy: Property mappings status for a database\n * @deprecated Use HealthCheckResponse instead\n */\nexport interface DatabasePropertyMappings {\n status: 'ok' | 'error' | 'not_connected';\n properties: Array<{\n key: string;\n configured: string;\n found: boolean;\n status: 'ok' | 'missing';\n }>;\n actualProperties: string[];\n}\n\n/**\n * Relation property discovered from todos database schema\n * Used for dynamic relation mapping\n */\nexport interface RelationPropertyMapping {\n propertyName: string; // Name of relation property in todos DB (e.g., \"Projects db\")\n targetDatabaseId: string; // The database this relation points to (normalized, no hyphens)\n}\n\n/**\n * Result of relation matching for a single block\n */\nexport interface RelationMatchResult {\n propertyName: string; // Relation property name to populate\n pageIds: string[]; // Page IDs to link\n matchType: 'mention' | 'contextual' | 'ai'; // How the match was made\n}\n\n/**\n * Structured result from AI summary generation\n * Contains both the summary text and extracted owner information\n */\nexport interface SummaryResult {\n summary: string;\n ownerName: string | null;\n ownerSource: null; // Deprecated - owner source is determined by system, not AI\n dueDate: string | null; // ISO format, e.g., \"2025-01-22\"\n mentionedPeople: string[]; // Names of people mentioned in todo text (not @mentions)\n}\n\n/**\n * Resolved owner value for property mapping\n * Type determines whether to use people or relation format\n */\nexport interface ResolvedOwner {\n type: 'people' | 'relation';\n value: Array<{ id: string }>;\n}\n",[],[327,329],"1cb975e0-b360-11f0-bc6f-0224a6c84d84","019a26dc-99bf-7269-a8f6-57913e4eabdf",["D",1761588582800],"# Shared\n\nThis directory contains code shared between frontend (browser) and backend (Deno).\n\n## Files\n\n- **types.ts** - TypeScript interfaces and types\n - Controller response format\n - Notion data structures\n - API request/response types\n\n- **utils.ts** - Utility functions\n - Must work in both browser AND Deno\n - ❌ **NEVER use Deno APIs** (won't work in browser)\n - ✅ Only use standard JavaScript/TypeScript\n\n## Guidelines\n\n### Adding Types\n- Define interfaces for data shared between layers\n- Use clear, descriptive names (PascalCase)\n- Document complex types with comments\n\n### Adding Utils\n- Test that functions work in both environments\n- Avoid platform-specific APIs (Deno, Node, browser-only)\n- Use pure functions when possible\n\n## Example\n\n```typescript\n// ✅ Good - works everywhere\nexport function formatDate(date: string): string {\n return new Date(date).toLocaleDateString();\n}\n\n// ❌ Bad - Deno-specific\nexport function readConfig() {\n return Deno.env.get('CONFIG'); // Won't work in browser!\n}\n```\n",[],[327,270],"1cb981de-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],"directory",[237,238,239],[327],"1cb97342-b360-11f0-bc6f-0224a6c84d84","index.tsx","019a26dc-99bf-7269-a8f6-4e21b80e62ce","019a26dc-99bf-7269-a8f6-51875d2ac446",["D",1761588582800],"/** @jsxImportSource https://esm.sh/react@18.2.0 */\nimport React from \"https://esm.sh/react@18.2.0\";\nimport ReactDOM from \"https://esm.sh/react-dom@18.2.0/client\";\nimport { App } from \"./components/App.tsx\";\n\nconst root = ReactDOM.createRoot(document.getElementById(\"root\")!);\nroot.render(\n <React.StrictMode>\n <App />\n </React.StrictMode>\n);\n",[],[356,349],"frontend","ba939fe6-f65b-11f0-9735-42dde27851f2","index.html","019a26dc-99bf-7269-a8f6-489546010958",321,["D",1768953428102],11,"<!DOCTYPE html>\n<html lang=\"en\">\n <head>\n <meta charset=\"UTF-8\" />\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\" />\n <title>Todo Sweeper - Recent Pages</title>\n\n \n <link\n rel=\"stylesheet\"\n href=\"https://cdn.jsdelivr.net/npm/@picocss/pico@2/css/pico.min.css\"\n />\n <link\n rel=\"stylesheet\"\n href=\"https://cdn.jsdelivr.net/npm/@picocss/pico@2/css/pico.colors.min.css\"\n />\n <style>\n #root > div {\n padding-block: var(--pico-block-spacing-vertical);\n }\n #root main section {\n margin-bottom: calc(var(--pico-spacing) * 4);\n }\n #root main section section {\n margin-bottom: calc(var(--pico-spacing) * 2);\n }\n #root main article footer details:last-child {\n margin-bottom: 0;\n }\n #root main details summary:hover {\n color: red;\n }\n </style>\n\n \n <script src=\"https://esm.town/v/std/catch\"></script>\n </head>\n <body>\n <div id=\"root\" class=\"container\"></div>\n\n \n <script type=\"module\" src=\"/frontend/index.tsx\"></script>\n </body>\n</html>\n",[],[356,358],"1cb97072-b360-11f0-bc6f-0224a6c84d84","019a26dc-99bf-7269-a8f6-463e83dc952c",["D",1761588582800],"# Frontend\n\nThis directory contains the React frontend for the Notion webhook integration application.\n\n## Structure\n\n- **index.html** - HTML shell that loads the React app\n- **index.tsx** - React app entry point\n- **components/** - React components\n - **App.tsx** - Main application component\n - **NotionBlock.tsx** - Renders Notion block content\n - **NotionProperty.tsx** - Displays Notion page properties\n\n## React Configuration\n\n**IMPORTANT**: All React files must:\n1. Include `/** @jsxImportSource https://esm.sh/react@18.2.0 */` at the top\n2. Pin React dependencies to version 18.2.0: `?deps=react@18.2.0`\n3. Use consistent React version across all imports\n\n## Styling\n\nUses Pico CSS (classless CSS framework loaded from CDN in index.html). Pico automatically styles semantic HTML elements without requiring CSS classes, keeping markup clean and maintainable.\n\n## API Communication\n\n- Fetch data from `/api/*` endpoints\n- Use shared types from `../shared/types.ts`\n- Handle loading and error states in components\n\n## Error Handling\n\nThe error catching script is included in index.html to capture client-side errors for better debugging.\n\n## Development\n\nSince this is a Val Town project, the frontend is served through Hono routes defined in `main.http.tsx`.\n",[],[356,270],"1cb96d48-b360-11f0-bc6f-0224a6c84d84","NotionProperty.tsx","019a26dc-99bf-7269-a8f6-3d6be1b615e8","019a26dc-99bf-7269-a8f6-40d88c93ec73",["D",1761588582800],"/** @jsxImportSource https://esm.sh/react@18.2.0 */\nimport { getPropertyValue } from \"../../shared/utils.ts\";\n\ninterface NotionPropertyProps {\n name: string;\n property: any;\n}\n\n/**\n * Component for rendering Notion page properties\n * Uses semantic HTML - Pico CSS provides automatic styling\n * Handles different property types\n */\nexport function NotionProperty({ name, property }: NotionPropertyProps) {\n if (!property) {\n return null;\n }\n\n const value = getPropertyValue({ [name]: property }, name);\n\n const renderValue = () => {\n switch (property.type) {\n case 'title':\n case 'rich_text':\n return <span>{value}</span>;\n\n case 'number':\n return <code>{value}</code>;\n\n case 'select':\n return <kbd>{value}</kbd>;\n\n case 'multi_select':\n return (\n <span>\n {Array.isArray(value) && value.map((item: string, i: number) => (\n <kbd key={i} style={{ marginRight: '0.5rem' }}>\n {item}\n </kbd>\n ))}\n </span>\n );\n\n case 'date':\n return (\n <time>\n {value ? new Date(value).toLocaleDateString() : '-'}\n </time>\n );\n\n case 'checkbox':\n return (\n <input\n type=\"checkbox\"\n checked={value}\n readOnly\n />\n );\n\n case 'url':\n return (\n <a\n href={value}\n target=\"_blank\"\n rel=\"noopener noreferrer\"\n >\n {value}\n </a>\n );\n\n case 'email':\n return (\n <a href={`mailto:${value}`}>\n {value}\n </a>\n );\n\n case 'phone_number':\n return (\n <a href={`tel:${value}`}>\n {value}\n </a>\n );\n\n case 'status':\n return <mark>{value}</mark>;\n\n default:\n return (\n <small>\n <em>Unsupported: {property.type}</em>\n </small>\n );\n }\n };\n\n return (\n <p>\n <strong>{name}:</strong> {renderValue()}\n </p>\n );\n}\n",[],[356,380,373],"components","1cb96a96-b360-11f0-bc6f-0224a6c84d84","NotionBlock.tsx","019a26dc-99bf-7269-a8f6-3ab802992fe4",["D",1761588582800],"/** @jsxImportSource https://esm.sh/react@18.2.0 */\n\ninterface NotionBlockProps {\n block: any;\n}\n\n/**\n * Component for rendering Notion blocks\n * Uses semantic HTML - Pico CSS provides automatic styling\n * Extend this to support different block types as needed\n */\nexport function NotionBlock({ block }: NotionBlockProps) {\n if (!block || !block.type) {\n return null;\n }\n\n const renderBlockContent = () => {\n switch (block.type) {\n case 'paragraph':\n return (\n <p>\n {block.paragraph?.rich_text?.map((text: any, i: number) => (\n <span key={i}>{text.plain_text}</span>\n ))}\n </p>\n );\n\n case 'heading_1':\n return (\n <h1>\n {block.heading_1?.rich_text?.map((text: any, i: number) => (\n <span key={i}>{text.plain_text}</span>\n ))}\n </h1>\n );\n\n case 'heading_2':\n return (\n <h2>\n {block.heading_2?.rich_text?.map((text: any, i: number) => (\n <span key={i}>{text.plain_text}</span>\n ))}\n </h2>\n );\n\n case 'heading_3':\n return (\n <h3>\n {block.heading_3?.rich_text?.map((text: any, i: number) => (\n <span key={i}>{text.plain_text}</span>\n ))}\n </h3>\n );\n\n case 'bulleted_list_item':\n return (\n <li>\n {block.bulleted_list_item?.rich_text?.map((text: any, i: number) => (\n <span key={i}>{text.plain_text}</span>\n ))}\n </li>\n );\n\n case 'numbered_list_item':\n return (\n <li>\n {block.numbered_list_item?.rich_text?.map((text: any, i: number) => (\n <span key={i}>{text.plain_text}</span>\n ))}\n </li>\n );\n\n case 'to_do':\n return (\n <label>\n <input\n type=\"checkbox\"\n checked={block.to_do?.checked}\n readOnly\n />\n <span style={block.to_do?.checked ? { textDecoration: 'line-through', opacity: 0.6 } : undefined}>\n {block.to_do?.rich_text?.map((text: any, i: number) => (\n <span key={i}>{text.plain_text}</span>\n ))}\n </span>\n </label>\n );\n\n case 'code':\n return (\n <pre>\n <code>\n {block.code?.rich_text?.map((text: any, i: number) => (\n <span key={i}>{text.plain_text}</span>\n ))}\n </code>\n </pre>\n );\n\n default:\n return (\n <small>\n <em>Unsupported block type: {block.type}</em>\n </small>\n );\n }\n };\n\n return <div>{renderBlockContent()}</div>;\n}\n",[],[356,380,382],"3203809a-02ce-11f1-bd33-42dde27851f2","App.tsx","019a26dc-99bf-7269-a8f6-34ccac805b70",422,["D",1770322005051],103,"/** @jsxImportSource https://esm.sh/react@18.2.0 */\nimport {\n useState,\n useEffect,\n} from \"https://esm.sh/react@18.2.0?deps=react@18.2.0\";\nimport type {\n PropertyStatus,\n HealthCheckResponse,\n} from \"../../shared/types.ts\";\n\n/**\n * Derived state for display purposes (UI-only)\n */\ntype DerivedState = \"found\" | \"missing\" | \"disabled\" | \"autodiscovered\";\n\n/**\n * Derive display state from property fields\n */\nfunction deriveState(prop: PropertyStatus): DerivedState {\n const effectiveName = prop.envValue ?? prop.defaultName;\n\n if (prop.foundName) {\n // Something was found\n if (prop.foundName === effectiveName) {\n return \"found\";\n }\n return \"autodiscovered\";\n }\n\n // Nothing found\n if (effectiveName) {\n return \"missing\";\n }\n return \"disabled\";\n}\n\n/**\n * Generic config row component - reusable template for all config sections\n * Summary: {label} → {value} {badge?}\n */\ntype ConfigRowProps = {\n label: string;\n purpose?: string;\n value: React.ReactNode;\n badge?: React.ReactNode;\n children: React.ReactNode;\n};\n\nfunction ConfigRow({ label, purpose, value, badge, children }: ConfigRowProps) {\n return (\n <details>\n <summary>\n {purpose ? <span data-tooltip={purpose}>{label}</span> : label} →{\" \"}\n {value} {badge}\n </summary>\n <article>{children}</article>\n </details>\n );\n}\n\n/**\n * Render a single property as stacked details\n * Summary: Name → column (or Name → ? for errors)\n */\nfunction PropertyRow({\n prop,\n envUrl,\n}: {\n prop: PropertyStatus;\n envUrl: string | null;\n}) {\n const state = deriveState(prop);\n const needsFix = state === \"autodiscovered\" || state === \"missing\";\n const isDisabled = state === \"disabled\";\n const configuredName = prop.envValue ?? prop.defaultName;\n\n // Status badge: only show \"disabled\" for disabled state\n const status = isDisabled ? (\n <code data-tooltip=\"optional - expand for details\">disabled</code>\n ) : null;\n\n // Column name: show ? with status for non-working states\n const columnName =\n state === \"found\" && prop.foundName ? (\n <code>{prop.foundName}{prop.foundType && ` (${prop.foundType})`}</code>\n ) : (\n <code\n className={\n needsFix && prop.required\n ? \"pico-background-red-500\"\n : needsFix\n ? \"pico-background-yellow-100\"\n : undefined\n }\n data-tooltip={needsFix ? \"Click for suggestions\" : undefined}\n >\n ?{needsFix && \" - not found\"}\n </code>\n );\n\n return (\n <ConfigRow label={prop.displayName} value={columnName} badge={status}>\n <dl>\n {/* Property description - always shown */}\n <dt>Property description</dt>\n <dd>{prop.purpose}</dd>\n\n {/* Success confirmation */}\n {state === \"found\" && (\n <>\n <dt>Status</dt>\n <dd className=\"pico-color-green-500\">\n ✓ Mapped to <code>{prop.foundName}</code>{\" \"}\n {prop.envValue ? \"(via env var)\" : \"(autodiscovered)\"}\n </dd>\n </>\n )}\n\n {/* Relation details - only for relation-type properties with metadata */}\n {state === \"found\" &&\n prop.foundType === \"relation\" &&\n prop.relationTargetDbId && (\n <>\n <dt>Target database</dt>\n <dd>\n {prop.relationTargetDbName || \"Unknown\"}{\" \"}\n <code>{prop.relationPageCount ?? 0} pages</code>\n </dd>\n <dt>How matching works</dt>\n <dd>\n Pages @mentioned in the todo will be linked. If the todo's\n source page is in or linked to this database, that connection\n carries over.\n </dd>\n </>\n )}\n\n {/* Mapping problem - only if there's an issue */}\n {(state === \"autodiscovered\" || state === \"missing\") && (\n <>\n <dt>Mapping problem</dt>\n <dd className=\"pico-color-red-500\">\n ✗ Your database doesn't have a property called{\" \"}\n <code>{configuredName}</code>\n </dd>\n </>\n )}\n\n {/* Similar properties - autodiscovered has foundName, missing may have candidates */}\n {state === \"autodiscovered\" && prop.foundName && (\n <>\n <dt>Properties with similar names</dt>\n <dd>\n <code>{prop.foundName}</code>\n </dd>\n </>\n )}\n\n {state === \"missing\" &&\n prop.candidates &&\n prop.candidates.length > 0 && (\n <>\n <dt>Properties with similar names</dt>\n <dd>\n {prop.candidates.map((candidate, i) => (\n <span key={candidate}>\n <code>{candidate}</code>\n {i < prop.candidates!.length - 1 && \", \"}\n </span>\n ))}\n </dd>\n </>\n )}\n\n {/* Fix options - only if there's a problem to fix */}\n {state === \"autodiscovered\" && prop.foundName && (\n <>\n <dt>To fix this mapping problem</dt>\n <dd>\n Try one of these options:\n <ul>\n <li>\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n Set env var\n </a>\n ) : (\n \"Set env var\"\n )}\n :{\" \"}\n <code>\n {prop.envVar}={prop.foundName}\n </code>\n </li>\n <li>\n Rename <code>{prop.foundName}</code> to{\" \"}\n <code>{configuredName}</code> in the database\n </li>\n </ul>\n </dd>\n </>\n )}\n\n {state === \"missing\" &&\n prop.candidates &&\n prop.candidates.length > 0 && (\n <>\n <dt>To fix this mapping problem</dt>\n <dd>\n Try one of these options:\n <ul>\n {prop.candidates.map((candidate) => (\n <li key={candidate}>\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n Set env var\n </a>\n ) : (\n \"Set env var\"\n )}\n :{\" \"}\n <code>\n {prop.envVar}={candidate}\n </code>\n </li>\n ))}\n {prop.candidates.map((candidate) => (\n <li key={`rename-${candidate}`}>\n Rename <code>{candidate}</code> to{\" \"}\n <code>{configuredName}</code> in the database\n </li>\n ))}\n <li>\n Create a <code>{configuredName}</code> property in the\n database\n </li>\n </ul>\n </dd>\n </>\n )}\n\n {state === \"missing\" &&\n (!prop.candidates || prop.candidates.length === 0) && (\n <>\n <dt>To fix this mapping problem</dt>\n <dd>\n Try one of these options:\n <ul>\n <li>\n Create a <code>{configuredName}</code> property in the\n database\n </li>\n <li>\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n Set env var\n </a>\n ) : (\n \"Set env var\"\n )}\n : <code>{prop.envVar}=your property name</code>\n </li>\n </ul>\n </dd>\n </>\n )}\n\n {/* Disabled state - optional property not configured */}\n {state === \"disabled\" &&\n prop.candidates &&\n prop.candidates.length > 0 && (\n <>\n <dt>To enable this property</dt>\n <dd>\n <ul>\n {prop.candidates.map((candidate) => (\n <li key={candidate}>\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n Set env var\n </a>\n ) : (\n \"Set env var\"\n )}\n :{\" \"}\n <code>\n {prop.envVar}={candidate}\n </code>\n </li>\n ))}\n </ul>\n </dd>\n </>\n )}\n\n {state === \"disabled\" &&\n (!prop.candidates || prop.candidates.length === 0) && (\n <>\n <dt>To enable this property</dt>\n <dd>\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n Set env var\n </a>\n ) : (\n \"Set env var\"\n )}\n : <code>{prop.envVar}=your property name</code>\n </dd>\n </>\n )}\n </dl>\n\n <footer>\n <details>\n <summary>JSON</summary>\n <pre>{JSON.stringify(prop, null, 2)}</pre>\n </details>\n </footer>\n </ConfigRow>\n );\n}\n\nexport function App() {\n const [loading, setLoading] = useState(true);\n const [error, setError] = useState<string | null>(null);\n const [healthData, setHealthCheckResponse] =\n useState<HealthCheckResponse | null>(null);\n\n const fetchHealthCheckResponse = async () => {\n setLoading(true);\n setError(null);\n\n try {\n const response = await fetch(\"/api/health\");\n const data: HealthCheckResponse = await response.json();\n\n if (!response.ok) {\n throw new Error(\"Failed to fetch health data\");\n }\n\n setHealthCheckResponse(data);\n } catch (err: any) {\n setError(err.message || \"An error occurred\");\n } finally {\n setLoading(false);\n }\n };\n\n // Fetch health data on mount\n useEffect(() => {\n fetchHealthCheckResponse();\n }, []);\n\n // Update page title with project info\n useEffect(() => {\n if (healthData?.project) {\n document.title = `${healthData.project.username}/${healthData.project.name}`;\n } else {\n document.title = \"todoSweeper - Health Check\";\n }\n }, [healthData]);\n\n // Construct Val Town environment variables URL\n const envUrl = healthData?.project\n ? `https://www.val.town/x/${healthData.project.username}/${healthData.project.name}/environment-variables`\n : null;\n\n return (\n <div className=\"container\">\n <header>\n <hgroup>\n {healthData?.project ? (\n <>\n <code>\n {healthData.project.url ? (\n <a href={healthData.project.url} target=\"_blank\">\n {healthData.project.username}/{healthData.project.name}\n </a>\n ) : (\n `${healthData.project.username}/${healthData.project.name}`\n )}\n </code>\n <small style={{ marginLeft: \"0.5rem\" }}>\n v{healthData.project.version}\n {healthData.project.branch !== \"main\" &&\n ` (${healthData.project.branch})`}\n </small>\n </>\n ) : (\n <code>todoSweeper</code>\n )}\n <p>Sweeps todos from Notion pages into a database for automations</p>\n </hgroup>\n </header>\n <hr />\n <main>\n <hgroup>\n <h1>Configuration</h1>\n <p>Connections, database property mappings, and settings</p>\n </hgroup>\n {loading && <p aria-busy=\"true\">Checking system health...</p>}\n\n {!loading && !error && healthData && (\n <>\n {/* Health Summary */}\n {(() => {\n // Compute issue counts\n const connectionIssues = [\n healthData.connections.notion,\n healthData.connections.database,\n ].filter((c) => c.status !== \"connected\").length;\n\n const requiredIssues = healthData.properties\n .filter((p) => p.required)\n .filter((p) => {\n const state = deriveState(p);\n return state === \"missing\" || state === \"autodiscovered\";\n }).length;\n\n const totalIssues = connectionIssues + requiredIssues;\n\n return (\n <section>\n {totalIssues > 0 && (\n <article className=\"pico-background-red-100\">\n <span className=\"pico-color-red-500\">✗</span>{\" \"}\n {totalIssues} issue{totalIssues > 1 ? \"s\" : \"\"}\n </article>\n )}\n </section>\n );\n })()}\n\n {/* Services */}\n <section>\n <hgroup>\n <h2>Connections</h2>\n <p></p>\n </hgroup>\n <dl>\n {/* Notion API */}\n <dt>\n <strong>Notion</strong>\n </dt>\n <dd>\n <details>\n <summary>\n {healthData.connections.notion.status === \"connected\" ? (\n <code className=\"pico-background-green-100\">\n Connected to{\" \"}\n {healthData.connections.notion.integrationName}\n </code>\n ) : (\n <code\n className=\"pico-background-red-500\"\n data-tooltip=\"Click for suggestions\"\n >\n Not connected\n </code>\n )}\n </summary>\n <article>\n <dl>\n {healthData.connections.notion.status ===\n \"connected\" && (\n <>\n <dt>Integration name</dt>\n <dd>\n <code>\n {healthData.connections.notion.integrationName}\n </code>\n </dd>\n </>\n )}\n {healthData.connections.notion.status === \"error\" && (\n <>\n <dt>Error</dt>\n <dd>{healthData.connections.notion.error}</dd>\n </>\n )}\n <dt>Environment variable</dt>\n <dd>\n <code>{healthData.connections.notion.envVar}</code>\n </dd>\n </dl>\n {healthData.connections.notion.status !== \"connected\" && (\n <p>\n To connect, add to your{\" \"}\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n Val Town environment variables\n </a>\n ) : (\n \"Val Town environment variables\"\n )}\n :\n <br />\n <code>\n {healthData.connections.notion.envVar}\n =your-integration-token\n </code>\n </p>\n )}\n <footer>\n <details>\n <summary>JSON</summary>\n <pre>\n {JSON.stringify(\n healthData.connections.notion,\n null,\n 2\n )}\n </pre>\n </details>\n </footer>\n </article>\n </details>\n </dd>\n\n {/* OpenAI */}\n <dt>\n <strong>OpenAI</strong>\n </dt>\n <dd>\n <details>\n <summary>\n {healthData.connections.openai.status === \"connected\" ? (\n <code className=\"pico-background-green-100\">\n Connected\n </code>\n ) : healthData.connections.openai.status === \"error\" ? (\n <code\n className=\"pico-background-red-500\"\n data-tooltip=\"Click for suggestions\"\n >\n Not connected\n </code>\n ) : (\n <code\n className=\"pico-background-yellow-100\"\n data-tooltip=\"Click for suggestions\"\n >\n Not configured\n </code>\n )}\n </summary>\n <article>\n <dl>\n {healthData.connections.openai.status === \"error\" && (\n <>\n <dt>Error</dt>\n <dd>{healthData.connections.openai.error}</dd>\n </>\n )}\n <dt>Environment variable</dt>\n <dd>\n <code>{healthData.connections.openai.envVar}</code>\n </dd>\n </dl>\n {healthData.connections.openai.status !== \"connected\" && (\n <p>\n To connect to OpenAI, add to your{\" \"}\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n Val Town environment variables\n </a>\n ) : (\n \"Val Town environment variables\"\n )}\n :\n <br />\n <code>\n {healthData.connections.openai.envVar}=your-api-key\n </code>\n </p>\n )}\n <footer>\n <details>\n <summary>JSON</summary>\n <pre>\n {JSON.stringify(\n healthData.connections.openai,\n null,\n 2\n )}\n </pre>\n </details>\n </footer>\n </article>\n </details>\n </dd>\n\n {/* Todos Database - only show if Notion is connected */}\n {healthData.connections.notion.status === \"connected\" && (\n <>\n <dt>\n <strong>Todos database</strong>\n </dt>\n <dd>\n <details>\n <summary>\n {healthData.connections.database.status ===\n \"connected\" ? (\n <code className=\"pico-background-green-100\">\n Connected to{\" \"}\n {healthData.connections.database.name}\n </code>\n ) : (\n <code\n className=\"pico-background-red-500\"\n data-tooltip=\"Click for suggestions\"\n >\n Not connected\n </code>\n )}\n </summary>\n <article>\n <dl>\n {healthData.connections.database.status ===\n \"connected\" && (\n <>\n <dt>Database name</dt>\n <dd>\n {healthData.connections.database.url ? (\n <a\n href={healthData.connections.database.url}\n target=\"_blank\"\n >\n {healthData.connections.database.name}\n </a>\n ) : (\n healthData.connections.database.name\n )}\n </dd>\n <dt>Database ID</dt>\n <dd>\n <code>\n {healthData.connections.database.id}\n </code>\n </dd>\n </>\n )}\n {healthData.connections.database.status ===\n \"error\" && (\n <>\n <dt>Error</dt>\n <dd>{healthData.connections.database.error}</dd>\n </>\n )}\n <dt>Environment variable</dt>\n <dd>\n <code>\n {healthData.connections.database.envVar}\n </code>\n </dd>\n </dl>\n {healthData.connections.database.status !==\n \"connected\" && (\n <p>\n To connect, add to your{\" \"}\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n Val Town environment variables\n </a>\n ) : (\n \"Val Town environment variables\"\n )}\n :\n <br />\n <code>\n {healthData.connections.database.envVar}\n =your-database-id\n </code>\n </p>\n )}\n <footer>\n <details>\n <summary>JSON</summary>\n <pre>\n {JSON.stringify(\n healthData.connections.database,\n null,\n 2\n )}\n </pre>\n </details>\n </footer>\n </article>\n </details>\n </dd>\n </>\n )}\n </dl>\n </section>\n\n {/* Property Configuration - only show if Notion is connected */}\n {healthData.connections.notion.status === \"connected\" && (\n <section>\n <hgroup>\n <h2>Todos database property mappings</h2>\n <p>\n For required, optional, and relation database properties\n </p>\n </hgroup>\n {/* <ul>\n <li>\n Default property display name →{\" \"}\n <code>Actual database property name</code>\n </li>\n </ul> */}\n {/* Required - sync will fail without these */}\n <section>\n <hgroup>\n <h5>Required property mappings</h5>\n <p>The sweep will fail if these properties have errors.</p>\n </hgroup>\n <ul>\n {healthData.properties\n .filter((p) => p.required)\n .map((prop) => (\n <li key={prop.key}>\n <PropertyRow prop={prop} envUrl={envUrl} />\n </li>\n ))}\n </ul>\n </section>\n\n {/* Optional - nice to have */}\n <section>\n <hgroup>\n <h5>Optional property mappings</h5>\n <p>\n These properties are not required for a successful sweep.\n </p>\n </hgroup>\n <ul>\n {healthData.properties\n .filter((p) => !p.required)\n .sort((a, b) =>\n a.displayName.localeCompare(b.displayName)\n )\n .map((prop) => (\n <li key={prop.key}>\n <PropertyRow prop={prop} envUrl={envUrl} />\n </li>\n ))}\n </ul>\n </section>\n </section>\n )}\n\n {/* Display Formula Helper */}\n {healthData.displayFormula && (\n <section>\n <hgroup>\n <h2>Display Formula</h2>\n <p>\n Copy this formula into a new formula property in your todos\n database to show: Owner + Task + Due date\n </p>\n </hgroup>\n <pre\n style={{\n whiteSpace: \"pre-wrap\",\n wordBreak: \"break-all\",\n fontSize: \"0.85em\",\n }}\n >\n <code>{healthData.displayFormula}</code>\n </pre>\n <p>\n <small>\n Example output: \"Alex to review the quarterly report by\n January 15, 2026\"\n </small>\n </p>\n </section>\n )}\n\n {/* Environment */}\n <section>\n <hgroup>\n <h2>Settings</h2>\n <p>\n Set by{\" \"}\n {envUrl ? (\n <a href={envUrl} target=\"_blank\">\n environment variables\n </a>\n ) : (\n \"environment variables\"\n )}{\" \"}\n in Val Town\n </p>\n </hgroup>\n <table>\n <thead>\n <tr>\n <th>Environment variable</th>\n <th>Value</th>\n </tr>\n </thead>\n <tbody>\n {healthData.settings.map((setting) => (\n <tr key={setting.key}>\n <td>\n <code>{setting.envVar}</code>\n <br />\n {setting.description}\n {setting.defaultValue !== null && (\n <>\n {\" \"}\n Default: <code>{setting.defaultValue}</code>\n </>\n )}\n </td>\n <td>\n {setting.value !== null ? (\n <code>{setting.value}</code>\n ) : (\n \"not set\"\n )}\n </td>\n </tr>\n ))}\n </tbody>\n </table>\n </section>\n\n {/* Raw JSON */}\n <section>\n <details>\n <summary>\n Raw health data (<code>JSON</code>)\n </summary>\n <article>\n <p>\n Also available at{\" \"}\n <code>\n <a href=\"/api/health\" target=\"_blank\">\n /api/health\n </a>\n </code>\n </p>\n <pre>{JSON.stringify(healthData, null, 2)}</pre>\n </article>\n </details>\n </section>\n </>\n )}\n\n {!loading && error && (\n <details open>\n <summary>\n <strong>✗ Connection Error</strong>\n </summary>\n <div>\n <p>{error}</p>\n <h4>Troubleshooting</h4>\n <ul>\n <li>\n Check that <code>NOTION_API_KEY</code> is set correctly\n </li>\n <li>Verify your integration has access to your workspace</li>\n <li>Ensure the API key hasn't expired</li>\n <li>Make sure pages are shared with your integration</li>\n </ul>\n </div>\n </details>\n )}\n </main>\n {healthData?.timestamp && (\n <footer>\n Last checked: {new Date(healthData.timestamp).toLocaleString()}\n </footer>\n )}\n </div>\n );\n}\n",[],[356,380,389],"1cb96f5a-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[230,231,232],[356,380],"1cb9746e-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[229,233,234,235],[356],"1cb96514-b360-11f0-bc6f-0224a6c84d84","019a26dc-99bf-7269-a8f6-2fbd5953f9c6","019a26dc-99bf-7269-a8f6-32bd7c2a6dc6",["D",1761588582800],"# Backend\n\nThis directory contains the backend logic for the Notion webhook integration application.\n\n## Structure\n\n- **controllers/** - Business logic and orchestration\n - Contains functions that validate data, orchestrate service calls, and return standardized responses\n - All controllers return `{success, data, error, details?}` format\n - Controllers are HTTP-agnostic (don't create HTTP responses)\n\n- **routes/** - HTTP request/response handling\n - **api/** - RESTful API endpoints for data operations\n - **tasks/** - Notion webhook handlers\n - **views/** - User-facing HTML/React views\n - **authCheck.ts** - Authentication middleware\n\n- **services/** - External API integrations\n - Handle calls to external APIs (Notion, Firecrawler, blob storage, etc.)\n - Return structured results with `{success, data, error}` format\n - Manage API authentication and error handling\n\n## Architecture Flow\n\n```\nRequest → Route → Controller → Service → External API\n```\n\n**Never skip layers!**\n- Routes call controllers (not services directly)\n- Controllers call services (not external APIs directly)\n- Services handle external API calls\n\n## Adding New Features\n\n1. Create route in appropriate directory (`api/`, `tasks/`, or `views/`)\n2. Create controller function with business logic\n3. Create/use service functions for external API calls\n4. Mount route in `main.http.tsx`\n\nSee `CLAUDE.md` for detailed guidelines.\n",[],[412,270],"backend","cbdb8792-ebfb-11f0-8ce9-42dde27851f2","valUtils.ts","cbdb8846-ebfb-11f0-8ce9-42dde27851f2","019a2d3c-4319-71ac-88e5-a0aa162b3be7",256,["D",1767812713786],"/**\n * Val Town Project Utilities\n *\n * Provides cached access to project metadata from Val Town's std utils.\n * Used for dynamic project name display and blob key prefixes.\n */\nimport { parseProject } from \"https://esm.town/v/std/utils@85-main/index.ts\";\n\n/**\n * Project information from Val Town\n */\nexport interface ProjectInfo {\n name: string;\n username: string;\n version: number;\n branch: string;\n url: string | null;\n}\n\nlet cachedProjectInfo: ProjectInfo | null = null;\n\n/**\n * Get project information from Val Town\n *\n * Results are cached since project info doesn't change at runtime.\n *\n * @returns Project info object\n */\nexport function getProjectInfo(): ProjectInfo {\n if (!cachedProjectInfo) {\n try {\n const info = parseProject(import.meta.url);\n cachedProjectInfo = {\n name: info.name || 'unknown',\n username: info.username || 'unknown',\n version: info.version || 0,\n branch: info.branch || 'main',\n url: info.links?.self?.project || null\n };\n } catch {\n cachedProjectInfo = {\n name: 'unknown',\n username: 'unknown',\n version: 0,\n branch: 'main',\n url: null\n };\n }\n }\n return cachedProjectInfo;\n}\n\n/**\n * Get just the project name\n *\n * Convenience function for cases where only the name is needed.\n *\n * @returns Project name string\n */\nexport function getProjectName(): string {\n return getProjectInfo().name;\n}\n",[],[412,422,414],"utils","b947264a-b659-11f0-934f-0224a6c84d84","textUtils.ts","b94726c2-b659-11f0-934f-0224a6c84d84",20,["D",1761915692505],"/**\n * Text utility functions for word counting and text validation\n * Pure functions with no external dependencies\n */\n\n/**\n * Get minimum block words setting from environment variable\n *\n * Reads MIN_BLOCK_WORDS env var\n * Falls back to 5 if not set or invalid\n *\n * Default of 5 accounts for:\n * - ~1 word for person mention (e.g., \"@John\")\n * - ~2-3 words for date mention (e.g., \"October 30, 2025\")\n * - ~2-3 words for action (e.g., \"buy groceries\")\n *\n * @returns Minimum word count (integer >= 1)\n */\nexport function getMinBlockWords(): number {\n const envValue = Deno.env.get('MIN_BLOCK_WORDS');\n\n if (!envValue) {\n return 5; // Default fallback\n }\n\n const parsed = parseInt(envValue.trim(), 10);\n\n // Validate: must be a positive integer\n if (isNaN(parsed) || parsed < 1) {\n console.warn(`⚠ Invalid MIN_BLOCK_WORDS value: \"${envValue}\". Using 5 as fallback.`);\n return 5;\n }\n\n return parsed;\n}\n\n/**\n * Count words in a text string\n *\n * Splits by whitespace and counts non-empty tokens\n * - Hyphenated words count as 1 word (e.g., \"buy-now\" = 1)\n * - Emojis count as words (e.g., \"🍋 🎉\" = 2)\n * - Mentions count as words (e.g., \"@John tomorrow\" = 2)\n * - Dates count as words (e.g., \"October 30, 2025\" = 3)\n *\n * @param text - Text string to count words in\n * @returns Number of words\n */\nexport function countWords(text: string): number {\n if (!text || typeof text !== 'string') {\n return 0;\n }\n\n // Split by whitespace, filter empty strings\n const words = text.trim().split(/\\s+/).filter(w => w.length > 0);\n\n return words.length;\n}\n",[],[412,422,424],"85daeb50-020f-11f1-8e35-42dde27851f2","notionUtils.ts","1582a536-b459-11f0-a09c-0224a6c84d84",["D",1770240111800],"/**\n * Notion utility functions for block transformations and processing\n */\n\nimport type { RelationMatchResult, ResolvedOwner } from '../../shared/types.ts';\nimport type { StatusPropertyInfo } from '../services/notion/databases.ts';\n\n/**\n * Format an ISO date string to human-readable format\n *\n * @param dateStr - ISO date string (e.g., \"2025-10-27\" or \"2025-10-27T09:00:00.000-04:00\")\n * @returns Formatted date string (e.g., \"October 27, 2025\" or \"October 27, 2025 at 9:00 AM EDT\")\n */\nexport function formatDateString(dateStr: string): string {\n try {\n // Check if it's date-only (no time component in original string)\n const isDateOnly = !dateStr.includes('T');\n\n if (isDateOnly) {\n // Date-only format: \"October 27, 2025\"\n const date = new Date(dateStr);\n return date.toLocaleDateString('en-US', {\n year: 'numeric',\n month: 'long',\n day: 'numeric',\n timeZone: 'UTC' // Use UTC for date-only to avoid timezone shifts\n });\n } else {\n // Datetime format: \"October 27, 2025 at 9:00 AM EDT\"\n // Parse the ISO string to extract components without timezone conversion\n const isoMatch = dateStr.match(/^(\\d{4})-(\\d{2})-(\\d{2})T(\\d{2}):(\\d{2}):(\\d{2})(?:\\.\\d+)?([-+]\\d{2}:\\d{2})?$/);\n\n if (!isoMatch) {\n // Fallback to original string if format doesn't match\n return dateStr;\n }\n\n const [_, year, month, day, hour24, minute, second, tzOffset] = isoMatch;\n\n // Create date object for formatting the date part only\n const date = new Date(`${year}-${month}-${day}`);\n const datePart = date.toLocaleDateString('en-US', {\n year: 'numeric',\n month: 'long',\n day: 'numeric',\n timeZone: 'UTC'\n });\n\n // Format time part - preserve the original hour/minute\n const hourNum = parseInt(hour24, 10);\n const minuteNum = parseInt(minute, 10);\n const isPM = hourNum >= 12;\n const hour12 = hourNum === 0 ? 12 : (hourNum > 12 ? hourNum - 12 : hourNum);\n const timePart = `${hour12}:${minute} ${isPM ? 'PM' : 'AM'}`;\n\n // Extract timezone abbreviation\n let tzAbbrev = '';\n if (tzOffset) {\n // Common timezone abbreviations based on offset\n const tzMap: Record<string, string> = {\n '-04:00': 'EDT',\n '-05:00': 'EST/CDT',\n '-06:00': 'CST/MDT',\n '-07:00': 'MST/PDT',\n '-08:00': 'PST',\n '+00:00': 'UTC',\n // Add more as needed\n };\n tzAbbrev = ` ${tzMap[tzOffset] || tzOffset}`;\n }\n\n return `${datePart} at ${timePart}${tzAbbrev}`;\n }\n } catch (err) {\n // If parsing fails, return original string\n return dateStr;\n }\n}\n\n/**\n * Remove hyphens from a string (for Notion URLs)\n */\nfunction removeHyphens(str: string): string {\n return str.replace(/-/g, '');\n}\n\nexport function reduceNotionBlock(block: any) {\n const blockId = block?.id;\n const lastEdited = block?.last_edited_time;\n\n // Generate Notion block URL\n const blockUrl = blockId ? `https://www.notion.so/${removeHyphens(blockId)}` : '';\n\n const richText = block?.[block.type]?.rich_text ?? [];\n\n // Extract ALL date mentions - AI will determine which is the due date\n const date_mentions: string[] = [];\n for (const part of richText) {\n if (part.type === \"mention\" && part.mention?.type === \"date\") {\n const dateStr = part.mention.date.start ?? part.mention.date.end;\n if (dateStr) {\n date_mentions.push(dateStr);\n }\n }\n }\n\n // Use same array for display formatting (previously was separate allDateStrings)\n const allDateStrings = date_mentions;\n\n // Build full sentence and replace ISO dates with formatted versions\n let fullSentence = richText\n .map((part: any) => part.plain_text)\n .join(\"\")\n .replace(/\\s+/g, \" \") // collapse multiple spaces\n .trim();\n\n // Replace ALL date mentions in the full sentence with formatted version\n for (const dateStr of allDateStrings) {\n const formattedDate = formatDateString(dateStr);\n // Replace the ISO date string with formatted version\n fullSentence = fullSentence.replace(dateStr, formattedDate);\n }\n\n // People mentions\n const people_mentions = richText\n .filter((part: any) => part.type === \"mention\" && part.mention?.type === \"user\")\n .map((part: any) => {\n const user = part.mention.user;\n return {\n id: user.id,\n name: user.name,\n email: user.person?.email ?? null,\n };\n });\n\n // Link mentions (Notion pages, external links - but NOT person mentions)\n // Enhanced with pageId for dynamic relation mapping\n const link_mentions = richText\n .filter((part: any) => {\n // Exclude person mentions - they should only go to people_mentions\n // (Person mentions have href pointing to user profile, which would incorrectly\n // send them to the Links property when not matched to a relation)\n if (part.type === \"mention\" && part.mention?.type === \"user\") {\n return false;\n }\n // Check for href property (page mentions and some links)\n // OR check for text.link.url (regular hyperlinks)\n return part.href || part.text?.link?.url;\n })\n .map((part: any) => {\n const url = part.href || part.text?.link?.url;\n const pageId = extractPageIdFromUrl(url); // Extract Notion page ID if present\n return {\n text: part.plain_text,\n url,\n pageId, // Normalized page ID (no hyphens) or null\n parentDatabaseId: null // Will be populated later in todoController\n };\n });\n\n // Extract checkbox state for to_do blocks\n const checked = block.type === 'to_do' ? (block.to_do?.checked === true) : undefined;\n\n return {\n todo_string: fullSentence,\n block_id: blockId,\n block_url: blockUrl,\n created_time: block?.created_time,\n last_edited_time: lastEdited,\n people_mentions,\n date_mentions,\n link_mentions,\n checked,\n };\n}\n\n/**\n * Map ReducedNotionBlock to Notion database properties\n *\n * Property mappings:\n * - todo_string → Name (title)\n * - block_id → Block ID (rich_text)\n * - block_url → Block URL (url)\n * - last_edited_time → Todo last edited time (date)\n * - author → Author (people) - block creator from created_by\n * - ownerValue → Owner (people or relation) - derived from AI Summary\n * - people_mentions → Other people (people) - ALL @mentions in order\n * - date_mentions[0] → Due date (date)\n * - link_mentions → Links (rich_text - only unmapped mentions)\n * - relationMatches → Dynamic relation properties\n * - checked + statusInfo → Status property (when checkbox is checked)\n *\n * @param reducedBlock - The reduced block data\n * @param validatedProperties - Validated property names (key → actual column name or null)\n * @param relationMatches - Dynamic relation matches from matchBlockToRelations\n * @param summary - AI-interpreted polished todo text (used for Name if available)\n * @param ownerValue - Resolved owner (person or relation format)\n * @param statusInfo - Status property info for checkbox sync (optional)\n */\nexport function mapReducedBlockToProperties(\n reducedBlock: any,\n validatedProperties: Record<string, string | null>,\n relationMatches?: RelationMatchResult[],\n summary?: string,\n ownerValue?: ResolvedOwner | null,\n statusInfo?: StatusPropertyInfo | null\n) {\n const properties: any = {};\n\n // Helper to get validated property name (returns null if property doesn't exist)\n const getPropName = (key: string): string | null => validatedProperties[key] || null;\n\n // Name (title) - AI summary if available, otherwise raw text\n const nameProp = getPropName('name');\n if (nameProp) {\n const nameText = summary || reducedBlock.todo_string;\n properties[nameProp] = {\n title: [\n {\n type: \"text\",\n text: { content: nameText }\n }\n ]\n };\n }\n\n // Source - always raw todo text from source block\n const sourceProp = getPropName('sourceText');\n if (sourceProp) {\n properties[sourceProp] = {\n rich_text: [\n {\n type: \"text\",\n text: { content: reducedBlock.todo_string }\n }\n ]\n };\n }\n\n // Block ID (rich_text)\n const blockIdProp = getPropName('sourceBlockId');\n if (blockIdProp) {\n properties[blockIdProp] = {\n rich_text: [\n {\n type: \"text\",\n text: { content: reducedBlock.block_id }\n }\n ]\n };\n }\n\n // Block URL (url)\n const blockUrlProp = getPropName('sourceBlockUrl');\n if (blockUrlProp && reducedBlock.block_url) {\n properties[blockUrlProp] = {\n url: reducedBlock.block_url\n };\n }\n\n // Page URL (url) - source page where the todo block was found\n const pageUrlProp = getPropName('sourcePageUrl');\n if (pageUrlProp && reducedBlock.page_url) {\n properties[pageUrlProp] = {\n url: reducedBlock.page_url\n };\n }\n\n // Todo last edited time (date)\n const lastEditedProp = getPropName('sourceLastEditedTime');\n if (lastEditedProp) {\n properties[lastEditedProp] = {\n date: {\n start: reducedBlock.last_edited_time\n }\n };\n }\n\n // Author (people) - block creator from created_by\n const authorProp = getPropName('sourceAuthor');\n if (authorProp && reducedBlock.author && reducedBlock.author.id) {\n properties[authorProp] = {\n people: [{ id: reducedBlock.author.id }]\n };\n }\n\n // Owner - derived from AI Summary, can be people or relation type\n const ownerProp = getPropName('owner');\n if (ownerProp && ownerValue) {\n if (ownerValue.type === 'people') {\n properties[ownerProp] = {\n people: ownerValue.value\n };\n } else if (ownerValue.type === 'relation') {\n properties[ownerProp] = {\n relation: ownerValue.value\n };\n }\n }\n\n // Other people (people) - ALL @mentions in order they appear\n const otherPeopleProp = getPropName('otherPeople');\n if (otherPeopleProp && reducedBlock.people_mentions && reducedBlock.people_mentions.length > 0) {\n properties[otherPeopleProp] = {\n people: reducedBlock.people_mentions.map((p: any) => ({ id: p.id }))\n };\n }\n\n // Due date (date) - first date\n const dueDateProp = getPropName('dueDate');\n if (dueDateProp && reducedBlock.date_mentions && reducedBlock.date_mentions.length > 0) {\n properties[dueDateProp] = {\n date: {\n start: reducedBlock.date_mentions[0]\n }\n };\n }\n\n // Collect all pageIds that were mapped to relations (for deduplication)\n const mappedPageIds = new Set<string>();\n\n // Apply dynamic relation matches\n if (relationMatches && relationMatches.length > 0) {\n for (const match of relationMatches) {\n // Skip Owner property - it's handled separately above with dedicated resolution logic\n // (prevents relation matching from overwriting heading/AI-determined owner)\n if (ownerProp && match.propertyName === ownerProp) {\n continue;\n }\n\n properties[match.propertyName] = {\n relation: match.pageIds.map(id => ({ id }))\n };\n // Track which pageIds were mapped to relations\n match.pageIds.forEach(id => mappedPageIds.add(id));\n }\n }\n\n // Links (rich_text) - only include mentions NOT mapped to relations\n const linksProp = getPropName('links');\n if (linksProp && reducedBlock.link_mentions && reducedBlock.link_mentions.length > 0) {\n // Filter out mentions that were mapped to relations\n const unmappedMentions = reducedBlock.link_mentions.filter(\n (lm: any) => !lm.pageId || !mappedPageIds.has(lm.pageId)\n );\n\n if (unmappedMentions.length > 0) {\n properties[linksProp] = {\n rich_text: unmappedMentions.flatMap((link: any, index: number) => {\n const parts = [];\n\n // Add bullet before each link\n parts.push({ type: \"text\", text: { content: \"• \" } });\n\n // Add the link\n parts.push({\n type: \"text\",\n text: {\n content: link.text,\n link: { url: link.url }\n }\n });\n\n // Add newline after each link except the last\n if (index < unmappedMentions.length - 1) {\n parts.push({ type: \"text\", text: { content: \"\\n\" } });\n }\n\n return parts;\n })\n };\n }\n }\n\n // Status - sync checkbox state to status property\n // When checked: set to \"Done\" (or equivalent)\n // When unchecked: set to first non-done option (e.g., \"Not started\")\n // This ensures unchecking a todo properly reverts the status\n const statusProp = getPropName('status');\n if (statusProp && statusInfo && reducedBlock.checked !== undefined) {\n if (statusInfo.type === 'checkbox') {\n properties[statusProp] = {\n checkbox: reducedBlock.checked\n };\n } else if (statusInfo.type === 'status' || statusInfo.type === 'select') {\n if (reducedBlock.checked && statusInfo.doneName) {\n // Checked → set to \"Done\"\n properties[statusProp] = {\n [statusInfo.type]: { name: statusInfo.doneName }\n };\n } else if (!reducedBlock.checked && statusInfo.notDoneName) {\n // Unchecked → set to first non-done option (e.g., \"Not started\")\n properties[statusProp] = {\n [statusInfo.type]: { name: statusInfo.notDoneName }\n };\n }\n }\n }\n\n return properties;\n}\n\n/**\n * Extract Notion page ID from a URL\n *\n * Handles various Notion URL formats:\n * - https://www.notion.so/2393adc4305780f9b923d904dfacb88c (32 chars, no hyphens)\n * - https://www.notion.so/2393adc4-3057-80f9-b923-d904dfacb88c (36 chars, with hyphens)\n * - https://www.notion.so/workspace/Page-Name-2393adc4305780f9b923d904dfacb88c\n *\n * @param url - Notion page URL\n * @returns Normalized page ID (32 chars, no hyphens) or null if invalid\n */\nexport function extractPageIdFromUrl(url: string): string | null {\n if (!url || typeof url !== 'string') {\n return null;\n }\n\n try {\n // Parse URL to get pathname\n const urlObj = new URL(url);\n const pathname = urlObj.pathname;\n\n // Extract the last segment (which should contain the page ID)\n const segments = pathname.split('/').filter(s => s.length > 0);\n const lastSegment = segments[segments.length - 1];\n\n if (!lastSegment) {\n return null;\n }\n\n // Page ID can be:\n // 1. At the end: \"Page-Name-2393adc4305780f9b923d904dfacb88c\"\n // 2. The entire segment: \"2393adc4305780f9b923d904dfacb88c\"\n // 3. With hyphens: \"2393adc4-3057-80f9-b923-d904dfacb88c\"\n\n // Try to extract 32 or 36 character UUID-like pattern at the end\n const uuidPattern32 = /([a-f0-9]{32})$/i; // 32 chars without hyphens\n const uuidPattern36 = /([a-f0-9]{8}-[a-f0-9]{4}-[a-f0-9]{4}-[a-f0-9]{4}-[a-f0-9]{12})$/i; // 36 chars with hyphens\n\n // Try 36-char pattern first (with hyphens)\n let match = lastSegment.match(uuidPattern36);\n if (match) {\n // Remove hyphens to normalize\n return match[1].replace(/-/g, '');\n }\n\n // Try 32-char pattern (no hyphens)\n match = lastSegment.match(uuidPattern32);\n if (match) {\n return match[1];\n }\n\n return null;\n\n } catch (err) {\n // Invalid URL format\n return null;\n }\n}\n",[],[412,422,432],"320354da-02ce-11f1-bd33-42dde27851f2","notionConfig.ts","67402628-eb09-11f0-9d1c-42dde27851f2",["D",1770322005051],18,"/**\n * PROPERTY CONFIGURATION - Single Source of Truth\n *\n * To add a new property:\n * 1. Add default name to TODOS_DEFAULTS below\n * 2. Add metadata to PROPERTY_METADATA below\n * 3. Done - types flow through API to frontend automatically\n *\n * To override a property name: Set env var TODOS_PROP_{KEY}=column name\n * To disable a property: Set env var TODOS_PROP_{KEY}= (empty string)\n *\n * Examples: TODOS_PROP_DUE_DATE=Deadline, TODOS_PROP_STATUS=\n */\n\nimport type { PropertyMetadata } from '../../shared/types.ts';\n\n// Get all env vars once at module load (avoids Val Town warning spam)\nconst ENV = Deno.env.toObject();\n\nconst TODOS_DEFAULTS = {\n name: \"Name\",\n sourceBlockId: \"Source block ID\",\n sourceBlockUrl: \"Source block URL\",\n sourcePageUrl: \"Source page URL\",\n sourceLastEditedTime: \"Source last edited time\",\n sourceAuthor: \"Source author\",\n owner: \"Owner\",\n otherPeople: \"Other people\",\n dueDate: \"Due date\",\n links: \"Links\",\n sourceText: \"Source text\",\n status: null as string | null, // disabled by default - set TODOS_PROP_STATUS to sync checkbox state\n projects: null as string | null, // disabled by default - set TODOS_PROP_PROJECTS to enable project linking\n};\n\n/**\n * Property metadata - declarative configuration for health check\n * Adjust category/required/autodiscoverable flags here to change behavior\n *\n * Categories:\n * - core: Essential todo properties (name, dueDate, status, owner)\n * - system: Infrastructure properties (blockId, urls, timestamps)\n * - enrichment: Additional context (author, links, source)\n */\nexport const PROPERTY_METADATA: Record<keyof typeof TODOS_DEFAULTS, PropertyMetadata> = {\n // === CORE: What makes a todo ===\n name: {\n displayName: \"Name\",\n purpose: \"What the todo says. May be cleaned up by AI if OpenAI is configured.\",\n category: \"core\",\n required: true,\n autodiscoverable: true,\n expectedTypes: [\"title\"]\n },\n dueDate: {\n displayName: \"Due date\",\n purpose: \"Due date determined by AI from the todo content, or defaults to DEFAULT_DUE_DATE setting.\",\n category: \"core\",\n required: false,\n autodiscoverable: true,\n expectedTypes: [\"date\"],\n namePatterns: [\"due\", \"due date\", \"deadline\", \"when\"]\n },\n status: {\n displayName: \"Status\",\n purpose: \"Tracks completion status. When the source checkbox is checked, this property is updated to reflect it.\",\n category: \"core\",\n required: false,\n autodiscoverable: true,\n expectedTypes: [\"status\", \"select\", \"checkbox\"]\n },\n owner: {\n displayName: \"Owner\",\n purpose: \"The person responsible for this todo, determined from page context or AI analysis.\",\n category: \"core\",\n required: false,\n autodiscoverable: true,\n expectedTypes: [\"people\", \"relation\"],\n namePatterns: [\"owner\", \"assigned\", \"assigned to\", \"assignee\", \"dri\"]\n },\n\n // === SYSTEM: Infrastructure ===\n sourceBlockId: {\n displayName: \"Source block ID\",\n purpose: \"Unique identifier of the source block. Used to prevent duplicates and link back to the original todo.\",\n category: \"system\",\n required: true,\n autodiscoverable: true,\n expectedTypes: [\"rich_text\"],\n namePatterns: [\"block id\", \"block_id\", \"blockid\", \"source block\"]\n },\n sourceBlockUrl: {\n displayName: \"Source block URL\",\n purpose: \"Direct link to the source block in Notion for quick navigation.\",\n category: \"system\",\n required: false,\n autodiscoverable: false,\n expectedTypes: [\"url\"]\n },\n sourcePageUrl: {\n displayName: \"Source page URL\",\n purpose: \"Link to the page containing the source block.\",\n category: \"system\",\n required: false,\n autodiscoverable: false,\n expectedTypes: [\"url\"]\n },\n sourceLastEditedTime: {\n displayName: \"Source last edited time\",\n purpose: \"When the source block was last modified. Used to detect changes. (Not Notion's \\\"Last edited time\\\", which tracks the database entry, not the source.)\",\n category: \"system\",\n required: false,\n autodiscoverable: false,\n expectedTypes: [\"date\"]\n },\n\n // === ENRICHMENT: Additional context ===\n sourceAuthor: {\n displayName: \"Source author\",\n purpose: \"The Notion user who created the source block.\",\n category: \"enrichment\",\n required: false,\n autodiscoverable: false,\n expectedTypes: [\"people\"]\n },\n otherPeople: {\n displayName: \"Other people\",\n purpose: \"People @mentioned in the source block, in order of appearance.\",\n category: \"enrichment\",\n required: false,\n autodiscoverable: false,\n expectedTypes: [\"people\", \"relation\"]\n },\n links: {\n displayName: \"Links\",\n purpose: \"URLs and page links found in the source block that weren't mapped to relations.\",\n category: \"enrichment\",\n required: false,\n autodiscoverable: false,\n expectedTypes: [\"rich_text\"]\n },\n sourceText: {\n displayName: \"Source text\",\n purpose: \"The raw text from the source block.\",\n category: \"system\",\n required: false,\n autodiscoverable: false,\n expectedTypes: [\"rich_text\"]\n },\n projects: {\n displayName: \"Projects\",\n purpose: \"Links todos to project pages. Supports @mention and contextual matching from source page.\",\n category: \"enrichment\",\n required: false,\n autodiscoverable: true,\n expectedTypes: [\"relation\"],\n namePatterns: [\"project\", \"projects\"]\n }\n};\n\n/**\n * Required property keys - derived from metadata (single source of truth)\n * These properties must be mapped correctly for sync to work\n */\nexport const REQUIRED_PROPERTY_KEYS = Object.entries(PROPERTY_METADATA)\n .filter(([_, meta]) => meta.required)\n .map(([key]) => key);\n\n/**\n * Category metadata - display info for each property category\n */\nexport const CATEGORY_METADATA: Record<string, { title: string; description: string }> = {\n core: {\n title: 'Core',\n description: 'Essential todo properties that define what a todo is'\n },\n system: {\n title: 'System',\n description: 'Infrastructure properties for sync and deduplication'\n },\n enrichment: {\n title: 'Enrichment',\n description: 'Additional context and metadata'\n }\n};\n\n// camelCase → SCREAMING_SNAKE_CASE\nconst toScreamingSnake = (s: string) => s.replace(/([a-z])([A-Z])/g, \"$1_$2\").toUpperCase();\n\n// Get env var, checking exact key first, then SCREAMING_SNAKE_CASE\nfunction getEnv(prefix: string, key: string): string | undefined {\n // Try exact key first: TODOS_PROP_name\n const exactKey = `${prefix}_${key}`;\n if (exactKey in ENV) return ENV[exactKey];\n\n // Try screaming snake: TODOS_PROP_NAME\n const screamingKey = `${prefix}_${toScreamingSnake(key)}`;\n if (screamingKey in ENV) return ENV[screamingKey];\n\n return undefined;\n}\n\n// Build config from defaults + env overrides\nfunction buildConfig<T extends Record<string, string | string[] | null>>(\n prefix: string,\n defaults: T,\n): { [K in keyof T]: T[K] extends string[] ? string[] : string | null } {\n return Object.fromEntries(\n Object.entries(defaults).map(([key, defaultVal]) => {\n const val = getEnv(prefix, key);\n\n if (Array.isArray(defaultVal)) {\n if (val === undefined) return [key, defaultVal];\n if (val === \"\") return [key, []];\n return [key, val.split(\",\").map((s) => s.trim()).filter(Boolean)];\n }\n\n if (val === undefined) return [key, defaultVal]; // keeps null if default is null\n if (val === \"\") return [key, null];\n return [key, val];\n }),\n ) as any;\n}\n\nexport const TODOS_PROPERTIES = buildConfig(\"TODOS_PROP\", TODOS_DEFAULTS);\n\n/**\n * Get the env var name for a property key\n * Returns the screaming snake case version (e.g., \"dueDate\" → \"TODOS_PROP_DUE_DATE\")\n */\nexport function getEnvVarName(key: string): string {\n return `TODOS_PROP_${toScreamingSnake(key)}`;\n}\n\n/**\n * Get the current env var value for a property key (if set)\n */\nexport function getEnvVarValue(key: string): string | null {\n const val = getEnv(\"TODOS_PROP\", key);\n return val === undefined ? null : val;\n}\n\n/**\n * Get the default property name for a key\n */\nexport function getDefaultPropertyName(key: string): string | null {\n return (TODOS_DEFAULTS as Record<string, string | null>)[key] ?? null;\n}\n\n/**\n * Get all property keys (for iteration)\n */\nexport function getPropertyKeys(): string[] {\n return Object.keys(TODOS_DEFAULTS);\n}\n",[],[412,422,439],"a9400ce0-f613-11f0-b05a-42dde27851f2","logger.ts","a9400e02-f613-11f0-b05a-42dde27851f2",263,["D",1768922475258],"/**\n * Simple logger with configurable log levels\n *\n * Set LOG_LEVEL env var to control verbosity:\n * - \"error\": Only errors\n * - \"warn\": Errors and warnings\n * - \"info\": Errors, warnings, and summary info (default, recommended for production)\n * - \"debug\": All logs including verbose per-item details (useful for development)\n */\n\ntype LogLevel = 'error' | 'warn' | 'info' | 'debug';\n\nconst LOG_LEVELS: Record<LogLevel, number> = {\n error: 0,\n warn: 1,\n info: 2,\n debug: 3\n};\n\nfunction getLogLevel(): LogLevel {\n const level = (Deno.env.get('LOG_LEVEL') || 'info').toLowerCase();\n if (level in LOG_LEVELS) {\n return level as LogLevel;\n }\n return 'info';\n}\n\nfunction shouldLog(level: LogLevel): boolean {\n const currentLevel = getLogLevel();\n return LOG_LEVELS[level] <= LOG_LEVELS[currentLevel];\n}\n\nexport const logger = {\n error: (...args: any[]) => {\n if (shouldLog('error')) console.error(...args);\n },\n warn: (...args: any[]) => {\n if (shouldLog('warn')) console.warn(...args);\n },\n info: (...args: any[]) => {\n if (shouldLog('info')) console.log(...args);\n },\n debug: (...args: any[]) => {\n if (shouldLog('debug')) console.log(...args);\n }\n};\n",[],[412,422,447],"1582a266-b459-11f0-a09c-0224a6c84d84","emojiUtils.ts","1582a2fc-b459-11f0-a09c-0224a6c84d84",16,["D",1761695515418],"/**\n * Emoji utility functions for extracting and detecting emojis\n */\n\n/**\n * Extract the first emoji from a text string\n *\n * @param text - Text to search for emoji\n * @returns First emoji character or null if none found\n *\n * @example\n * extractFirstEmoji(\"📝 Write documentation\") // Returns \"📝\"\n * extractFirstEmoji(\"No emoji here\") // Returns null\n */\nexport function extractFirstEmoji(text: string): string | null {\n if (!text) return null;\n\n // Comprehensive emoji regex pattern covering most common emoji ranges\n const emojiRegex = /[\\u{1F300}-\\u{1F9FF}\\u{2600}-\\u{26FF}\\u{2700}-\\u{27BF}\\u{1F000}-\\u{1F02F}\\u{1F0A0}-\\u{1F0FF}\\u{1F100}-\\u{1F64F}\\u{1F680}-\\u{1F6FF}\\u{1F900}-\\u{1F9FF}\\u{1FA00}-\\u{1FA6F}\\u{1FA70}-\\u{1FAFF}]/u;\n\n const match = text.match(emojiRegex);\n return match ? match[0] : null;\n}\n\n/**\n * Check if text contains at least one emoji\n *\n * @param text - Text to check\n * @returns True if emoji found, false otherwise\n */\nexport function hasEmoji(text: string): boolean {\n return extractFirstEmoji(text) !== null;\n}\n",[],[412,422,455],"98409e32-f618-11f0-ba7e-42dde27851f2","dateUtils.ts","b9471dda-b659-11f0-934f-0224a6c84d84",264,["D",1768924594216],"/**\n * Date utility functions for calculating default due dates\n * Pure functions with no external dependencies\n */\n\n/**\n * Supported default due date settings\n */\nexport type DueDateSetting =\n | 'today'\n | 'tomorrow'\n | 'one_week'\n | 'end_of_week'\n | 'next_business_day';\n\n/**\n * Get default due date setting from environment variable\n *\n * Reads DEFAULT_DUE_DATE env var\n * Falls back to \"today\" if not set or invalid\n *\n * @returns Valid due date setting\n */\nexport function getDefaultDueDateSetting(): DueDateSetting {\n const envSetting = Deno.env.get('DEFAULT_DUE_DATE');\n\n if (!envSetting) {\n return 'today'; // Default fallback\n }\n\n const normalized = envSetting.trim().toLowerCase();\n\n // Validate setting\n const validSettings: DueDateSetting[] = [\n 'today',\n 'tomorrow',\n 'one_week',\n 'end_of_week',\n 'next_business_day'\n ];\n\n if (validSettings.includes(normalized as DueDateSetting)) {\n return normalized as DueDateSetting;\n }\n\n console.warn(`⚠ Invalid DEFAULT_DUE_DATE value: \"${envSetting}\". Using \"today\" as fallback.`);\n return 'today';\n}\n\n/**\n * Calculate due date based on setting\n *\n * Returns ISO date string (YYYY-MM-DD format)\n * All calculations are relative to the reference date (defaults to now)\n *\n * @param setting - Due date setting\n * @param referenceDate - Date to calculate from (defaults to now, use block creation time for deterministic results)\n * @returns ISO date string (e.g., \"2025-10-31\")\n */\nexport function calculateDueDate(setting: DueDateSetting, referenceDate?: Date): string {\n const now = referenceDate || new Date();\n let targetDate = new Date(now);\n\n switch (setting) {\n case 'today':\n // Already set to today\n break;\n\n case 'tomorrow':\n targetDate.setDate(targetDate.getDate() + 1);\n break;\n\n case 'one_week':\n targetDate.setDate(targetDate.getDate() + 7);\n break;\n\n case 'end_of_week':\n // Next Friday\n const currentDay = targetDate.getDay(); // 0 = Sunday, 5 = Friday\n let daysUntilFriday = (5 - currentDay + 7) % 7;\n\n // If today is Friday, go to next Friday\n if (daysUntilFriday === 0) {\n daysUntilFriday = 7;\n }\n\n targetDate.setDate(targetDate.getDate() + daysUntilFriday);\n break;\n\n case 'next_business_day':\n // Skip weekends (Saturday = 6, Sunday = 0)\n targetDate.setDate(targetDate.getDate() + 1); // Start with tomorrow\n\n while (targetDate.getDay() === 0 || targetDate.getDay() === 6) {\n targetDate.setDate(targetDate.getDate() + 1);\n }\n break;\n\n default:\n // Fallback to today for any unexpected values\n console.warn(`⚠ Unexpected due date setting: \"${setting}\". Using today.`);\n }\n\n // Format as ISO date string (YYYY-MM-DD)\n return toISODateString(targetDate);\n}\n\n/**\n * Convert Date object to ISO date string (YYYY-MM-DD)\n *\n * @param date - Date object\n * @returns ISO date string\n */\nfunction toISODateString(date: Date): string {\n const year = date.getFullYear();\n const month = String(date.getMonth() + 1).padStart(2, '0');\n const day = String(date.getDate()).padStart(2, '0');\n\n return `${year}-${month}-${day}`;\n}\n",[],[412,422,463],"85da684c-020f-11f1-8e35-42dde27851f2","configUtils.ts","43d0f650-b962-11f0-8efb-0224a6c84d84",["D",1770240111800],"/**\n * Configuration utility functions\n * Read and validate environment variables with sensible defaults\n */\n\n/**\n * Get recent pages lookback window from environment variable\n *\n * Reads RECENT_PAGES_LOOKBACK_HOURS env var to determine how far back\n * to search for recent Notion pages.\n *\n * Used by:\n * - Frontend dashboard initial load\n * - /api/pages/recent endpoint\n * - /tasks/todos batch processing\n * - Cron jobs (todoSearch.cron.ts)\n *\n * @returns Number of hours to look back (positive integer)\n * - Default: 24 if not set or invalid\n * - Must be positive integer\n */\nexport function getRecentPagesLookbackHours(): number {\n const envValue = Deno.env.get('RECENT_PAGES_LOOKBACK_HOURS');\n\n // If not set, return default of 24 hours\n if (envValue === undefined || envValue === null || envValue === '') {\n return 24;\n }\n\n // Parse as integer\n const parsed = parseInt(envValue, 10);\n\n // If invalid, return default\n if (isNaN(parsed) || parsed <= 0) {\n console.warn(`⚠ Invalid RECENT_PAGES_LOOKBACK_HOURS value \"${envValue}\", using default of 24 hours`);\n return 24;\n }\n\n return parsed;\n}\n",[],[412,422,471],"5c0f8d2e-cdf5-11f0-a2c0-42dde27851f2","blockUtils.ts","35489a72-b8ac-11f0-a94f-0224a6c84d84",142,["D",1764511414357],"/**\n * Block stability utilities\n *\n * Provides functions to check if blocks are \"stable\" (not recently edited)\n * Used to prevent saving blocks that are actively being edited to blob storage\n */\n\n/**\n * Get block stability threshold from environment variable\n *\n * Reads BLOCK_STABILITY_MINUTES env var to determine minimum age\n * before a block is considered stable enough to save to blob storage.\n *\n * @returns Number of minutes (0 or positive integer)\n * - Default: 0 if not set (no delay - immediate sync)\n * - Set env var to add a delay (e.g., \"2\" for 2 minutes)\n */\nexport function getBlockStabilityMinutes(): number {\n const envValue = Deno.env.get('BLOCK_STABILITY_MINUTES');\n\n // If not set, return default of 0 (no delay)\n if (envValue === undefined || envValue === null || envValue === '') {\n return 0;\n }\n\n // Parse as integer\n const parsed = parseInt(envValue, 10);\n\n // If invalid, return default\n if (isNaN(parsed) || parsed < 0) {\n console.warn(`⚠ Invalid BLOCK_STABILITY_MINUTES value \"${envValue}\", using default of 0 (no delay)`);\n return 0;\n }\n\n return parsed;\n}\n\n/**\n * Calculate the age of a block in minutes\n *\n * @param lastEditedTime - ISO 8601 timestamp string from Notion\n * @returns Number of minutes since the block was last edited\n */\nexport function getBlockAgeMinutes(lastEditedTime: string): number {\n const lastEditedDate = new Date(lastEditedTime);\n const now = Date.now();\n const ageMs = now - lastEditedDate.getTime();\n const ageMinutes = ageMs / (1000 * 60);\n\n return ageMinutes;\n}\n\n/**\n * Check if a block is \"stable\" (old enough to save)\n *\n * A block is considered stable if:\n * - stabilityMinutes is 0 (delay disabled), OR\n * - Block age >= stabilityMinutes\n *\n * @param lastEditedTime - ISO 8601 timestamp string from Notion\n * @param stabilityMinutes - Minimum age threshold in minutes\n * @returns true if block is stable, false if too recent\n */\nexport function isBlockStable(lastEditedTime: string, stabilityMinutes: number): boolean {\n // If stability check is disabled (0 minutes), always return true\n if (stabilityMinutes === 0) {\n return true;\n }\n\n // Calculate block age\n const ageMinutes = getBlockAgeMinutes(lastEditedTime);\n\n // Block is stable if age meets or exceeds threshold\n return ageMinutes >= stabilityMinutes;\n}\n\n/**\n * Valid Notion background colors\n */\nconst VALID_BACKGROUND_COLORS = new Set([\n 'gray_background',\n 'brown_background',\n 'orange_background',\n 'yellow_background',\n 'green_background',\n 'blue_background',\n 'purple_background',\n 'pink_background',\n 'red_background'\n]);\n\n/**\n * Get block background color from environment variable\n *\n * When set to a valid Notion background color, matched todo blocks\n * will have their background color changed in Notion as a visual indicator.\n *\n * @returns The background color if valid, null otherwise\n */\nexport function getBlockBackgroundColor(): string | null {\n const color = Deno.env.get('BLOCK_BACKGROUND_COLOR')?.toLowerCase();\n if (color && VALID_BACKGROUND_COLORS.has(color)) {\n return color;\n }\n return null;\n}\n",[],[412,422,478],"1582a054-b459-11f0-a09c-0224a6c84d84","blobUtils.ts","1582a0ea-b459-11f0-a09c-0224a6c84d84",["D",1761695515418],"/**\n * Blob storage utility functions\n */\n\n/**\n * Extract block ID from blob key\n * Key format: projectName--category--blockId\n *\n * @param key - Blob key string (e.g., \"demo--todo--abc-123-def-456\")\n * @returns Block ID (e.g., \"abc-123-def-456\")\n *\n * @example\n * extractBlockIdFromKey(\"demo--todo--abc-123\") // Returns \"abc-123\"\n */\nexport function extractBlockIdFromKey(key: string): string {\n const parts = key.split('--');\n return parts[parts.length - 1]; // Last part is the block ID\n}\n",[],[412,422,486],"1582acde-b459-11f0-a09c-0224a6c84d84",["D",1761695515418],[217,218,219,220,221,222,223,224,225,226],[412,422],"fa4ed9f6-b363-11f0-87a0-0224a6c84d84","faviconizerService.ts","b7748178-b361-11f0-b7fc-0224a6c84d84","019a26dc-99bf-7269-a8f6-2bcf04eec442",10,["D",1761590243094],"import type { FaviconizerResponse } from \"../../shared/types.ts\";\n\n/**\n * Get favicon URL for a domain using Faviconizer API\n */\nexport async function getFaviconForDomain(domain: string) {\n try {\n const response = await fetch(\"https://faviconizer.val.run/api/domain\", {\n method: \"POST\",\n headers: {\n \"Content-Type\": \"application/json\",\n },\n body: JSON.stringify({ domain }),\n });\n\n if (!response.ok) {\n return {\n success: false,\n data: null,\n error: `Faviconizer API returned status ${response.status}`,\n };\n }\n\n const data: FaviconizerResponse = await response.json();\n\n // Check if the API's success field is true\n if (!data.success) {\n return {\n success: false,\n data: null,\n error: \"Faviconizer API returned success: false\",\n };\n }\n\n return {\n success: true,\n data: data,\n error: null,\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || \"Failed to fetch favicon\",\n };\n }\n}\n",[],[412,505,497],"services","3f9695f0-f7b4-11f0-bfe0-42dde27851f2","blobService.ts","15829b9a-b459-11f0-a09c-0224a6c84d84",364,["D",1769101398135],12,"import { blob } from \"https://esm.town/v/std/blob\";\nimport { getProjectName } from '../utils/valUtils.ts';\n\n/**\n * Save a block to Val Town blob storage\n *\n * @param category - Category of block (e.g., \"todo\", \"mention\", \"date\")\n * @param blockId - Notion block ID\n * @param data - Block data to save (typically ReducedNotionBlock)\n * @returns {success, key, error}\n */\nexport async function saveBlockToBlob(\n category: string,\n blockId: string,\n data: any\n) {\n try {\n // Construct blob key: projectName--category--blockId\n const projectName = getProjectName();\n const key = `${projectName}--${category}--${blockId}`;\n\n // Save to blob storage (overwrites if exists)\n await blob.setJSON(key, data);\n\n return {\n success: true,\n key: key,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n key: null,\n error: err.message || 'Failed to save to blob storage'\n };\n }\n}\n\n/**\n * Get a block from Val Town blob storage\n * (Future implementation)\n */\nexport async function getBlockFromBlob(\n category: string,\n blockId: string\n) {\n try {\n const projectName = getProjectName();\n const key = `${projectName}--${category}--${blockId}`;\n\n const data = await blob.getJSON(key);\n\n return {\n success: true,\n data: data,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to get from blob storage'\n };\n }\n}\n\n/**\n * Delete a block from Val Town blob storage\n * (Future implementation)\n */\nexport async function deleteBlockFromBlob(\n category: string,\n blockId: string\n) {\n try {\n const projectName = getProjectName();\n const key = `${projectName}--${category}--${blockId}`;\n\n await blob.delete(key);\n\n return {\n success: true,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n error: err.message || 'Failed to delete from blob storage'\n };\n }\n}\n\n/**\n * List all blocks in a category from Val Town blob storage\n *\n * @param category - Category of blocks to list (e.g., \"todo\")\n * @returns {success, keys, error}\n */\nexport async function listBlocksByCategory(category: string) {\n try {\n const projectName = getProjectName();\n const prefix = `${projectName}--${category}--`;\n\n // List all blobs with the prefix\n const result = await blob.list(prefix);\n\n // blob.list() returns an array of objects with 'key' property\n // Extract just the keys as strings\n const keys = Array.isArray(result)\n ? result.map((item: any) => typeof item === 'string' ? item : item.key)\n : [];\n\n return {\n success: true,\n keys: keys,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n keys: null,\n error: err.message || 'Failed to list blobs'\n };\n }\n}\n\n/**\n * Update sync metadata for a blob\n *\n * MERGES with existing sync_metadata to preserve fields not being updated.\n * This prevents accidentally clearing flags like `syncing` or `archived`.\n *\n * @param key - Full blob key (projectName--category--blockId)\n * @param syncMetadata - Sync metadata fields to update (merged with existing)\n * @returns {success, error}\n */\nexport async function updateBlobSyncMetadata(\n key: string,\n syncMetadata: { synced: boolean; syncing?: boolean; target_page_id?: string; archived?: boolean }\n) {\n try {\n // Read existing blob\n const existingData = await blob.getJSON(key);\n\n if (!existingData) {\n return {\n success: false,\n error: 'Blob not found'\n };\n }\n\n // MERGE with existing sync_metadata to preserve fields not being updated\n // This prevents accidentally clearing the syncing lock or other flags\n const updatedData = {\n ...existingData,\n sync_metadata: {\n ...existingData.sync_metadata,\n ...syncMetadata\n }\n };\n\n // Save back to blob\n await blob.setJSON(key, updatedData);\n\n return {\n success: true,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n error: err.message || 'Failed to update blob sync metadata'\n };\n }\n}\n\n// Lock timeout in milliseconds (5 minutes)\n// If a lock is held longer than this, it's considered stale and can be overwritten\nconst SYNC_LOCK_TIMEOUT_MS = 5 * 60 * 1000;\n\n/**\n * Acquire sync lock for a blob (prevents concurrent syncs)\n *\n * Uses optimistic locking pattern with timeout:\n * 1. Read current state\n * 2. If syncing=true AND block was recently edited (< 5 min), return failure\n * 3. If syncing=false/undefined OR lock is stale, acquire lock\n *\n * Uses last_edited_time as proxy for lock age - if block was edited > 5 min ago\n * and syncing is still true, the sync is stuck (should complete in < 1 min).\n *\n * @param key - Full blob key (projectName--category--blockId)\n * @returns {success, error}\n */\nexport async function acquireSyncLock(key: string) {\n try {\n const existingData = await blob.getJSON(key);\n\n if (!existingData) {\n return { success: false, error: 'Blob not found' };\n }\n\n // Check if lock already held AND not stale\n if (existingData.sync_metadata?.syncing === true) {\n // Use last_edited_time as proxy for when sync should have started\n const lastEditedTime = existingData.last_edited_time\n ? new Date(existingData.last_edited_time).getTime()\n : 0;\n const isStale = Date.now() - lastEditedTime > SYNC_LOCK_TIMEOUT_MS;\n\n if (!isStale) {\n return { success: false, error: 'Lock already held' };\n }\n // Lock is stale (block edited > 5 min ago but sync never completed) - allow override\n }\n\n // Acquire lock by setting syncing=true\n const updatedData = {\n ...existingData,\n sync_metadata: {\n ...existingData.sync_metadata,\n syncing: true\n }\n };\n\n await blob.setJSON(key, updatedData);\n return { success: true, error: null };\n } catch (err: any) {\n return { success: false, error: err.message || 'Failed to acquire sync lock' };\n }\n}\n\n/**\n * Release sync lock without marking as synced (for error recovery)\n *\n * Called when sync fails partway through, allowing the block to be retried.\n *\n * @param key - Full blob key (projectName--category--blockId)\n * @returns {success, error}\n */\nexport async function releaseSyncLock(key: string) {\n try {\n const existingData = await blob.getJSON(key);\n\n if (!existingData) {\n return { success: false, error: 'Blob not found' };\n }\n\n const updatedData = {\n ...existingData,\n sync_metadata: {\n ...existingData.sync_metadata,\n syncing: false\n }\n };\n\n await blob.setJSON(key, updatedData);\n return { success: true, error: null };\n } catch (err: any) {\n return { success: false, error: err.message || 'Failed to release sync lock' };\n }\n}\n\n/**\n * Count blobs in a category\n *\n * @param category - Category of blocks to count (e.g., \"todo\")\n * @returns {success, count, prefix, error}\n */\nexport async function countBlobsByCategory(category: string) {\n try {\n const result = await listBlocksByCategory(category);\n\n if (!result.success) {\n return {\n success: false,\n count: 0,\n prefix: null,\n error: result.error\n };\n }\n\n const projectName = getProjectName();\n const prefix = `${projectName}--${category}--`;\n\n return {\n success: true,\n count: result.keys?.length || 0,\n prefix: prefix,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n count: 0,\n prefix: null,\n error: err.message || 'Failed to count blobs'\n };\n }\n}\n\n// ============================================================================\n// CACHE FUNCTIONS\n// ============================================================================\n\n/**\n * Cached data structure stored in blob\n */\ninterface CachedData<T> {\n data: T;\n timestamp: number;\n}\n\n/**\n * Get cached data from blob storage\n *\n * Key format: {projectName}--cache--{cacheKey}\n *\n * @param cacheKey - Cache identifier (e.g., \"health\", \"relations\")\n * @returns Cached data with timestamp, or null if not found\n */\nexport async function getCache<T>(cacheKey: string): Promise<CachedData<T> | null> {\n try {\n const projectName = getProjectName();\n const key = `${projectName}--cache--${cacheKey}`;\n\n const cached = await blob.getJSON(key) as CachedData<T> | null;\n\n if (cached && typeof cached.timestamp === 'number' && cached.data !== undefined) {\n return cached;\n }\n\n return null;\n } catch {\n // Cache miss or error - return null\n return null;\n }\n}\n\n/**\n * Set cached data in blob storage\n *\n * Key format: {projectName}--cache--{cacheKey}\n *\n * @param cacheKey - Cache identifier (e.g., \"health\", \"relations\")\n * @param data - Data to cache\n */\nexport async function setCache<T>(cacheKey: string, data: T): Promise<void> {\n try {\n const projectName = getProjectName();\n const key = `${projectName}--cache--${cacheKey}`;\n\n const cached: CachedData<T> = {\n data,\n timestamp: Date.now()\n };\n\n await blob.setJSON(key, cached);\n } catch {\n // Cache set failure is non-fatal - just log and continue\n console.warn(`Failed to set cache for key: ${cacheKey}`);\n }\n}\n",[],[412,505,507],"1e0a9cd2-03cc-11f1-a1cf-42dde27851f2","aiService.ts","6f6e36ea-cc7e-11f0-a038-42dde27851f2",428,["D",1770431063716],31,"/**\n * AI service for summaries and owner disambiguation using OpenAI\n *\n * Uses OPENAI_API_KEY env var if set (recommended for production).\n * Falls back to Val Town's shared @std/openai (has 10 req/min limit).\n */\nimport { OpenAI as ValTownOpenAI } from \"https://esm.town/v/std/openai\";\nimport OpenAINpm from \"npm:openai@4\";\nimport type { SummaryResult } from '../../shared/types.ts';\n\n// Use own API key if available, otherwise fall back to Val Town's shared client\nfunction getOpenAIClient() {\n const apiKey = Deno.env.get(\"OPENAI_API_KEY\");\n if (apiKey) {\n return new OpenAINpm({ apiKey });\n }\n // Fall back to Val Town's shared OpenAI (has rate limits)\n return new ValTownOpenAI();\n}\n\n/**\n * Validate OpenAI API key by making a simple API call\n * Uses models.list() as a lightweight validation endpoint\n *\n * @returns {success, error?}\n */\nexport async function validateOpenAIConnection(): Promise<{\n success: boolean;\n error?: string;\n}> {\n const apiKey = Deno.env.get('OPENAI_API_KEY');\n if (!apiKey) {\n return { success: false, error: 'OPENAI_API_KEY not set' };\n }\n\n try {\n const openai = getOpenAIClient();\n const response = await openai.models.list();\n // Access .data to force the API call (lazy evaluation)\n const models = response.data;\n if (!models) {\n return { success: false, error: 'No response from OpenAI' };\n }\n return { success: true };\n } catch (err: any) {\n return { success: false, error: err.message || 'Invalid API key' };\n }\n}\n\n/**\n * Use AI to pick the best match from multiple name matches\n *\n * Called when multiple pages' names match text in the todo or parent blocks.\n * AI decides based on semantic similarity between todo text and page names.\n *\n * @param todoText - The todo text for context\n * @param candidates - Array of matching pages with id and name\n * @returns Best matching page ID, or null if no confident match\n */\nexport async function disambiguateOwnerMatch(\n todoText: string,\n candidates: Array<{ id: string; name: string }>\n): Promise<string | null> {\n if (candidates.length === 0) {\n return null;\n }\n\n if (candidates.length === 1) {\n return candidates[0].id;\n }\n\n const openai = getOpenAIClient();\n\n const candidateList = candidates\n .map(p => `- ID: ${p.id} | Name: \"${p.name}\"`)\n .join('\\n');\n\n try {\n const completion = await openai.chat.completions.create({\n model: \"gpt-4o-mini\",\n max_tokens: 100,\n messages: [\n {\n role: \"system\",\n content: `Given a todo and candidate names that partially matched, pick the best match based on context. Return ONLY the ID of the best match, or \"NONE\" if no candidate is clearly the right match.`\n },\n {\n role: \"user\",\n content: `Todo: \"${todoText}\"\\n\\nCandidates:\\n${candidateList}\\n\\nReturn the best matching ID or NONE:`\n }\n ]\n });\n\n let response = completion.choices[0]?.message?.content?.trim();\n\n if (!response || response === \"NONE\" || response.toUpperCase() === \"NONE\") {\n return null;\n }\n\n // Strip common prefixes\n if (response.startsWith('ID: ')) {\n response = response.substring(4);\n }\n\n // Validate response is a known candidate ID\n const matchedCandidate = candidates.find(p => p.id === response);\n return matchedCandidate ? matchedCandidate.id : null;\n\n } catch (err: any) {\n console.error('AI owner disambiguation error:', err.message);\n return null;\n }\n}\n\n/**\n * Context for generating AI-interpreted todo summary\n */\nexport interface SummaryContext {\n todoString: string;\n precedingHeading: string | null;\n peopleMentions: Array<{ name: string | null }>;\n dateMentions: string[];\n matchedProject: string | null;\n sourcePage: string | null;\n defaultDueDate: string; // Formatted default date (use if no date in content)\n}\n\n/**\n * Generate an AI-interpreted, polished version of a todo\n *\n * Takes all raw data and metadata, then synthesizes a clear, standalone\n * todo statement that resolves ambiguities about owner, due date, and context.\n * Also extracts the owner name and due date for downstream processing.\n *\n * Due date handling:\n * - AI examines all dates mentioned in the content\n * - AI determines which date (if any) is the due date\n * - If no date in content, AI uses the defaultDueDate from context\n * - AI always returns a dueDate in ISO format (never null for successful calls)\n * - Both the summary text and dueDate field include the same date for consistency\n *\n * @param context - All available context about the todo\n * @returns Structured result with summary, ownerName, ownerSource, and dueDate\n */\nexport async function generateSummary(context: SummaryContext): Promise<SummaryResult> {\n const openai = getOpenAIClient();\n\n // Build context sections for the prompt\n const sections: string[] = [];\n\n sections.push(`Raw todo: \"${context.todoString}\"`);\n\n if (context.precedingHeading) {\n sections.push(`Preceding heading: \"${context.precedingHeading}\"`);\n }\n\n if (context.peopleMentions.length > 0) {\n const names = context.peopleMentions\n .map(p => p.name)\n .filter(Boolean)\n .join(', ');\n if (names) {\n sections.push(`People @mentioned: ${names}`);\n }\n }\n\n if (context.dateMentions.length > 0) {\n sections.push(`Dates mentioned: ${context.dateMentions.join(', ')}`);\n }\n\n if (context.matchedProject) {\n sections.push(`Related project: \"${context.matchedProject}\"`);\n }\n\n if (context.sourcePage) {\n sections.push(`Source page: \"${context.sourcePage}\"`);\n }\n\n // Always include default due date (AI uses this if no date in content)\n sections.push(`Default due date (use only if no date in todo): ${context.defaultDueDate}`);\n\n const contextBlock = sections.join('\\n');\n\n try {\n const completion = await openai.chat.completions.create({\n model: \"gpt-4o-mini\",\n max_tokens: 200,\n messages: [\n {\n role: \"system\",\n content: `You rewrite todos into clear, standalone statements, identify the owner, extract the due date, and identify mentioned people.\n\nGiven raw todo text and metadata, return JSON with:\n1. summary: A polished todo statement (start with action verb, no owner name)\n2. ownerName: The person who should PERFORM this task (extracted from todo text, or null)\n3. ownerSource: Always null (owner source is determined by the system, not AI)\n4. dueDate: The due date in ISO format (YYYY-MM-DD)\n5. mentionedPeople: Array of people names mentioned in the todo (not @mentions, not the owner)\n\nOwner extraction rules:\n- Look for a person's name at the START of the todo that indicates who should do the task\n- Common patterns: \"Taylor to call Jordan\", \"Casey: review the doc\", \"Riley - follow up on proposal\"\n- The owner is whoever should PERFORM the action, NOT the target of the action\n- \"Email Jordan\" → owner is NOT Jordan (Jordan is the target); owner is null unless prefixed\n- \"Taylor to email Jordan\" → owner IS Taylor (performs the action)\n- \"Taylor to talk to Jordan about the project\" → owner IS Taylor, NOT Jordan (Taylor performs, Jordan is target)\n- \"Talk to Jordan\" → owner is null (no performer specified), Jordan is in mentionedPeople\n- If no clear performer is indicated at the start, ownerName is null\n\nMentioned people rules:\n- Extract ALL people mentioned in the todo text who are NOT the owner\n- Include people who are targets of actions (call, email, ping, etc.) - they are still relevant to the todo\n- \"ping Jordan about the project\" → mentionedPeople: [\"Jordan\"]\n- \"follow up with Morgan and Casey\" → mentionedPeople: [\"Morgan\", \"Casey\"]\n- \"Taylor to call Jordan\" → mentionedPeople: [\"Jordan\"] (Taylor is owner, Jordan is mentioned)\n- \"discuss with Morgan and Riley\" → mentionedPeople: [\"Morgan\", \"Riley\"]\n- Do NOT include places, companies, or common words that happen to look like names\n- Return the FULL name if apparent from context (e.g., \"Jamie R\" with context suggesting \"Jamie Rodriguez\" → use \"Jamie Rodriguez\")\n\nDue date rules:\n- If the todo mentions a specific date, extract it as dueDate\n- If multiple dates are mentioned, use the one that appears to be the deadline/due date\n- If NO date is mentioned in the todo, use the default due date provided in the context\n- ALWAYS return a date in ISO format: YYYY-MM-DD (never return null for dueDate)\n\nSummary format:\n- Start directly with the action verb (do NOT include owner name in summary)\n- Include WHAT the task is about\n- PRESERVE project/page names mentioned in the todo (e.g., \"[Project Alpha]\" → \"...for Project Alpha\")\n- Do NOT include due date in the summary text\n- Owner and due date are returned separately and stored in their own properties\n\nExamples:\n{\"summary\": \"Prepare the quarterly proposal\", \"ownerName\": null, \"ownerSource\": null, \"dueDate\": \"2026-01-09\", \"mentionedPeople\": []}\n{\"summary\": \"Call Jordan about the project status\", \"ownerName\": null, \"ownerSource\": null, \"dueDate\": \"2025-01-15\", \"mentionedPeople\": [\"Jordan\"]}\n{\"summary\": \"Review the Q1 Planning document with Morgan\", \"ownerName\": \"Taylor\", \"ownerSource\": null, \"dueDate\": \"2025-01-15\", \"mentionedPeople\": [\"Morgan\"]}\n{\"summary\": \"Follow up on the proposal\", \"ownerName\": \"Casey\", \"ownerSource\": null, \"dueDate\": \"2025-01-20\", \"mentionedPeople\": []}\n{\"summary\": \"Call Jamie R about the partnership\", \"ownerName\": \"Riley\", \"ownerSource\": null, \"dueDate\": \"2025-01-22\", \"mentionedPeople\": [\"Jamie R\"]}\n\nReturn ONLY valid JSON, no other text.`\n },\n {\n role: \"user\",\n content: contextBlock\n }\n ]\n });\n\n const response = completion.choices[0]?.message?.content?.trim();\n\n if (!response) {\n // Fallback if AI returns nothing\n return {\n summary: context.todoString,\n ownerName: null,\n ownerSource: null,\n dueDate: null,\n mentionedPeople: []\n };\n }\n\n // Try to parse JSON response\n try {\n const parsed = JSON.parse(response);\n return {\n summary: parsed.summary || context.todoString,\n ownerName: parsed.ownerName || null,\n ownerSource: parsed.ownerSource || null,\n dueDate: parsed.dueDate || null,\n mentionedPeople: Array.isArray(parsed.mentionedPeople) ? parsed.mentionedPeople : []\n };\n } catch (parseErr) {\n // If JSON parsing fails, treat entire response as summary (backward compatibility)\n console.warn('AI returned non-JSON response, treating as summary:', response.substring(0, 50));\n return {\n summary: response,\n ownerName: null,\n ownerSource: null,\n dueDate: null,\n mentionedPeople: []\n };\n }\n\n } catch (err: any) {\n console.error('AI summary generation error:', err.message);\n // Fallback to original on error\n return {\n summary: context.todoString,\n ownerName: null,\n ownerSource: null,\n dueDate: null,\n mentionedPeople: []\n };\n }\n}\n",[],[412,505,516],"f6ee89d4-ebd3-11f0-955f-42dde27851f2","users.ts","49d474b8-ea41-11f0-a04d-42dde27851f2","019a26dc-99bf-7269-a8f6-24515668c88f",254,["D",1767795606157],"import { notion } from \"./index.ts\";\n\n/**\n * Workspace user entry for caching\n */\ninterface WorkspaceUser {\n id: string;\n name: string;\n email: string | null;\n}\n\n/**\n * Cache for workspace users\n * Refreshed every 5 minutes to balance freshness and performance\n */\nlet userCache: WorkspaceUser[] | null = null;\nlet cacheTimestamp: number = 0;\nconst CACHE_TTL_MS = 5 * 60 * 1000; // 5 minutes\n\n/**\n * Get all workspace users\n *\n * Fetches all users in the workspace with caching for performance.\n * Requires the integration to have \"Read user information\" capability.\n *\n * @returns Array of workspace users\n */\nexport async function getWorkspaceUsers(): Promise<{\n success: boolean;\n users: WorkspaceUser[];\n error: string | null;\n}> {\n // Check cache freshness\n const now = Date.now();\n if (userCache && (now - cacheTimestamp) < CACHE_TTL_MS) {\n console.log(`getWorkspaceUsers: Returning ${userCache.length} cached users`);\n return {\n success: true,\n users: userCache,\n error: null\n };\n }\n\n console.log('getWorkspaceUsers: Fetching users from Notion API...');\n try {\n const users: WorkspaceUser[] = [];\n let hasMore = true;\n let startCursor: string | undefined = undefined;\n\n while (hasMore) {\n const response = await notion.users.list({\n start_cursor: startCursor,\n page_size: 100\n });\n\n for (const user of response.results) {\n // Only include people (not bots)\n if (user.type === 'person') {\n users.push({\n id: user.id,\n name: user.name || '',\n email: (user as any).person?.email || null\n });\n }\n }\n\n hasMore = response.has_more;\n startCursor = response.next_cursor || undefined;\n }\n\n // Update cache\n userCache = users;\n cacheTimestamp = now;\n\n console.log(`getWorkspaceUsers: Fetched ${users.length} users: ${users.map(u => u.name).join(', ')}`);\n return {\n success: true,\n users,\n error: null\n };\n\n } catch (err: any) {\n // Common error: integration lacks \"Read user information\" capability\n if (err.status === 403) {\n console.warn('getWorkspaceUsers: API access denied (403) - integration may lack \"Read user information\" capability');\n } else {\n console.error(`getWorkspaceUsers: Error fetching users: ${err.message}`);\n }\n\n return {\n success: false,\n users: [],\n error: err.message || 'Failed to get workspace users'\n };\n }\n}\n\n/**\n * Find a workspace user by name\n *\n * Searches workspace users for a matching name.\n * Uses case-insensitive partial matching (e.g., \"Alex\" matches \"Alex Johnson\").\n *\n * @param name - Name to search for\n * @returns Matching user or null\n */\nexport async function findUserByName(name: string): Promise<{\n success: boolean;\n user: WorkspaceUser | null;\n error: string | null;\n}> {\n if (!name) {\n return {\n success: true,\n user: null,\n error: null\n };\n }\n\n const usersResult = await getWorkspaceUsers();\n\n if (!usersResult.success) {\n console.log(`findUserByName(\"${name}\"): getWorkspaceUsers failed - ${usersResult.error}`);\n return {\n success: false,\n user: null,\n error: usersResult.error\n };\n }\n\n const nameLower = name.toLowerCase().trim();\n const searchFirstName = nameLower.split(' ')[0]; // Extract first name from search term\n console.log(`findUserByName: Searching for \"${name}\" (first name: \"${searchFirstName}\") among ${usersResult.users.length} users`);\n\n // Try exact match first\n let match = usersResult.users.find(u =>\n u.name.toLowerCase() === nameLower\n );\n\n if (match) {\n console.log(`findUserByName: Exact match found: \"${match.name}\"`);\n return {\n success: true,\n user: match,\n error: null\n };\n }\n\n // Try partial match (name starts with search term, or search term is first name)\n match = usersResult.users.find(u => {\n const userNameLower = u.name.toLowerCase();\n const userFirstName = userNameLower.split(' ')[0];\n // \"Alex\" matches \"Alex Johnson\"\n // \"Alex Johnson\" matches \"Alex Johnson\"\n // NEW: \"Alex Johnson\" matches \"Alex @ Acme Corp\" (first names match)\n return userNameLower.startsWith(nameLower) ||\n userFirstName === nameLower ||\n userFirstName === searchFirstName; // Match first names\n });\n\n if (match) {\n console.log(`findUserByName: Partial match found: \"${match.name}\"`);\n } else {\n console.log(`findUserByName: No match found for \"${name}\"`);\n }\n\n return {\n success: true,\n user: match || null,\n error: null\n };\n}\n\n/**\n * Find a workspace user by ID\n *\n * Searches cached workspace users for a matching ID.\n * Uses the same cache as getWorkspaceUsers() - no extra API call.\n *\n * @param userId - User ID to search for\n * @returns Matching user or null\n */\nexport async function findUserById(userId: string): Promise<{\n success: boolean;\n user: WorkspaceUser | null;\n error: string | null;\n}> {\n if (!userId) {\n return {\n success: true,\n user: null,\n error: null\n };\n }\n\n const usersResult = await getWorkspaceUsers();\n\n if (!usersResult.success) {\n return {\n success: false,\n user: null,\n error: usersResult.error\n };\n }\n\n const match = usersResult.users.find(u => u.id === userId);\n\n return {\n success: true,\n user: match || null,\n error: null\n };\n}\n\n/**\n * Clear the user cache\n * Useful for testing or forcing a refresh\n */\nexport function clearUserCache(): void {\n userCache = null;\n cacheTimestamp = 0;\n}\n",[],[412,505,533,525],"notion","fd178ad2-b4f0-11f0-8e48-0224a6c84d84","search.ts","fd178b4a-b4f0-11f0-8e48-0224a6c84d84",17,["D",1761760757953],"/**\n * Notion search operations\n */\n\nimport { notion } from \"./index.ts\";\n\n/**\n * Normalize Notion ID by removing hyphens for comparison\n */\nfunction normalizeNotionId(id: string): string {\n return id.replace(/-/g, '');\n}\n\n/**\n * Search for pages edited within the last X hours\n * Filters out:\n * - Archived items\n * - Database objects\n * - Pages inside TODOS_DB_ID database\n *\n * @param hoursAgo - Number of hours to look back (default: 24)\n * @returns {success, data, error} with array of simplified page objects\n */\nexport async function searchRecentPages(hoursAgo: number = 24) {\n try {\n // Get TODOS_DB_ID to filter out those pages\n const todosDbId = Deno.env.get('TODOS_DB_ID');\n\n // Calculate cutoff timestamp\n const cutoffTime = new Date();\n cutoffTime.setHours(cutoffTime.getHours() - hoursAgo);\n\n const results: any[] = [];\n let hasMore = true;\n let startCursor: string | undefined = undefined;\n\n // Paginate through all results\n while (hasMore) {\n const response = await notion.search({\n sort: {\n direction: 'descending',\n timestamp: 'last_edited_time'\n },\n page_size: 100,\n start_cursor: startCursor\n });\n\n // Add results to array\n results.push(...response.results);\n\n // Check if we've reached the cutoff time\n // Since results are sorted by last_edited_time descending,\n // once we hit a page older than cutoff, we can stop\n const lastResult = response.results[response.results.length - 1];\n if (lastResult) {\n const lastEditedTime = new Date(lastResult.last_edited_time);\n if (lastEditedTime < cutoffTime) {\n // We've gone past the cutoff, stop paginating\n hasMore = false;\n break;\n }\n }\n\n hasMore = response.has_more;\n startCursor = response.next_cursor ?? undefined;\n }\n\n // Filter results:\n // 1. Within time range\n // 2. Not archived\n // 3. Only pages (not databases)\n // 4. Not in TODOS_DB_ID database\n const filteredResults = results.filter((item: any) => {\n const lastEditedTime = new Date(item.last_edited_time);\n const isWithinTimeRange = lastEditedTime >= cutoffTime;\n const isNotArchived = !item.archived;\n const isPage = item.object === 'page';\n\n // Filter out pages in the todos database (normalize IDs for comparison)\n const isInTodosDb = todosDbId\n && item.parent?.type === 'database_id'\n && normalizeNotionId(item.parent?.database_id || '') === normalizeNotionId(todosDbId);\n const isNotInTodosDb = !isInTodosDb;\n\n return isWithinTimeRange && isNotArchived && isPage && isNotInTodosDb;\n });\n\n // Simplify the results - extract only needed fields\n const simplifiedResults = filteredResults.map((item: any) => {\n const simplified: any = {\n id: item.id,\n object: item.object, // 'page' or 'database'\n last_edited_time: item.last_edited_time,\n url: item.url\n };\n\n // Extract title from page properties\n const titleProp = Object.values(item.properties || {}).find(\n (prop: any) => prop.type === 'title'\n ) as any;\n\n if (titleProp?.title?.[0]?.plain_text) {\n simplified.title = titleProp.title[0].plain_text;\n } else {\n simplified.title = 'Untitled';\n }\n\n // Extract parent information\n if (item.parent) {\n simplified.parent = {\n type: item.parent.type\n };\n\n // Add the appropriate ID based on parent type\n if (item.parent.type === 'page_id' && item.parent.page_id) {\n simplified.parent.id = item.parent.page_id;\n } else if (item.parent.type === 'database_id' && item.parent.database_id) {\n simplified.parent.id = item.parent.database_id;\n } else if (item.parent.type === 'workspace' && item.parent.workspace) {\n // Workspace parent doesn't have an ID, it's a boolean\n simplified.parent.id = null;\n }\n }\n\n return simplified;\n });\n\n return {\n success: true,\n data: simplifiedResults,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message\n };\n }\n}\n",[],[412,505,533,535],"6ad87a76-fabb-11f0-903c-42dde27851f2","relations.ts","3d6cb198-e71e-11f0-835d-42dde27851f2",394,["D",1769434330639],"import { notion } from \"./index.ts\";\nimport type { RelationPropertyMapping } from \"../../../shared/types.ts\";\n\n/**\n * Discover all relation properties from a database schema\n *\n * Queries the database to find all relation-type properties and their target databases.\n * Used for dynamic relation mapping without manual configuration.\n *\n * @param databaseId - The database to inspect (typically the todos database)\n * @returns Array of relation property mappings with target database IDs\n */\nexport async function discoverRelationProperties(databaseId: string): Promise<{\n success: boolean;\n relations: RelationPropertyMapping[];\n error: string | null;\n}> {\n try {\n const database = await notion.databases.retrieve({ database_id: databaseId });\n const props = (database as any).properties || {};\n\n const relations: RelationPropertyMapping[] = [];\n\n for (const [name, value] of Object.entries(props)) {\n const propAny = value as any;\n if (propAny.type === 'relation' && propAny.relation?.database_id) {\n // Normalize database ID (remove hyphens)\n const targetDbId = propAny.relation.database_id.replace(/-/g, '');\n\n relations.push({\n propertyName: name,\n targetDatabaseId: targetDbId\n });\n }\n }\n\n return {\n success: true,\n relations,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n relations: [],\n error: err.message || 'Failed to discover relation properties'\n };\n }\n}\n\n/**\n * Get the parent database ID for a Notion page\n *\n * Returns the database ID if the page is in a database, or null if\n * the page is a standalone page (parent is workspace or another page).\n *\n * @param pageId - Page ID to lookup\n * @returns Parent database ID (normalized, no hyphens) or null\n */\nexport async function getPageParentDatabaseId(pageId: string): Promise<{\n success: boolean;\n databaseId: string | null;\n error: string | null;\n}> {\n try {\n const page = await notion.pages.retrieve({ page_id: pageId });\n\n const parent = (page as any).parent;\n if (parent?.type === 'database_id') {\n // Normalize database ID (remove hyphens)\n return {\n success: true,\n databaseId: parent.database_id.replace(/-/g, ''),\n error: null\n };\n }\n\n // Parent is not a database (page, workspace, etc.)\n return {\n success: true,\n databaseId: null,\n error: null\n };\n\n } catch (err: any) {\n // Page might be deleted, archived, or inaccessible\n return {\n success: false,\n databaseId: null,\n error: err.message || 'Failed to get page parent database'\n };\n }\n}\n\n/**\n * Batch lookup parent database IDs for multiple pages\n *\n * Efficiently looks up parent database IDs for a list of page IDs.\n * Uses internal caching to avoid duplicate lookups within the same batch.\n *\n * @param pageIds - Array of page IDs to lookup\n * @returns Map of pageId -> parentDatabaseId (null if not in a database)\n */\nexport async function batchGetPageParentDatabaseIds(pageIds: string[]): Promise<{\n success: boolean;\n mapping: Map<string, string | null>;\n error: string | null;\n}> {\n const mapping = new Map<string, string | null>();\n\n // Deduplicate page IDs\n const uniquePageIds = [...new Set(pageIds)];\n\n // Process in parallel with concurrency limit\n const BATCH_SIZE = 5; // Limit concurrent requests to avoid rate limits\n\n for (let i = 0; i < uniquePageIds.length; i += BATCH_SIZE) {\n const batch = uniquePageIds.slice(i, i + BATCH_SIZE);\n\n const results = await Promise.all(\n batch.map(async (pageId) => {\n const result = await getPageParentDatabaseId(pageId);\n return { pageId, databaseId: result.success ? result.databaseId : null };\n })\n );\n\n for (const { pageId, databaseId } of results) {\n mapping.set(pageId, databaseId);\n }\n }\n\n return {\n success: true,\n mapping,\n error: null\n };\n}\n\n/**\n * Get the type of a specific property in a database\n *\n * Used to detect whether a property like \"Owner\" is a people type or relation type.\n * Returns the property type and relation target database ID if applicable.\n *\n * @param databaseId - Database to query\n * @param propertyName - Name of the property to look up\n * @returns Property type and relation target if applicable\n */\nexport async function getPropertyType(databaseId: string, propertyName: string): Promise<{\n success: boolean;\n propertyType: string | null; // 'people', 'relation', 'rich_text', etc.\n relationTargetDbId?: string | null; // Only populated if type is 'relation'\n error: string | null;\n}> {\n try {\n const database = await notion.databases.retrieve({ database_id: databaseId });\n const props = (database as any).properties || {};\n\n const property = props[propertyName];\n\n if (!property) {\n return {\n success: true,\n propertyType: null,\n relationTargetDbId: null,\n error: null\n };\n }\n\n const propertyType = property.type;\n\n // If it's a relation, also return the target database ID\n if (propertyType === 'relation' && property.relation?.database_id) {\n return {\n success: true,\n propertyType,\n relationTargetDbId: property.relation.database_id.replace(/-/g, ''),\n error: null\n };\n }\n\n return {\n success: true,\n propertyType,\n relationTargetDbId: null,\n error: null\n };\n\n } catch (err: any) {\n return {\n success: false,\n propertyType: null,\n relationTargetDbId: null,\n error: err.message || 'Failed to get property type'\n };\n }\n}\n\n/**\n * Get all page IDs from a database\n *\n * Fetches all pages from a database and returns their IDs.\n * Used to build a lookup set for matching mentions to relation targets.\n *\n * @param databaseId - Database to query\n * @returns Set of page IDs (normalized, no hyphens)\n */\nexport async function getDatabasePageIds(databaseId: string): Promise<{\n success: boolean;\n pageIds: string[];\n error: string | null;\n}> {\n try {\n const pageIds: string[] = [];\n let hasMore = true;\n let startCursor: string | undefined = undefined;\n\n while (hasMore) {\n const response = await notion.databases.query({\n database_id: databaseId,\n start_cursor: startCursor,\n page_size: 100\n });\n\n for (const page of response.results) {\n // Normalize page ID (remove hyphens)\n pageIds.push(page.id.replace(/-/g, ''));\n }\n\n hasMore = response.has_more;\n startCursor = response.next_cursor || undefined;\n }\n\n return {\n success: true,\n pageIds,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n pageIds: [],\n error: err.message || 'Failed to get database page IDs'\n };\n }\n}\n",[],[412,505,533,543],"72f1cbee-d5c8-11f0-b312-42dde27851f2","pages.ts","019a26dc-99bf-7269-a8f6-23db6afe7f06",180,["D",1765371734701],"import { notion } from \"./index.ts\";\n\n/**\n * Block types that can contain nested to_do blocks\n * Used to optimize recursive fetching in block type mode\n */\nexport const TODO_CONTAINER_TYPES = new Set([\n 'to_do', // to_do blocks can nest inside other to_do blocks\n 'toggle', // common pattern for organizing todos\n 'heading_1', // toggle headings can contain nested content\n 'heading_2', // toggle headings can contain nested content\n 'heading_3', // toggle headings can contain nested content\n 'column_list', // layout container\n 'column', // layout container\n 'synced_block', // can contain any block type\n 'callout', // can contain nested content\n 'quote', // can contain nested blocks\n 'bulleted_list_item', // can have nested content\n 'numbered_list_item', // can have nested content\n 'template' // can contain any block type\n]);\n\n/**\n * Get a Notion page by ID\n */\nexport async function getNotionPage(pageId: string) {\n try {\n const page = await notion.pages.retrieve({ page_id: pageId });\n\n return {\n success: true,\n data: page,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to fetch Notion page'\n };\n }\n}\n\n/**\n * Update a Notion page's properties\n */\nexport async function updateNotionPage(\n pageId: string,\n properties: any,\n icon?: { type: string; emoji?: string; external?: { url: string } }\n) {\n try {\n const updateData: any = {\n page_id: pageId,\n properties: properties\n };\n\n // Add icon if provided\n if (icon) {\n updateData.icon = icon;\n }\n\n const page = await notion.pages.update(updateData);\n\n return {\n success: true,\n data: page,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to update Notion page'\n };\n }\n}\n\n/**\n * Create a new page in a database\n */\nexport async function createNotionPage(\n databaseId: string,\n properties: any,\n icon?: { type: string; emoji?: string; external?: { url: string } }\n) {\n try {\n const pageData: any = {\n parent: { database_id: databaseId },\n properties: properties\n };\n\n // Add icon if provided\n if (icon) {\n pageData.icon = icon;\n }\n\n const page = await notion.pages.create(pageData);\n\n return {\n success: true,\n data: page,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to create Notion page'\n };\n }\n}\n\n/**\n * Get page blocks (content)\n */\nexport async function getPageBlocks(pageId: string) {\n try {\n const response = await notion.blocks.children.list({ block_id: pageId });\n\n return {\n success: true,\n data: response.results,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to fetch page blocks'\n };\n }\n}\n\n/**\n * Update a Notion page's icon with an external URL\n */\nexport async function updatePageIcon(pageId: string, iconUrl: string) {\n try {\n const page = await notion.pages.update({\n page_id: pageId,\n icon: {\n type: \"external\",\n external: {\n url: iconUrl\n }\n }\n });\n\n return {\n success: true,\n data: page,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to update page icon'\n };\n }\n}\n\n/**\n * Append a paragraph block to a page (for error notes)\n */\nexport async function appendBlockToPage(pageId: string, message: string) {\n try {\n const response = await notion.blocks.children.append({\n block_id: pageId,\n children: [\n {\n object: \"block\",\n type: \"paragraph\",\n paragraph: {\n rich_text: [\n {\n type: \"text\",\n text: {\n content: message\n }\n }\n ]\n }\n }\n ]\n });\n\n return {\n success: true,\n data: response,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to append block to page'\n };\n }\n}\n\n/**\n * Append a callout block to a page indicating the source has additional details\n *\n * @param pageId - The page to append to\n * @param blockUrl - URL to the source block\n */\nexport async function appendHasChildrenCallout(pageId: string, blockUrl: string) {\n try {\n const response = await notion.blocks.children.append({\n block_id: pageId,\n children: [\n {\n object: \"block\",\n type: \"callout\",\n callout: {\n rich_text: [\n {\n type: \"text\",\n text: { content: \"This todo has additional details. \" }\n },\n {\n type: \"text\",\n text: {\n content: \"View details →\",\n link: { url: blockUrl }\n }\n }\n ],\n icon: {\n type: \"external\",\n external: { url: \"https://www.notion.so/icons/info-alternate_blue.svg\" }\n },\n color: \"blue_background\"\n }\n } as any\n ]\n });\n\n return {\n success: true,\n data: response,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to append callout to page'\n };\n }\n}\n\n/**\n * Recursively get all blocks from a page, including nested children\n *\n * @param blockId - Starting block/page ID\n * @param containerFilter - Optional set of block types to recurse into. If provided, only\n * recurses into blocks whose type is in this set. If null/undefined,\n * recurses into all blocks with children.\n */\nexport async function getPageBlocksRecursive(\n blockId: string,\n containerFilter?: Set<string>\n): Promise<any[]> {\n const allBlocks: any[] = [];\n\n try {\n const response = await notion.blocks.children.list({ block_id: blockId });\n\n for (const block of response.results) {\n // Add the current block\n allBlocks.push(block);\n\n // If block has children, check if we should recurse\n if ((block as any).has_children) {\n const blockType = (block as any).type;\n\n // Recurse if no filter, or if block type is in the filter set\n if (!containerFilter || containerFilter.has(blockType)) {\n const children = await getPageBlocksRecursive(block.id, containerFilter);\n allBlocks.push(...children);\n }\n }\n }\n\n return allBlocks;\n } catch (err: any) {\n // Silently skip validation errors for unsupported block types (transcription, audio, etc.)\n // These are expected - Notion API doesn't support all block types\n if (err.code !== 'validation_error') {\n console.error(`Notion API Error for block ${blockId}: [${err.code}] ${err.message}`);\n }\n return allBlocks; // Return what we have so far\n }\n}\n\n/**\n * Get page title from a Notion page\n *\n * Extracts the title from page properties (looks for property with type \"title\")\n * Returns plain text version of the title\n *\n * @param pageId - Notion page ID\n * @returns Object with success, title (or null), and error\n */\nexport async function getPageTitle(pageId: string) {\n try {\n const page = await notion.pages.retrieve({ page_id: pageId });\n\n // Find the title property (it's the property with type \"title\")\n const properties = (page as any).properties;\n\n if (!properties) {\n return {\n success: false,\n title: null,\n error: 'Page has no properties'\n };\n }\n\n // Find the title property\n let titleProperty = null;\n for (const [key, value] of Object.entries(properties)) {\n if ((value as any).type === 'title') {\n titleProperty = value;\n break;\n }\n }\n\n if (!titleProperty) {\n return {\n success: false,\n title: null,\n error: 'Page has no title property'\n };\n }\n\n // Extract plain text from title rich_text array\n const titleRichText = (titleProperty as any).title;\n if (!Array.isArray(titleRichText) || titleRichText.length === 0) {\n return {\n success: true,\n title: 'Untitled', // Notion pages can legitimately have empty titles\n error: null\n };\n }\n\n const title = titleRichText\n .map((rt: any) => rt.plain_text || '')\n .join('');\n\n return {\n success: true,\n title: title || 'Untitled',\n error: null\n };\n\n } catch (err: any) {\n return {\n success: false,\n title: null,\n error: err.message || 'Failed to fetch page title'\n };\n }\n}\n",[],[412,505,533,551],"ffe22f08-02dc-11f1-8dcf-42dde27851f2","index.ts","019a26dc-99bf-7269-a8f6-1efef3029063",423,["D",1770328363391],"import { Client } from \"npm:@notionhq/client@2\";\n\n// Initialize Notion client - shared across all Notion service modules\nexport const notion = new Client({ auth: Deno.env.get(\"NOTION_API_KEY\") });\n\n/**\n * Get the bot user (integration) info\n * Returns integration id, name, and workspace name\n */\nexport async function getBotUser(): Promise<{\n success: boolean;\n data: { id: string; name: string; workspaceName?: string } | null;\n error: string | null;\n}> {\n try {\n const user = await notion.users.me({});\n return {\n success: true,\n data: {\n id: user.id,\n name: user.name || \"Unknown\",\n workspaceName: (user as any).bot?.workspace_name\n },\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || \"Failed to get bot user\"\n };\n }\n}\n\n// Export all page operations\nexport * from \"./pages.ts\";\n\n// Export all database operations\nexport * from \"./databases.ts\";\n\n// Export all search operations\nexport * from \"./search.ts\";\n\n// Export all block operations\nexport * from \"./blocks.ts\";\n\n// Export all relation discovery operations\nexport * from \"./relations.ts\";\n\n// Export all user operations\nexport * from \"./users.ts\";\n",[],[412,505,533,559],"85da0cd0-020f-11f1-8e35-42dde27851f2","databases.ts","019a26dc-99bf-7269-a8f6-1887e096fe60",["D",1770240111800],"import { notion } from \"./index.ts\";\nimport { TODOS_PROPERTIES } from \"../../utils/notionConfig.ts\";\n\n/**\n * Query a Notion database\n */\nexport async function queryNotionDatabase(databaseId: string, filter?: any, sorts?: any) {\n try {\n const response = await notion.databases.query({\n database_id: databaseId,\n filter: filter,\n sorts: sorts\n });\n\n return {\n success: true,\n data: response.results,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to query Notion database'\n };\n }\n}\n\n/**\n * Get a database by ID\n */\nexport async function getNotionDatabase(databaseId: string) {\n try {\n const database = await notion.databases.retrieve({ database_id: databaseId });\n\n return {\n success: true,\n data: database,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to fetch Notion database'\n };\n }\n}\n\n/**\n * Get all page IDs from a database\n * Used to cache project IDs for matching against link_mentions\n *\n * @param databaseId - Notion database ID\n * @returns Object with success, pageIds array (normalized, no hyphens), and error\n */\nexport async function getProjectPageIds(databaseId: string): Promise<{\n success: boolean;\n pageIds: string[];\n error: string | null;\n}> {\n try {\n const pageIds: string[] = [];\n let hasMore = true;\n let startCursor: string | undefined = undefined;\n\n while (hasMore) {\n const response = await notion.databases.query({\n database_id: databaseId,\n page_size: 100,\n start_cursor: startCursor\n });\n\n // Extract and normalize page IDs (remove hyphens for consistent comparison)\n for (const page of response.results) {\n const normalizedId = page.id.replace(/-/g, '');\n pageIds.push(normalizedId);\n }\n\n hasMore = response.has_more;\n startCursor = response.next_cursor ?? undefined;\n }\n\n return {\n success: true,\n pageIds,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n pageIds: [],\n error: err.message || 'Failed to fetch project page IDs'\n };\n }\n}\n\n/**\n * Get property names from a database schema\n * Used by health check to validate property mappings\n *\n * @param databaseId - Notion database ID\n * @returns Object with success, property names array, and error\n */\nexport async function getDatabasePropertyNames(databaseId: string): Promise<{\n success: boolean;\n properties: string[];\n error: string | null;\n}> {\n try {\n const database = await notion.databases.retrieve({ database_id: databaseId });\n const props = (database as any).properties || {};\n\n // Extract property names, excluding buttons (not data, just action triggers)\n const properties: string[] = [];\n for (const [name, value] of Object.entries(props)) {\n const type = (value as any).type;\n if (type !== 'button') {\n properties.push(name);\n }\n }\n\n return {\n success: true,\n properties,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n properties: [],\n error: err.message || 'Failed to fetch database properties'\n };\n }\n}\n\n/**\n * Get all page IDs and names from a database\n * Used for owner name matching in relation mapping\n *\n * @param databaseId - Notion database ID\n * @returns Object with success, array of {id, name}, and error\n */\nexport async function getDatabasePageNames(databaseId: string): Promise<{\n success: boolean;\n pages: Array<{ id: string; name: string }>;\n error: string | null;\n}> {\n try {\n const pages: Array<{ id: string; name: string }> = [];\n let hasMore = true;\n let startCursor: string | undefined = undefined;\n\n while (hasMore) {\n const response = await notion.databases.query({\n database_id: databaseId,\n start_cursor: startCursor,\n page_size: 100\n });\n\n for (const page of response.results) {\n // Extract page ID (normalized, no hyphens)\n const id = page.id.replace(/-/g, '');\n\n // Extract page title from properties\n const props = (page as any).properties || {};\n let name = 'Untitled';\n\n // Find the title property\n for (const [, value] of Object.entries(props)) {\n const propAny = value as any;\n if (propAny.type === 'title' && propAny.title) {\n name = propAny.title.map((t: any) => t.plain_text || '').join('') || 'Untitled';\n break;\n }\n }\n\n pages.push({ id, name });\n }\n\n hasMore = response.has_more;\n startCursor = response.next_cursor || undefined;\n }\n\n return {\n success: true,\n pages,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n pages: [],\n error: err.message || 'Failed to fetch database page names'\n };\n }\n}\n\n/**\n * Query database for a page with specific Block ID\n *\n * @param databaseId - Database ID to query\n * @param blockId - Block ID value to search for\n * @param propertyName - Property name to search in (uses validated name from cache for consistency)\n */\nexport async function queryDatabaseByBlockId(databaseId: string, blockId: string, propertyName?: string) {\n try {\n // Use provided property name, or fall back to configured name\n const propName = propertyName || TODOS_PROPERTIES.sourceBlockId;\n\n if (!propName) {\n return {\n success: false,\n data: null,\n error: 'No Block ID property configured'\n };\n }\n\n const filter = {\n property: propName,\n rich_text: {\n equals: blockId\n }\n };\n\n const response = await notion.databases.query({\n database_id: databaseId,\n filter: filter\n });\n\n // Return first matching page or null\n const page = response.results.length > 0 ? response.results[0] : null;\n\n return {\n success: true,\n data: page,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to query database by Block ID'\n };\n }\n}\n\n/**\n * Status property info for checkbox sync\n *\n * When a todo checkbox is checked, status is set to `doneName`.\n * When unchecked, status is reset to `notDoneName` (first non-done option).\n * This ensures unchecking a todo properly reverts the status.\n */\nexport interface StatusPropertyInfo {\n type: 'status' | 'select' | 'checkbox';\n doneName?: string; // Option name for \"done\" state (e.g., \"Done\", \"Complete\")\n notDoneName?: string; // Option name for \"not done\" state (e.g., \"Not started\", \"To Do\")\n}\n\n/**\n * Get status property info from database schema\n * Used for syncing checkbox state to a status/select/checkbox property\n *\n * @param databaseId - Notion database ID\n * @param propertyName - Name of the status property\n * @returns Status property info or null if not found/unsupported\n */\nexport async function getStatusPropertyInfo(\n databaseId: string,\n propertyName: string\n): Promise<{\n success: boolean;\n data: StatusPropertyInfo | null;\n error: string | null;\n}> {\n try {\n const database = await notion.databases.retrieve({ database_id: databaseId });\n const props = (database as any).properties || {};\n const property = props[propertyName];\n\n if (!property) {\n return {\n success: false,\n data: null,\n error: `Property \"${propertyName}\" not found in database schema`\n };\n }\n\n const propType = property.type;\n\n // Checkbox type - no options needed\n if (propType === 'checkbox') {\n return {\n success: true,\n data: { type: 'checkbox' },\n error: null\n };\n }\n\n // Status or select type - find \"done\" and \"not done\" options\n if (propType === 'status' || propType === 'select') {\n const options = property[propType]?.options || [];\n\n // Patterns for \"done\" state (case-insensitive)\n const donePatterns = ['done', 'complete', 'completed', 'finished'];\n\n let doneName: string | undefined;\n let notDoneName: string | undefined;\n\n for (const option of options) {\n const optionNameLower = (option.name || '').toLowerCase();\n if (donePatterns.includes(optionNameLower)) {\n doneName = option.name;\n } else if (!notDoneName) {\n // First non-done option becomes the default \"not done\" state\n notDoneName = option.name;\n }\n }\n\n if (doneName) {\n return {\n success: true,\n data: {\n type: propType,\n doneName,\n notDoneName // First non-done option (e.g., \"Not started\", \"To Do\")\n },\n error: null\n };\n }\n\n // No \"done\" option found\n const availableOptions = options.map((o: any) => o.name).join(', ');\n return {\n success: false,\n data: null,\n error: `No \"done\" option found in \"${propertyName}\". Available: ${availableOptions}`\n };\n }\n\n // Unsupported property type\n return {\n success: false,\n data: null,\n error: `Property \"${propertyName}\" has unsupported type: ${propType}`\n };\n\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to get status property info'\n };\n }\n}\n\n/**\n * Database schema property (name, type, id)\n */\nexport interface SchemaProperty {\n name: string;\n type: string;\n id: string;\n}\n\n/**\n * Get full database schema with property names and types\n * Used by health check for property validation and autodiscovery\n *\n * @param databaseId - Notion database ID\n * @returns Schema properties array\n */\nexport async function getDatabaseSchema(databaseId: string): Promise<{\n success: boolean;\n properties: SchemaProperty[];\n databaseName: string | null;\n error: string | null;\n}> {\n try {\n const database = await notion.databases.retrieve({ database_id: databaseId });\n const props = (database as any).properties || {};\n\n // Extract database name from title\n const titleArray = (database as any).title || [];\n const databaseName = titleArray.map((t: any) => t.plain_text || '').join('') || null;\n\n // Extract property details, excluding buttons\n const properties: SchemaProperty[] = [];\n for (const [name, value] of Object.entries(props)) {\n const propAny = value as any;\n if (propAny.type !== 'button') {\n properties.push({\n name,\n type: propAny.type,\n id: propAny.id\n });\n }\n }\n\n return {\n success: true,\n properties,\n databaseName,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n properties: [],\n databaseName: null,\n error: err.message || 'Failed to fetch database schema'\n };\n }\n}\n\n/**\n * Find properties by type (for autodiscovery)\n * Returns all properties matching any of the given types\n *\n * @param properties - Schema properties from getDatabaseSchema\n * @param types - Notion property types to match (e.g., [\"status\", \"select\", \"checkbox\"])\n * @returns Matching properties\n */\nexport function findPropertiesByType(\n properties: SchemaProperty[],\n types: string[]\n): SchemaProperty[] {\n return properties.filter(p => types.includes(p.type));\n}\n\n/**\n * Find the title property (every database has exactly one)\n *\n * @param properties - Schema properties from getDatabaseSchema\n * @returns Title property or null if not found\n */\nexport function findTitleProperty(properties: SchemaProperty[]): SchemaProperty | null {\n return properties.find(p => p.type === 'title') || null;\n}\n\n/**\n * Find a property by name (case-insensitive)\n *\n * @param properties - Schema properties from getDatabaseSchema\n * @param name - Property name to find\n * @returns Matching property or null\n */\nexport function findPropertyByName(\n properties: SchemaProperty[],\n name: string\n): SchemaProperty | null {\n const nameLower = name.toLowerCase();\n return properties.find(p => p.name.toLowerCase() === nameLower) || null;\n}\n\n/**\n * Find candidate properties by partial name match\n * Used for suggesting fixes when a configured property is missing\n *\n * @param properties - Schema properties from getDatabaseSchema\n * @param name - Name to match against\n * @returns Matching property names (sorted by relevance)\n */\nexport function findCandidateProperties(\n properties: SchemaProperty[],\n name: string\n): string[] {\n const nameLower = name.toLowerCase();\n const candidates: Array<{ name: string; score: number }> = [];\n\n for (const prop of properties) {\n const propLower = prop.name.toLowerCase();\n\n // Exact match (case-insensitive)\n if (propLower === nameLower) {\n candidates.push({ name: prop.name, score: 100 });\n continue;\n }\n\n // Contains match\n if (propLower.includes(nameLower) || nameLower.includes(propLower)) {\n candidates.push({ name: prop.name, score: 50 });\n continue;\n }\n\n // Word match\n const nameWords = nameLower.split(/\\s+/);\n const propWords = propLower.split(/\\s+/);\n const hasWordMatch = nameWords.some(nw =>\n propWords.some(pw => pw.includes(nw) || nw.includes(pw))\n );\n if (hasWordMatch) {\n candidates.push({ name: prop.name, score: 25 });\n }\n }\n\n // Sort by score (highest first) and return names\n return candidates\n .sort((a, b) => b.score - a.score)\n .map(c => c.name);\n}\n\n/**\n * Split a property name into words\n * Handles spaces, underscores, hyphens, and camelCase\n */\nfunction getWords(name: string): string[] {\n return name\n .replace(/([a-z])([A-Z])/g, '$1 $2') // camelCase → camel Case\n .replace(/[_-]/g, ' ') // snake_case → snake case\n .toLowerCase()\n .split(/\\s+/)\n .filter(Boolean);\n}\n\n/**\n * Find a property matching any of the given name patterns\n * Uses word boundary matching: \"Due by\" matches pattern \"due\", but \"Overdue\" doesn't\n *\n * @param properties - Schema properties from getDatabaseSchema\n * @param patterns - Name patterns to match (e.g., [\"due\", \"deadline\"])\n * @param expectedTypes - Property types to filter by (e.g., [\"date\"])\n * @returns Matching property or null\n */\nexport function findPropertyByNamePatterns(\n properties: SchemaProperty[],\n patterns: string[],\n expectedTypes: string[]\n): SchemaProperty | null {\n const normalizedPatterns = patterns.map(p => p.toLowerCase());\n\n // First pass: exact match (case-insensitive)\n for (const prop of properties) {\n if (!expectedTypes.includes(prop.type)) continue;\n if (normalizedPatterns.includes(prop.name.toLowerCase())) {\n return prop;\n }\n }\n\n // Second pass: word boundary match\n // Property name contains a word that matches a pattern\n for (const prop of properties) {\n if (!expectedTypes.includes(prop.type)) continue;\n const propWords = getWords(prop.name);\n for (const pattern of normalizedPatterns) {\n // Check if any word in the property name matches the pattern exactly\n if (propWords.includes(pattern)) {\n return prop;\n }\n // For multi-word patterns like \"due date\", ALL pattern words must be present\n const patternWords = getWords(pattern);\n if (patternWords.length > 1 && patternWords.every(pw => propWords.includes(pw))) {\n return prop;\n }\n }\n }\n\n return null;\n}\n",[],[412,505,533,567],"6072067c-cdf4-11f0-8ccd-42dde27851f2","blocks.ts","fef229ce-cc7f-11f0-a637-42dde27851f2",139,["D",1764510992220],"import { notion } from \"./index.ts\";\n\n/**\n * Get a single block by ID\n *\n * Used for parent block traversal in project matching\n */\nexport async function getBlock(blockId: string) {\n try {\n const block = await notion.blocks.retrieve({ block_id: blockId });\n\n return {\n success: true,\n data: block,\n error: null\n };\n } catch (err: any) {\n return {\n success: false,\n data: null,\n error: err.message || 'Failed to fetch block'\n };\n }\n}\n\n/**\n * Update a block's background color\n *\n * Valid colors: gray_background, brown_background, orange_background,\n * yellow_background, green_background, blue_background, purple_background,\n * pink_background, red_background, default (no background)\n *\n * @param blockId - The block ID to update\n * @param color - The background color to set\n */\nexport async function updateBlockColor(blockId: string, color: string) {\n try {\n // First retrieve the block to get its type\n const block = await notion.blocks.retrieve({ block_id: blockId });\n const blockType = (block as any).type;\n\n // Update the block with the new color\n await notion.blocks.update({\n block_id: blockId,\n [blockType]: {\n color: color\n }\n });\n\n return { success: true, error: null };\n } catch (err: any) {\n return { success: false, error: err.message || 'Failed to update block color' };\n }\n}\n",[],[412,505,533,574],"1cb962ee-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[206,207,208,209,210,211,212],[412,505,533],"1cb96406-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[205,213,214,215],[412,505],"ebd4a5ce-d51c-11f0-8fa4-42dde27851f2","authCheck.ts","019a26dc-99bf-7269-a8f6-1140d0020917","019a26dc-99bf-7269-a8f6-14fb6878e1e4",151,["D",1765298064073],3,"import { Context, Next } from \"npm:hono@4\";\n\n/**\n * Authentication middleware for Hono routes\n *\n * This middleware checks for authentication credentials in requests.\n * Customize this based on your authentication strategy:\n * - API key in header\n * - Bearer token\n * - Notion webhook signature verification\n * - etc.\n */\nexport async function authCheck(c: Context, next: Next) {\n // Example: Check for API key in header\n const apiKey = c.req.header('x-api-key');\n const expectedKey = Deno.env.get('API_KEY');\n\n // If no API_KEY is configured, skip auth (development mode)\n if (!expectedKey) {\n console.warn('Warning: API_KEY not configured, skipping authentication');\n await next();\n return;\n }\n\n // Validate API key\n if (!apiKey || apiKey !== expectedKey) {\n return c.json({ error: 'Unauthorized', details: 'Invalid or missing API key' }, 401);\n }\n\n // Authentication successful, proceed\n await next();\n}\n\n/**\n * Webhook authentication middleware\n *\n * Protects /tasks/* and /api/* endpoints using X-API-KEY header\n * Header value must match WEBHOOK_SECRET env var\n *\n * Usage in Notion webhooks:\n * Add custom header: X-API-KEY: <your-secret>\n *\n * Usage in API requests:\n * curl -H \"X-API-KEY: <your-secret>\" https://your-val.val.town/api/pages/recent\n */\nexport async function webhookAuth(c: Context, next: Next) {\n const apiKey = c.req.header('x-api-key');\n const webhookSecret = Deno.env.get('WEBHOOK_SECRET');\n\n // If no webhook secret is configured, skip authentication (development mode)\n if (!webhookSecret) {\n console.warn('⚠ WEBHOOK_SECRET not configured, skipping authentication');\n await next();\n return;\n }\n\n // Validate API key\n if (!apiKey || apiKey !== webhookSecret) {\n return c.json({\n error: 'Unauthorized',\n details: 'Invalid or missing X-API-KEY header'\n }, 401);\n }\n\n // Authentication successful\n await next();\n}\n",[],[412,599,590],"routes","1cb956be-b360-11f0-bc6f-0224a6c84d84","views","019a26dc-99bf-7269-a8f6-0e71a3295bf6",["D",1761588582800],[],[412,599,601],"eee93626-d51c-11f0-83d7-42dde27851f2","todos.ts","fd1783e8-b4f0-11f0-8e48-0224a6c84d84","019a26dc-99bf-7269-a8f6-0b691fa0aef1",152,["D",1765298069212],"import { Hono } from \"npm:hono@4\";\nimport * as todoOrchestrationController from '../../controllers/todoOrchestrationController.ts';\n\nconst todos = new Hono();\n\n/**\n * POST /tasks/todos\n * Batch process keyword search across recent pages:\n * 1. Get recent pages (filtered by hours parameter)\n * 2. Search each page for keywords and save to blobs\n * 3. Sync all todo blobs to Notion database\n *\n * Keywords are read from SEARCH_KEYWORDS environment variable (comma-separated)\n * Defaults to \"todo\" if not set\n *\n * Note: This endpoint returns immediately (202 Accepted) and processes in background.\n * This prevents timeout errors when called from Notion webhooks or manual triggers.\n * Check console logs for detailed progress and results.\n *\n * Query parameters:\n * - hours (optional): Number of hours to look back (default: 24)\n *\n * Examples:\n * - POST /tasks/todos (defaults: 24 hours)\n * - POST /tasks/todos?hours=48\n *\n * Response (immediate):\n * {\n * accepted: true,\n * message: \"Processing started in background\",\n * parameters: { hours: 24 }\n * }\n */\ntodos.post('/', async (c) => {\n try {\n // Extract hours parameter from query string\n const hoursParam = c.req.query('hours');\n let hours: number | undefined = undefined;\n\n if (hoursParam) {\n hours = parseInt(hoursParam, 10);\n\n // Validate that it's a valid number\n if (isNaN(hours) || hours <= 0) {\n return c.json({\n success: false,\n error: \"Invalid hours parameter\",\n details: \"hours must be a positive number\"\n }, 400);\n }\n }\n\n // Get request origin for building internal URL\n const origin = new URL(c.req.url).origin;\n\n // Get webhook secret for internal authentication\n const webhookSecret = Deno.env.get('WEBHOOK_SECRET');\n\n // Trigger internal processing endpoint (don't await)\n fetch(`${origin}/tasks/todos/internal`, {\n method: 'POST',\n headers: {\n 'Content-Type': 'application/json',\n 'X-API-KEY': webhookSecret || ''\n },\n body: JSON.stringify({ hours })\n }).catch((err) => {\n // Log fetch errors (not processing errors)\n console.error('Error triggering internal processing:', err);\n });\n\n // Return immediately with 202 Accepted\n return c.json({\n accepted: true,\n message: \"Processing started in background. Check console logs for progress and results.\",\n parameters: {\n hours: hours ?? 24\n }\n }, 202);\n\n } catch (err: any) {\n // Return error for validation failures\n console.error('Error starting batch keyword search processing:', err);\n return c.json({\n success: false,\n error: \"Failed to start processing\",\n details: err.message\n }, 500);\n }\n});\n\n/**\n * POST /tasks/todos/internal\n * Internal endpoint for actual batch processing.\n * Called by /tasks/todos to perform work in background.\n *\n * This endpoint is not intended for external use.\n *\n * Keywords are read from SEARCH_KEYWORDS environment variable (comma-separated)\n * Defaults to \"todo\" if not set\n *\n * Body:\n * {\n * hours?: number\n * }\n */\ntodos.post('/internal', async (c) => {\n try {\n // Extract parameters from body\n const body = await c.req.json();\n const hours = body.hours;\n\n console.log('Internal processing endpoint triggered');\n console.log(`Parameters: hours=${hours ?? 24}`);\n console.log('Keywords from SEARCH_KEYWORDS env var (or default \"todo\")');\n\n // Process batch keyword search (await to ensure completion)\n const result = await todoOrchestrationController.processBatchTodos(hours);\n\n console.log('Internal processing completed successfully');\n console.log(`Result summary: ${result.pagesSearched} pages searched, ${result.totalTodosFound} matches found`);\n\n // Return result (for logging purposes, no one is listening)\n return c.json(result, 200);\n\n } catch (err: any) {\n // Log errors comprehensively\n console.error('ERROR in internal processing:', err);\n console.error('Stack trace:', err.stack);\n\n // Return 200 even on errors (no one is listening anyway)\n return c.json({\n success: false,\n error: err.message,\n details: err.stack\n }, 200);\n }\n});\n\nexport default todos;\n",[],[412,599,615,607],"tasks","15829730-b459-11f0-a09c-0224a6c84d84","todoSave.ts","158297c6-b459-11f0-a09c-0224a6c84d84",["D",1761695515418],"import { Hono } from \"npm:hono@4\";\nimport * as todoSaveController from '../../controllers/todoSaveController.ts';\n\nconst todoSave = new Hono();\n\n// POST /tasks/todo-save - Save todos from blob storage to Notion database\ntodoSave.post('/', async (c) => {\n try {\n // Process all todo blobs and save to Notion\n // Note: This processes ALL blobs, not just one page\n const result = await todoSaveController.processTodoSave();\n\n // Always return 200 OK to webhook\n return c.json({\n success: result.success,\n totalBlobs: result.totalBlobs,\n pagesCreated: result.pagesCreated,\n pagesUpdated: result.pagesUpdated,\n pagesSkipped: result.pagesSkipped,\n pagesFailed: result.pagesFailed,\n failures: result.failures,\n message: result.success\n ? `Processed ${result.totalBlobs} blobs: ${result.pagesCreated} created, ${result.pagesUpdated} updated, ${result.pagesSkipped} skipped, ${result.pagesFailed} failed`\n : `Todo save failed: ${result.error}`\n }, 200);\n\n } catch (err: any) {\n // Even on unexpected errors, return 200 to prevent webhook retries\n console.error('Unexpected error in todo-save webhook:', err);\n return c.json({\n success: false,\n error: \"Internal server error\",\n details: err.message\n }, 200);\n }\n});\n\nexport default todoSave;\n",[],[412,599,615,617],"4391e99e-f790-11f0-bff4-42dde27851f2","todoPage.ts","dfc4f0f4-d5cb-11f0-bcbf-42dde27851f2",350,["D",1769085942893],"import { Hono } from \"npm:hono@4\";\nimport * as todoController from '../../controllers/todoController.ts';\nimport * as todoSaveController from '../../controllers/todoSaveController.ts';\n\nconst todoPage = new Hono();\n\n/**\n * POST /tasks/todos/page\n * Handle Notion webhook for single-page todo search and sync.\n *\n * Returns immediately (202 Accepted) and processes in background.\n * This prevents timeout errors when called from Notion buttons.\n *\n * Keywords are read from SEARCH_KEYWORDS environment variable (comma-separated)\n * Defaults to \"todo\" if not set\n *\n * Searches only the page specified in the webhook payload, then syncs to database\n * so users see todos appear immediately when they click the button in Notion.\n */\ntodoPage.post('/', async (c) => {\n try {\n // Extract payload from request body\n const payload = await c.req.json();\n\n // Extract page_id from webhook payload - try multiple possible locations\n const pageId = payload?.page_id || // Direct property\n payload?.data?.id || // Notion webhook event format\n payload?.page?.id; // Alternative format\n\n if (!pageId) {\n console.error('Failed to extract page_id from payload:', JSON.stringify(payload));\n return c.json({\n error: \"Missing page_id in webhook payload\",\n receivedPayload: payload\n }, 400);\n }\n\n // Get request origin for building internal URL\n const origin = new URL(c.req.url).origin;\n\n // Get webhook secret for internal authentication\n const webhookSecret = Deno.env.get('WEBHOOK_SECRET');\n\n // Trigger internal processing endpoint (don't await)\n fetch(`${origin}/tasks/todos/page/internal`, {\n method: 'POST',\n headers: {\n 'Content-Type': 'application/json',\n 'X-API-KEY': webhookSecret || ''\n },\n body: JSON.stringify({ pageId })\n }).catch((err) => {\n console.error('Error triggering internal processing:', err);\n });\n\n // Return immediately with 202 Accepted\n return c.json({\n accepted: true,\n message: \"Processing started\",\n pageId\n }, 202);\n\n } catch (err: any) {\n // Return error for validation failures\n console.error('Error in todo page webhook:', err);\n return c.json({\n success: false,\n error: \"Failed to start processing\",\n details: err.message\n }, 500);\n }\n});\n\n/**\n * POST /tasks/todos/page/internal\n * Internal endpoint for actual single-page processing.\n * Called by /tasks/todos/page to perform work in background.\n *\n * This endpoint is not intended for external use.\n *\n * Body:\n * {\n * pageId: string\n * }\n */\ntodoPage.post('/internal', async (c) => {\n try {\n const body = await c.req.json();\n const { pageId } = body;\n\n if (!pageId) {\n return c.json({ error: \"Missing pageId\" }, 400);\n }\n\n // Search page for todos (webhooks process immediately, ignoring BLOCK_STABILITY_MINUTES)\n const searchResult = await todoController.processTodoSearch(pageId, false);\n\n // Sync blobs to database so user sees results immediately\n const saveResult = await todoSaveController.processTodoSave();\n\n // Log concise summary\n console.log(`[Page] ${pageId.slice(0, 8)}... → ${searchResult.blocksFound} found, ${saveResult.pagesCreated} created, ${saveResult.pagesUpdated} updated`);\n\n // Return result (for logging purposes)\n return c.json({\n success: searchResult.success,\n blocksFound: searchResult.blocksFound,\n saveResult: {\n pagesCreated: saveResult.pagesCreated,\n pagesUpdated: saveResult.pagesUpdated,\n pagesSkipped: saveResult.pagesSkipped,\n pagesFailed: saveResult.pagesFailed\n }\n }, 200);\n\n } catch (err: any) {\n console.error('ERROR in page processing:', err.message);\n return c.json({\n success: false,\n error: err.message\n }, 200);\n }\n});\n\nexport default todoPage;\n",[],[412,599,615,624],"7a2cba30-b369-11f0-a1ff-0224a6c84d84","favicon.ts","b76fa216-b361-11f0-8f90-0224a6c84d84",["D",1761592605104],"import { Hono } from \"npm:hono@4\";\nimport * as faviconController from '../../controllers/faviconController.ts';\n\nconst favicon = new Hono();\n\n// POST /tasks/favicon - Handle Notion webhook for favicon updates\nfavicon.post('/', async (c) => {\n try {\n // Extract payload from request body\n const payload = await c.req.json();\n\n // Extract page_id from webhook payload - try multiple possible locations\n const pageId = payload?.page_id || // Direct property\n payload?.data?.id || // Notion webhook event format\n payload?.page?.id; // Alternative format\n\n if (!pageId) {\n console.error('Failed to extract page_id from payload:', JSON.stringify(payload));\n return c.json({\n error: \"Missing page_id in webhook payload\",\n receivedPayload: payload\n }, 400);\n }\n\n // Process favicon update (always return 200 to webhook)\n const result = await faviconController.processFaviconUpdate(pageId);\n\n // Always return 200 OK to Notion webhook (even on errors)\n // Errors are logged as notes in the page itself\n return c.json({\n success: result.success,\n iconUpdated: result.iconUpdated,\n errorNoteAdded: result.errorNoteAdded,\n successNoteAdded: result.successNoteAdded,\n message: result.success\n ? \"Favicon updated successfully\"\n : `Favicon update failed: ${result.error}`\n }, 200);\n\n } catch (err: any) {\n // Even on unexpected errors, return 200 to prevent webhook retries\n console.error('Unexpected error in favicon webhook:', err);\n return c.json({\n success: false,\n error: \"Internal server error\",\n details: err.message\n }, 200);\n }\n});\n\nexport default favicon;\n",[],[412,599,615,632],"1cb955b0-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[198,199,200,201],[412,599,615],"fd17b930-b4f0-11f0-8e48-0224a6c84d84","019a26dc-99bf-7269-a8f6-01b5d543f467","019a26dc-99bf-7269-a8f6-0782b134ef20",["D",1761760757953],"import { Hono } from \"npm:hono@4\";\nimport * as pageController from '../../controllers/pageController.ts';\n\nconst pages = new Hono();\n\n/**\n * GET /api/pages/recent\n * Get pages edited within the last X hours\n * Filters out:\n * - Database objects\n * - Archived pages\n * - Pages inside TODOS_DB_ID database\n *\n * Query parameters:\n * - hours (optional): Number of hours to look back (default: 24)\n *\n * Response:\n * {\n * pages: [{ id, object, title, url, last_edited_time, parent }],\n * count: number,\n * timeRange: string\n * }\n */\npages.get('/recent', async (c) => {\n // Extract hours parameter from query string\n const hoursParam = c.req.query('hours');\n let hours: number | undefined = undefined;\n\n if (hoursParam) {\n hours = parseInt(hoursParam, 10);\n\n // Validate that it's a valid number\n if (isNaN(hours)) {\n return c.json({\n error: \"Invalid hours parameter\",\n details: \"hours must be a valid number\"\n }, 400);\n }\n }\n\n // Call controller\n const result = await pageController.getRecentPages(hours);\n\n if (!result.success) {\n return c.json({\n error: result.error,\n details: result.details\n }, 500);\n }\n\n return c.json(result.data);\n});\n\n// GET /api/pages/:id - Get a Notion page\npages.get('/:id', async (c) => {\n const pageId = c.req.param('id');\n const result = await pageController.getPage(pageId);\n\n if (!result.success) {\n return c.json({ error: result.error, details: result.details }, 400);\n }\n\n return c.json(result.data);\n});\n\n// POST /api/pages/:id - Update a Notion page\npages.post('/:id', async (c) => {\n const pageId = c.req.param('id');\n const body = await c.req.json();\n\n const result = await pageController.updatePage(pageId, body.properties);\n\n if (!result.success) {\n return c.json({ error: result.error, details: result.details }, 400);\n }\n\n return c.json(result.data);\n});\n\n// GET /api/pages/:id/content - Get page blocks\npages.get('/:id/content', async (c) => {\n const pageId = c.req.param('id');\n const result = await pageController.getPageContent(pageId);\n\n if (!result.success) {\n return c.json({ error: result.error, details: result.details }, 400);\n }\n\n return c.json(result.data);\n});\n\nexport default pages;\n",[],[412,599,649,551],"api","43d12b70-b962-11f0-8efb-0224a6c84d84","019a26dc-99bf-7269-a8f5-fdd1422bebc3",27,["D",1762249214392],"import { Hono } from \"npm:hono@4\";\nimport pages from './pages.ts';\nimport health from './health.ts';\n\nconst api = new Hono();\n\n// Mount health check at /api/health\napi.route('/health', health);\n\n// Mount page routes at /api/pages\napi.route('/pages', pages);\n\n// Add more routes here as needed\n// Example:\n// import databases from './databases.ts';\n// api.route('/databases', databases);\n\nexport default api;\n",[],[412,599,649,559],"0047e9c0-f80a-11f0-8375-42dde27851f2","health.ts","43d0f498-b962-11f0-8efb-0224a6c84d84",368,["D",1769138228616],"import { Hono } from \"npm:hono@4\";\nimport * as healthController from '../../controllers/healthController.ts';\n\nconst health = new Hono();\n\n/**\n * GET /api/health\n * System health check endpoint\n *\n * Property-centric response showing:\n * - Connection status (Notion API)\n * - Database status (todos database)\n * - All expected properties with their state\n * - Discovered relations with page counts\n * - Settings configuration\n *\n * Use for:\n * - Monitoring system status\n * - Debugging connection issues\n * - Validating property configuration\n */\nhealth.get('/', async (c) => {\n const forceRefresh = c.req.query('refresh') === 'true';\n const result = await healthController.getSystemHealth(forceRefresh);\n return c.json(result);\n});\n\nexport default health;\n",[],[412,599,649,658],"1cb95484-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[194,195,196],[412,599,649],"1cb95e2a-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[193,197,202,203],[412,599],"951b54c2-eb0f-11f0-8d70-42dde27851f2","todoSync.cron.ts","fd17812c-b4f0-11f0-8e48-0224a6c84d84","019a311f-c8bf-720b-97f6-3e820327a344",228,["D",1767711260690],"interval","import * as todoSaveController from '../controllers/todoSaveController.ts';\n\n/**\n * Cron: Sync todo blobs to Notion database\n *\n * Recommended schedule: Less frequent than search cron (e.g., every 8-12 hours)\n */\nexport default async function() {\n if (Deno.env.get(\"CRONS_DISABLED\") === \"true\") {\n console.log(\"[Sync] Skipped - CRONS_DISABLED=true\");\n return;\n }\n\n try {\n const result = await todoSaveController.processTodoSave();\n\n if (!result.success) {\n console.error(`[Sync] ERROR: ${result.error}${result.details ? ` - ${result.details}` : ''}`);\n return;\n }\n\n // Single summary line\n console.log(`[Sync] ${result.pagesCreated} created, ${result.pagesUpdated} updated, ${result.pagesSkipped} skipped, ${result.pagesFailed} failed (${result.totalBlobs} blobs)`);\n\n // Log failures if any\n if (result.failures && result.failures.length > 0) {\n result.failures.forEach((failure: any) => {\n console.error(` ✗ ${failure.blockId}: ${failure.error}`);\n });\n }\n } catch (err: any) {\n console.error(`[Sync] ERROR: ${err.message}`);\n }\n}\n",[],[412,683,674],"crons","951b4aa4-eb0f-11f0-8d70-42dde27851f2","todoSearch.cron.ts","fd177934-b4f0-11f0-8e48-0224a6c84d84",["D",1767711260690],"import * as todoOrchestrationController from '../controllers/todoOrchestrationController.ts';\n\n/**\n * Cron: Search for todos in recent pages and save to blob storage\n *\n * Lookback window: 15 minutes (cron runs frequently)\n * Schedule: Set in Val Town UI (runs every minute)\n */\nexport default async function() {\n if (Deno.env.get(\"CRONS_DISABLED\") === \"true\") {\n console.log(\"[Search] Skipped - CRONS_DISABLED=true\");\n return;\n }\n\n try {\n const result = await todoOrchestrationController.processBatchSearch(0.25);\n\n if (!result.success) {\n console.error(`[Search] ERROR: ${result.error}${result.details ? ` - ${result.details}` : ''}`);\n return;\n }\n\n // Single summary line\n const summary = `[Search] ${result.totalNew || 0} new, ${result.totalUpdated || 0} updated, ${result.totalSkipped || 0} skipped (${result.pagesSearched} pages)`;\n\n // Only log page details when there's actual work\n const pagesWithChanges = result.searchResults.filter(r => r.blocksFound > 0);\n if (pagesWithChanges.length > 0) {\n console.log(summary);\n pagesWithChanges.forEach(page => {\n console.log(` ${page.pageTitle}: ${page.blocksNew || 0} new, ${page.blocksUpdated || 0} updated`);\n });\n } else {\n console.log(summary);\n }\n } catch (err: any) {\n console.error(`[Search] ERROR: ${err.message}`);\n }\n}\n",[],[412,683,685],"83915e98-f7aa-11f0-a5e2-42dde27851f2","cacheWarm.cron.ts","83915f42-f7aa-11f0-a5e2-42dde27851f2",352,["D",1769097217211],"import { getSystemHealth } from '../controllers/healthController.ts';\n\n/**\n * Cron: Warm caches to ensure fast first-load\n *\n * Currently warms:\n * - Health cache (ensures dashboard loads instantly)\n *\n * Future: Can add more cache warming here (relations, project names, etc.)\n *\n * Recommended schedule: Every 1 minute\n */\nexport default async function() {\n // Respects CRONS_DISABLED for \"setup mode\" vs \"production mode\":\n // - Setup mode (CRONS_DISABLED=true): No cache warming, config changes reflect immediately\n // - Production mode: Cache warming enabled, fast dashboard loads\n if (Deno.env.get(\"CRONS_DISABLED\") === \"true\") {\n console.log(\"[CacheWarm] Skipped - CRONS_DISABLED=true\");\n return;\n }\n\n try {\n // Warm health cache by calling getSystemHealth()\n // This checks cache TTL internally - if fresh, returns immediately\n // If stale, builds fresh data and saves to cache\n const result = await getSystemHealth();\n\n const cacheStatus = result.cached ? `hit (age: ${Math.round((result.cacheAge || 0) / 1000)}s)` : 'refreshed';\n console.log(`[CacheWarm] Health cache ${cacheStatus}`);\n } catch (err: any) {\n console.error(`[CacheWarm] ERROR: ${err.message}`);\n }\n}\n",[],[412,683,692],"fd1792f2-b4f0-11f0-8e48-0224a6c84d84",["D",1761760757953],[189,190,191],[412,683],"8972c49c-03d4-11f1-9535-42dde27851f2","todoSaveController.ts","158284c0-b459-11f0-a09c-0224a6c84d84","019a26dc-99bf-7269-a8f5-f9aacbe46421",430,["D",1770434679895],93,"import * as notionService from '../services/notion/index.ts';\nimport * as blobService from '../services/blobService.ts';\nimport * as aiService from '../services/aiService.ts';\nimport { logger } from '../utils/logger.ts';\nimport { discoverRelationProperties, getPropertyType } from '../services/notion/relations.ts';\nimport { getStatusPropertyInfo, getDatabaseSchema, findPropertyByName, type StatusPropertyInfo, type SchemaProperty } from '../services/notion/databases.ts';\nimport { findUserByName, findUserById } from '../services/notion/users.ts';\nimport type { TodoSaveResult, ReducedNotionBlock, RelationPropertyMapping, RelationMatchResult, ResolvedOwner, SummaryResult } from '../../shared/types.ts';\nimport { extractFirstEmoji } from '../utils/emojiUtils.ts';\nimport { getBlockBackgroundColor } from '../utils/blockUtils.ts';\nimport { mapReducedBlockToProperties, extractPageIdFromUrl, formatDateString } from '../utils/notionUtils.ts';\nimport { getDefaultDueDateSetting, calculateDueDate } from '../utils/dateUtils.ts';\nimport { extractBlockIdFromKey } from '../utils/blobUtils.ts';\nimport { TODOS_PROPERTIES, REQUIRED_PROPERTY_KEYS } from '../utils/notionConfig.ts';\n\n// Cache configuration\nconst RELATIONS_CACHE_KEY = 'relations';\nconst RELATIONS_CACHE_TTL_MS = 60_000; // 1 minute\n\n/**\n * Enhance Notion API errors with helpful context about property type mismatches\n */\nfunction enhanceNotionError(error: string): string {\n // Check if this looks like a property type validation error\n if (error.includes('body.properties.') && error.includes('should be')) {\n // Extract property name from error like \"body.properties.Due date.date should be...\"\n const propertyMatch = error.match(/body\\.properties\\.([^.]+)/);\n const propertyName = propertyMatch?.[1] || 'unknown';\n\n return `Property type mismatch for \"${propertyName}\". ` +\n `Check that the column type in your Notion database matches what the code expects. ` +\n `Original error: ${error}`;\n }\n\n // Check for validation errors\n if (error.includes('validation') || error.includes('Invalid')) {\n return `Notion API validation error. This may indicate a property type mismatch ` +\n `between your database schema and the code. Original error: ${error}`;\n }\n\n return error;\n}\n\n/**\n * Page name entry for owner name matching\n */\ntype PageNameEntry = {\n id: string;\n name: string;\n nameLower: string; // Pre-computed lowercase for matching\n};\n\n/**\n * Relation schema cache - initialized once per save operation\n * Contains discovered relations and their target database page IDs\n */\ntype RelationCache = {\n // Discovered relations from todos database schema\n relations: RelationPropertyMapping[];\n // Map of targetDatabaseId -> Set of page IDs in that database (for fast ID lookup)\n targetDbPageIds: Map<string, Set<string>>;\n // Map of targetDatabaseId -> page name entries (for heading/name matching)\n relationPageNames: Map<string, PageNameEntry[]>;\n // Owner property configuration (detected at runtime)\n ownerPropertyType: 'people' | 'relation' | null;\n ownerTargetDbId?: string; // Only populated if owner is relation type\n // Other people property configuration (detected at runtime)\n otherPeoplePropertyType: 'people' | 'relation' | null;\n otherPeopleTargetDbId?: string; // Only populated if otherPeople is relation type\n // Status property configuration for checkbox sync (detected at runtime)\n statusInfo: StatusPropertyInfo | null;\n // Validated property names - only set properties exist in database\n // key = TODOS_PROPERTIES key, value = actual column name or null if not found\n validatedProperties: Record<string, string | null>;\n};\n\n/**\n * Build relation cache by discovering relations from todos database schema (no caching)\n *\n * @param todosDbId - Todos database ID\n * @returns Relation cache or null if discovery fails\n */\nasync function buildRelationCache(todosDbId: string): Promise<RelationCache | null> {\n // Wave 1: Get database schema (required first for property validation)\n const schemaResult = await getDatabaseSchema(todosDbId);\n if (!schemaResult.success) {\n logger.error(`Failed to get database schema: ${schemaResult.error}`);\n return null;\n }\n\n // Validate all configured properties against database schema\n const validatedProperties: Record<string, string | null> = {};\n\n for (const [key, configuredName] of Object.entries(TODOS_PROPERTIES)) {\n if (configuredName && typeof configuredName === 'string') {\n const found = findPropertyByName(schemaResult.properties, configuredName);\n validatedProperties[key] = found ? found.name : null;\n } else {\n validatedProperties[key] = null; // Property not configured\n }\n }\n\n // Check if required properties exist (REQUIRED_PROPERTY_KEYS from notionConfig.ts)\n const missingRequired = REQUIRED_PROPERTY_KEYS.filter(k => {\n const configuredName = (TODOS_PROPERTIES as Record<string, any>)[k];\n return configuredName && !validatedProperties[k];\n });\n\n if (missingRequired.length > 0) {\n const missing = missingRequired.map(k => `${k}=\"${(TODOS_PROPERTIES as Record<string, any>)[k]}\"`).join(', ');\n logger.error(`Sync cannot proceed - required properties not found: ${missing}. Run health check for fix suggestions.`);\n return null;\n }\n\n // Log property validation summary\n const foundProps = Object.entries(validatedProperties).filter(([k, v]) => v && (TODOS_PROPERTIES as Record<string, any>)[k]);\n const missingProps = Object.entries(validatedProperties).filter(([k, v]) => !v && (TODOS_PROPERTIES as Record<string, any>)[k]);\n logger.info(`Properties: ${foundProps.length} found${missingProps.length > 0 ? `, ${missingProps.length} skipped` : ''}`);\n\n // Debug: show individual property mappings\n for (const [key, validatedName] of Object.entries(validatedProperties)) {\n const configuredName = (TODOS_PROPERTIES as Record<string, any>)[key];\n if (configuredName && typeof configuredName === 'string') {\n if (validatedName) {\n logger.debug(` ✓ ${key}: \"${configuredName}\" → found`);\n } else {\n logger.debug(` ⚠ ${key}: \"${configuredName}\" → not found (will skip)`);\n }\n }\n }\n\n // Wave 2: Parallel fetch of relations, owner type, otherPeople type, and status info\n // These are independent and can run concurrently after schema is confirmed\n const [relationsResult, ownerTypeResult, otherPeopleTypeResult, statusResult] = await Promise.all([\n discoverRelationProperties(todosDbId),\n TODOS_PROPERTIES.owner\n ? getPropertyType(todosDbId, TODOS_PROPERTIES.owner)\n : Promise.resolve(null),\n TODOS_PROPERTIES.otherPeople\n ? getPropertyType(todosDbId, TODOS_PROPERTIES.otherPeople)\n : Promise.resolve(null),\n TODOS_PROPERTIES.status\n ? getStatusPropertyInfo(todosDbId, TODOS_PROPERTIES.status)\n : Promise.resolve(null)\n ]);\n\n if (!relationsResult.success) {\n logger.warn(`Failed to discover relations: ${relationsResult.error}`);\n return null;\n }\n\n // Note: cache.relations is set after filtering below (relationsToLoad)\n const cache: RelationCache = {\n relations: [], // Will be set to relationsToLoad after filtering\n targetDbPageIds: new Map(),\n relationPageNames: new Map(),\n ownerPropertyType: null,\n ownerTargetDbId: undefined,\n otherPeoplePropertyType: null,\n otherPeopleTargetDbId: undefined,\n statusInfo: null,\n validatedProperties\n };\n\n // Process owner type result\n if (ownerTypeResult?.success && ownerTypeResult.propertyType) {\n if (ownerTypeResult.propertyType === 'people') {\n cache.ownerPropertyType = 'people';\n } else if (ownerTypeResult.propertyType === 'relation') {\n cache.ownerPropertyType = 'relation';\n cache.ownerTargetDbId = ownerTypeResult.relationTargetDbId || undefined;\n }\n logger.debug(`Owner property \"${TODOS_PROPERTIES.owner}\" is type: ${cache.ownerPropertyType}`);\n }\n\n // Process otherPeople type result\n if (otherPeopleTypeResult?.success && otherPeopleTypeResult.propertyType) {\n if (otherPeopleTypeResult.propertyType === 'people') {\n cache.otherPeoplePropertyType = 'people';\n } else if (otherPeopleTypeResult.propertyType === 'relation') {\n cache.otherPeoplePropertyType = 'relation';\n cache.otherPeopleTargetDbId = otherPeopleTypeResult.relationTargetDbId || undefined;\n }\n logger.debug(`Other people property \"${TODOS_PROPERTIES.otherPeople}\" is type: ${cache.otherPeoplePropertyType}`);\n }\n\n // Process status result\n if (statusResult?.success && statusResult.data) {\n cache.statusInfo = statusResult.data;\n const stateInfo = statusResult.data.doneName\n ? ` (done=\"${statusResult.data.doneName}\"${statusResult.data.notDoneName ? `, notDone=\"${statusResult.data.notDoneName}\"` : ''})`\n : '';\n logger.info(`Status property \"${TODOS_PROPERTIES.status}\" is type: ${statusResult.data.type}${stateInfo}`);\n } else if (TODOS_PROPERTIES.status && statusResult) {\n logger.warn(`Status property \"${TODOS_PROPERTIES.status}\": ${statusResult.error}`);\n }\n\n // Wave 3: Filter to only mapped relations, then fetch their target databases\n // Build set of mapped relation property names from TODOS_PROPERTIES\n const mappedRelationNames = new Set(\n Object.values(TODOS_PROPERTIES).filter((v): v is string => typeof v === 'string' && v !== '')\n );\n\n // Filter discovered relations to only those with explicit TODOS_PROP_xxx mappings\n const relationsToLoad = relationsResult.relations.filter(\n relation => mappedRelationNames.has(relation.propertyName)\n );\n\n logger.debug(`Filtered relations: ${relationsToLoad.length} mapped of ${relationsResult.relations.length} discovered`);\n\n // Set cache.relations to the filtered list\n cache.relations = relationsToLoad;\n\n // Fetch page names for mapped relations only\n // Used for: mention matching (IDs), heading matching (names), owner resolution (names)\n const relationDataPromises = relationsToLoad.map(async (relation) => {\n const namesResult = await notionService.getDatabasePageNames(relation.targetDatabaseId);\n return { relation, namesResult };\n });\n\n const relationDataResults = await Promise.all(relationDataPromises);\n\n // Process relation data results - build both ID sets and name arrays\n for (const { relation, namesResult } of relationDataResults) {\n if (namesResult.success) {\n // Build page name entries\n const entries: PageNameEntry[] = namesResult.pages.map(p => ({\n id: p.id,\n name: p.name,\n nameLower: p.name.toLowerCase()\n }));\n cache.relationPageNames.set(relation.targetDatabaseId, entries);\n\n // Also build ID set for fast lookup during mention matching\n cache.targetDbPageIds.set(\n relation.targetDatabaseId,\n new Set(entries.map(e => e.id))\n );\n }\n }\n\n // Log mapped relations\n logger.info(`Relations: ${cache.relations.length} mapped of ${relationsResult.relations.length} discovered`);\n for (const rel of cache.relations) {\n const pageCount = cache.targetDbPageIds.get(rel.targetDatabaseId)?.size || 0;\n logger.debug(` - ${rel.propertyName} → ${pageCount} pages`);\n }\n\n return cache;\n}\n\n/**\n * Initialize relation cache with blob caching\n *\n * Caching:\n * - Caches relation data in blob storage for 1 minute\n * - Returns cached data if available and fresh\n * - Fetches fresh data if cache is stale or missing\n *\n * Note: Maps are serialized to arrays in blob storage and restored on retrieval\n *\n * @param todosDbId - Todos database ID\n * @returns Relation cache or null if discovery fails\n */\nasync function initializeRelationCache(todosDbId: string): Promise<RelationCache | null> {\n // Check cache first\n const cached = await blobService.getCache<SerializedRelationCache>(RELATIONS_CACHE_KEY);\n\n if (cached && Date.now() - cached.timestamp < RELATIONS_CACHE_TTL_MS) {\n const cacheAge = Math.round((Date.now() - cached.timestamp) / 1000);\n logger.info(`Using cached relation data (age: ${cacheAge}s)`);\n\n // Restore Maps from serialized arrays\n return deserializeRelationCache(cached.data);\n }\n\n // Cache miss or stale - fetch fresh data\n logger.debug('Relation cache miss - fetching fresh data');\n const freshCache = await buildRelationCache(todosDbId);\n\n if (freshCache) {\n // Serialize Maps to arrays for blob storage\n const serialized = serializeRelationCache(freshCache);\n await blobService.setCache(RELATIONS_CACHE_KEY, serialized);\n logger.debug('Relation cache saved to blob storage');\n }\n\n return freshCache;\n}\n\n/**\n * Serialized relation cache for blob storage\n * Maps are converted to arrays of [key, value] tuples\n */\ntype SerializedRelationCache = {\n relations: RelationPropertyMapping[];\n targetDbPageIds: Array<[string, string[]]>;\n relationPageNames: Array<[string, PageNameEntry[]]>;\n ownerPropertyType: 'people' | 'relation' | null;\n ownerTargetDbId?: string;\n otherPeoplePropertyType: 'people' | 'relation' | null;\n otherPeopleTargetDbId?: string;\n statusInfo: StatusPropertyInfo | null;\n validatedProperties: Record<string, string | null>;\n};\n\n/**\n * Serialize RelationCache for blob storage (Maps → Arrays)\n */\nfunction serializeRelationCache(cache: RelationCache): SerializedRelationCache {\n return {\n relations: cache.relations,\n targetDbPageIds: Array.from(cache.targetDbPageIds.entries()).map(\n ([key, set]) => [key, Array.from(set)]\n ),\n relationPageNames: Array.from(cache.relationPageNames.entries()),\n ownerPropertyType: cache.ownerPropertyType,\n ownerTargetDbId: cache.ownerTargetDbId,\n otherPeoplePropertyType: cache.otherPeoplePropertyType,\n otherPeopleTargetDbId: cache.otherPeopleTargetDbId,\n statusInfo: cache.statusInfo,\n validatedProperties: cache.validatedProperties\n };\n}\n\n/**\n * Deserialize RelationCache from blob storage (Arrays → Maps)\n */\nfunction deserializeRelationCache(serialized: SerializedRelationCache): RelationCache {\n return {\n relations: serialized.relations,\n targetDbPageIds: new Map(\n serialized.targetDbPageIds.map(([key, arr]) => [key, new Set(arr)])\n ),\n relationPageNames: new Map(serialized.relationPageNames),\n ownerPropertyType: serialized.ownerPropertyType,\n ownerTargetDbId: serialized.ownerTargetDbId,\n otherPeoplePropertyType: serialized.otherPeoplePropertyType ?? null,\n otherPeopleTargetDbId: serialized.otherPeopleTargetDbId,\n statusInfo: serialized.statusInfo,\n validatedProperties: serialized.validatedProperties\n };\n}\n\n/**\n * Match a block to relations dynamically\n *\n * Strategies (in priority order):\n * 1a. Explicit mention mapping - @mention a page, check its parent DB against relation targets\n * 1b. Heading text matching - preceding heading exactly matches a page name in a relation target\n * 2. Contextual mapping - Source page's DB matches a relation target (fallback)\n *\n * @param blockData - Reduced block with enriched link_mentions\n * @param cache - Relation cache\n * @returns Array of relation match results\n */\nasync function matchBlockToRelations(\n blockData: ReducedNotionBlock,\n cache: RelationCache | null\n): Promise<RelationMatchResult[]> {\n if (!cache || cache.relations.length === 0) {\n return [];\n }\n\n const results: RelationMatchResult[] = [];\n const matchedMentionPageIds = new Set<string>();\n\n // Strategy 1a: Explicit mention mapping\n // Check each link mention's parentDatabaseId against relation targets\n for (const linkMention of blockData.link_mentions || []) {\n if (!linkMention.pageId || !linkMention.parentDatabaseId) continue;\n\n // Find which relation this mention's parent DB matches\n for (const relation of cache.relations) {\n if (linkMention.parentDatabaseId === relation.targetDatabaseId) {\n // Verify the page exists in the target database\n const targetPageIds = cache.targetDbPageIds.get(relation.targetDatabaseId);\n if (targetPageIds?.has(linkMention.pageId)) {\n // This mention matches this relation\n let existingResult = results.find(r => r.propertyName === relation.propertyName);\n if (!existingResult) {\n existingResult = { propertyName: relation.propertyName, pageIds: [], matchType: 'mention' };\n results.push(existingResult);\n }\n if (!existingResult.pageIds.includes(linkMention.pageId)) {\n existingResult.pageIds.push(linkMention.pageId);\n }\n matchedMentionPageIds.add(linkMention.pageId);\n }\n break; // A mention can only match one relation\n }\n }\n }\n\n // Strategy 1b: Heading text matching\n // Treat preceding heading text as an explicit match if it exactly matches a page name\n if (blockData.preceding_heading) {\n const headingLower = blockData.preceding_heading.toLowerCase().trim();\n\n for (const relation of cache.relations) {\n const pageNames = cache.relationPageNames.get(relation.targetDatabaseId);\n if (!pageNames) continue;\n\n // Find exact match for heading text\n const matchingPage = pageNames.find(p => p.nameLower === headingLower);\n if (matchingPage) {\n // Add to results (alongside any existing matches for this relation)\n let existingResult = results.find(r => r.propertyName === relation.propertyName);\n if (!existingResult) {\n existingResult = { propertyName: relation.propertyName, pageIds: [], matchType: 'mention' };\n results.push(existingResult);\n }\n if (!existingResult.pageIds.includes(matchingPage.id)) {\n existingResult.pageIds.push(matchingPage.id);\n logger.debug(`Heading \"${blockData.preceding_heading}\" matched to \"${matchingPage.name}\" in ${relation.propertyName}`);\n }\n }\n }\n }\n\n // Strategy 2: Contextual mapping - FALLBACK\n // Uses source page metadata to match relations not already matched by Strategy 1\n\n // Strategy 2a: Source page has relation properties with values\n // e.g., a Meeting Notes page has \"Project: Acme Redesign\" filled in\n // WHITELIST: Only Projects gets contextual matching - other relations require explicit @mentions\n if (blockData.source_page_relations && TODOS_PROPERTIES.projects) {\n for (const relation of cache.relations) {\n if (results.some(r => r.propertyName === relation.propertyName)) continue;\n if (relation.propertyName !== TODOS_PROPERTIES.projects) continue; // Only Projects gets contextual\n\n const targetPageIds = cache.targetDbPageIds.get(relation.targetDatabaseId);\n if (!targetPageIds) continue;\n\n // Check all source page relation values for page IDs in this target DB\n for (const [_propName, pageIds] of Object.entries(blockData.source_page_relations)) {\n const matchingIds = pageIds.filter(id => targetPageIds.has(id));\n if (matchingIds.length > 0) {\n results.push({\n propertyName: relation.propertyName,\n pageIds: matchingIds,\n matchType: 'contextual',\n });\n break; // One match per relation\n }\n }\n }\n }\n\n // Strategy 2b: Source page is a direct child of a target database\n // e.g., todo lives on a page that IS an entry in the Projects database\n // Only fires when source_page_database_id is set (parent.type === 'database_id')\n // WHITELIST: Only Projects gets contextual matching\n if (blockData.source_page_database_id && TODOS_PROPERTIES.projects) {\n for (const relation of cache.relations) {\n if (blockData.source_page_database_id === relation.targetDatabaseId) {\n if (results.some(r => r.propertyName === relation.propertyName)) continue;\n if (relation.propertyName !== TODOS_PROPERTIES.projects) continue; // Only Projects gets contextual\n\n const sourcePageId = extractPageIdFromUrl(blockData.page_url || '');\n if (sourcePageId) {\n results.push({\n propertyName: relation.propertyName,\n pageIds: [sourcePageId],\n matchType: 'contextual'\n });\n }\n }\n }\n }\n\n // Strategy 2c: Parent page traversal for child pages\n // When source_page_database_id is null (page is a child page, not a database entry),\n // walk up parent pages to find an ancestor that matches via 2a or 2b logic\n // WHITELIST: Only Projects gets contextual matching\n if (!blockData.source_page_database_id && blockData.page_url && TODOS_PROPERTIES.projects) {\n const sourcePageId = extractPageIdFromUrl(blockData.page_url);\n if (sourcePageId) {\n const ancestorMatches = await traverseParentPagesForContextualMatch(\n sourcePageId, cache, new Set(results.map(r => r.propertyName)), TODOS_PROPERTIES.projects\n );\n results.push(...ancestorMatches);\n }\n }\n\n return results;\n}\n\n/**\n * Find a contact by name using exact word matching\n * Used to map workspace users (Owner, @mentions) to Contact pages\n *\n * @param name - Person name to search for\n * @param contactNames - Array of contact page name entries\n * @returns Matching contact entry or null\n */\nfunction findContactByName(name: string, contactNames: PageNameEntry[]): PageNameEntry | null {\n const nameLower = name.toLowerCase();\n const nameWords = nameLower.split(/\\s+/);\n\n // Exact match first\n const exactMatch = contactNames.find(c => c.nameLower === nameLower);\n if (exactMatch) return exactMatch;\n\n // Word match (first name or last name)\n for (const contact of contactNames) {\n const contactWords = contact.nameLower.split(/\\s+/);\n if (nameWords.some(nw => contactWords.some(cw => cw === nw && nw.length >= 2))) {\n return contact;\n }\n }\n\n return null;\n}\n\n/**\n * Add resolved owner and @mentioned people to Contacts relation\n * This preserves connections even if Owner is later changed in the todos database\n *\n * @param resolvedOwner - Resolved owner (people or relation type)\n * @param peopleMentions - @mentions from block text\n * @param relationMatches - Existing relation matches to merge into\n * @param cache - Relation cache with contact names\n */\nasync function addPeopleToContactsRelation(\n resolvedOwner: ResolvedOwner | null,\n peopleMentions: Array<{id: string; name: string | null}>,\n relationMatches: RelationMatchResult[],\n cache: RelationCache\n): Promise<void> {\n // Use the Owner relation's target database for contacts matching\n if (!cache.ownerTargetDbId) return;\n\n const contactNames = cache.relationPageNames.get(cache.ownerTargetDbId);\n if (!contactNames || contactNames.length === 0) return;\n\n // Find the property name for the Owner relation (by property name, not target DB)\n // Multiple relations may target the same DB (e.g., DRI and Steels DB both target Contacts)\n const contactsPropName = TODOS_PROPERTIES.owner;\n if (!contactsPropName) return;\n\n const contactIdsToAdd: string[] = [];\n\n // 1. Add Owner to Contacts\n if (resolvedOwner) {\n if (resolvedOwner.type === 'relation') {\n // Owner is already a page ID - add directly\n contactIdsToAdd.push(resolvedOwner.value[0].id);\n logger.debug(`Adding owner (relation) to Contacts: ${resolvedOwner.value[0].id}`);\n } else if (resolvedOwner.type === 'people') {\n // Owner is a workspace user - look up name and find matching contact\n const userId = resolvedOwner.value[0].id;\n const userResult = await findUserById(userId);\n if (userResult.success && userResult.user?.name) {\n const match = findContactByName(userResult.user.name, contactNames);\n if (match) {\n contactIdsToAdd.push(match.id);\n logger.debug(`Adding owner \"${userResult.user.name}\" to Contacts: ${match.name}`);\n }\n }\n }\n }\n\n // 2. Add Other People (@mentions) to Contacts\n for (const mention of peopleMentions) {\n if (mention.name) {\n const match = findContactByName(mention.name, contactNames);\n if (match && !contactIdsToAdd.includes(match.id)) {\n contactIdsToAdd.push(match.id);\n logger.debug(`Adding @mention \"${mention.name}\" to Contacts: ${match.name}`);\n }\n }\n }\n\n // 3. Merge into existing relation match (deduplicated)\n if (contactIdsToAdd.length > 0) {\n let existingMatch = relationMatches.find(r => r.propertyName === contactsPropName);\n if (!existingMatch) {\n existingMatch = { propertyName: contactsPropName, pageIds: [], matchType: 'mention' };\n relationMatches.push(existingMatch);\n }\n for (const id of contactIdsToAdd) {\n if (!existingMatch.pageIds.includes(id)) {\n existingMatch.pageIds.push(id);\n }\n }\n logger.debug(`Contacts relation now has ${existingMatch.pageIds.length} entries`);\n }\n}\n\n/**\n * Add AI-extracted mentioned people to the Other People relation\n *\n * This function processes names that AI extracts from the todo text\n * (e.g., \"follow up with Sarah and Mike\" → [\"Sarah\", \"Mike\"])\n * and adds matching contacts to the Other People relation.\n *\n * Only works when otherPeople is configured as a relation type.\n * Uses stricter matching rules than owner:\n * - Exact match only (no fuzzy/word matching)\n * - Skip if ambiguous (multiple matches)\n * - Skip if already in Owner or @mentions\n *\n * @param mentionedPeople - Names extracted by AI from todo text\n * @param resolvedOwnerName - Owner name (to skip duplicates)\n * @param peopleMentions - @mentions from block text (to skip duplicates)\n * @param relationMatches - Existing relation matches to merge into\n * @param cache - Relation cache with contact names\n */\nfunction addAIMentionedPeopleToOtherPeople(\n mentionedPeople: string[],\n resolvedOwnerName: string | null,\n peopleMentions: Array<{id: string; name: string}>,\n relationMatches: RelationMatchResult[],\n cache: RelationCache\n): void {\n // Only process if otherPeople is a relation type\n if (cache.otherPeoplePropertyType !== 'relation' || !cache.otherPeopleTargetDbId) {\n return;\n }\n\n // Get the otherPeople property name\n if (!TODOS_PROPERTIES.otherPeople) return;\n\n const contactNames = cache.relationPageNames.get(cache.otherPeopleTargetDbId);\n if (!contactNames || contactNames.length === 0) return;\n\n const contactIdsToAdd: string[] = [];\n const ownerNameLower = resolvedOwnerName?.toLowerCase();\n\n for (const name of mentionedPeople) {\n const nameLower = name.toLowerCase();\n\n // Skip if matches Owner\n if (ownerNameLower && nameLower === ownerNameLower) {\n logger.debug(`Skipping AI name \"${name}\": matches owner`);\n continue;\n }\n\n // Skip if matches any @mention\n const matchesMention = peopleMentions.some(m =>\n m.name.toLowerCase() === nameLower ||\n m.name.toLowerCase().includes(nameLower) ||\n nameLower.includes(m.name.toLowerCase())\n );\n if (matchesMention) {\n logger.debug(`Skipping AI name \"${name}\": matches @mention`);\n continue;\n }\n\n // Try exact match first\n const exactMatches = contactNames.filter(c => c.nameLower === nameLower);\n\n if (exactMatches.length === 1) {\n const match = exactMatches[0];\n if (!contactIdsToAdd.includes(match.id)) {\n contactIdsToAdd.push(match.id);\n logger.debug(`Adding AI name \"${name}\" to Other People (exact): ${match.name}`);\n }\n } else if (exactMatches.length > 1) {\n logger.debug(`Skipping AI name \"${name}\": ambiguous (${exactMatches.length} exact matches)`);\n } else {\n // No exact match - try prefix matching\n // \"Kevin M\" should match \"Kevin McIntyre\" (all words are prefixes)\n // Single-letter initials like \"M\" are allowed\n const nameWords = nameLower.split(/\\s+/);\n const prefixMatches = contactNames.filter(contact => {\n const contactWords = contact.nameLower.split(/\\s+/);\n // Every word in the AI name must be a prefix of some word in the contact name\n return nameWords.every(nw =>\n contactWords.some(cw => cw.startsWith(nw))\n );\n });\n\n if (prefixMatches.length === 1) {\n const match = prefixMatches[0];\n if (!contactIdsToAdd.includes(match.id)) {\n contactIdsToAdd.push(match.id);\n logger.debug(`Adding AI name \"${name}\" to Other People (prefix): ${match.name}`);\n }\n } else if (prefixMatches.length > 1) {\n logger.debug(`Skipping AI name \"${name}\": ambiguous (${prefixMatches.length} prefix matches)`);\n } else {\n logger.debug(`Skipping AI name \"${name}\": no match found`);\n }\n }\n }\n\n // Merge into existing relation match (deduplicated)\n if (contactIdsToAdd.length > 0) {\n let existingMatch = relationMatches.find(r => r.propertyName === TODOS_PROPERTIES.otherPeople);\n if (!existingMatch) {\n existingMatch = { propertyName: TODOS_PROPERTIES.otherPeople!, pageIds: [], matchType: 'ai' };\n relationMatches.push(existingMatch);\n }\n for (const id of contactIdsToAdd) {\n if (!existingMatch.pageIds.includes(id)) {\n existingMatch.pageIds.push(id);\n }\n }\n logger.debug(`Other People relation now has ${existingMatch.pageIds.length} entries`);\n }\n}\n\n/**\n * Traverse parent pages to find contextual relation matches (Strategy 2c)\n *\n * For child pages (where source_page_database_id is null because parent.type is 'page_id'),\n * walks up the page hierarchy checking each ancestor for:\n * - 2a: Relation properties with values matching a target database\n * - 2b: Being a direct child of a target database\n *\n * Stops when a database entry is found or max depth is reached.\n *\n * @param startPageId - The source page ID to start traversing from\n * @param cache - Relation cache with target database info\n * @param alreadyMatched - Set of relation property names already matched by earlier strategies\n * @param onlyPropertyName - WHITELIST: Only match this property (typically Projects)\n * @returns Array of contextual relation match results\n */\nasync function traverseParentPagesForContextualMatch(\n startPageId: string,\n cache: RelationCache,\n alreadyMatched: Set<string>,\n onlyPropertyName: string // WHITELIST: Only match this property (Projects)\n): Promise<RelationMatchResult[]> {\n const results: RelationMatchResult[] = [];\n let currentPageId = startPageId;\n const maxDepth = 3;\n\n for (let depth = 0; depth < maxDepth; depth++) {\n const pageResult = await notionService.getNotionPage(currentPageId);\n if (!pageResult.success || !pageResult.data) break;\n\n const page = pageResult.data as any;\n const parent = page.parent;\n\n // Check 3a: Does this page have relation properties with matching values?\n if (page.properties) {\n for (const [_propName, propValue] of Object.entries(page.properties)) {\n const prop = propValue as any;\n if (prop.type === 'relation' && Array.isArray(prop.relation) && prop.relation.length > 0) {\n const pageIds = prop.relation.map((r: any) => r.id.replace(/-/g, ''));\n for (const relation of cache.relations) {\n if (alreadyMatched.has(relation.propertyName)) continue;\n if (relation.propertyName !== onlyPropertyName) continue; // WHITELIST\n const targetPageIds = cache.targetDbPageIds.get(relation.targetDatabaseId);\n if (!targetPageIds) continue;\n const matchingIds = pageIds.filter((id: string) => targetPageIds.has(id));\n if (matchingIds.length > 0) {\n results.push({\n propertyName: relation.propertyName,\n pageIds: matchingIds,\n matchType: 'contextual',\n });\n alreadyMatched.add(relation.propertyName);\n }\n }\n }\n }\n }\n\n // Check 3b: Is this page a direct child of a target database?\n if (parent?.type === 'database_id') {\n const dbId = parent.database_id.replace(/-/g, '');\n for (const relation of cache.relations) {\n if (alreadyMatched.has(relation.propertyName)) continue;\n if (relation.propertyName !== onlyPropertyName) continue; // WHITELIST\n if (dbId === relation.targetDatabaseId) {\n results.push({\n propertyName: relation.propertyName,\n pageIds: [currentPageId],\n matchType: 'contextual',\n });\n alreadyMatched.add(relation.propertyName);\n }\n }\n break; // Reached a database entry, stop traversing\n }\n\n // Move up to parent page\n if (parent?.type === 'page_id') {\n currentPageId = parent.page_id.replace(/-/g, '');\n } else {\n break; // No more page parents to traverse\n }\n }\n\n return results;\n}\n\n/**\n * Find all owner matches for a search name using flexible word matching\n *\n * Matching rules:\n * - Exact match: \"brad noble\" === \"brad noble\"\n * - Word match: any search word matches any entry word\n * - Prefix match: \"n\" matches \"noble\" via startsWith\n *\n * Examples that match \"Brad Noble\":\n * - \"Brad\" (word match)\n * - \"Brad N\" (word match + prefix)\n * - \"Brad Noble\" (exact or word match)\n * - \"Noble\" (word match)\n *\n * @param searchName - Name to search for (will be lowercased)\n * @param ownerNames - Array of page name entries to search\n * @returns All matching entries (may be empty)\n */\n// Common words to ignore when matching names\nconst STOPWORDS = new Set([\n 'the', 'a', 'an', 'and', 'or', 'but', 'in', 'on', 'at', 'to', 'for', 'of', 'with',\n 'by', 'from', 'as', 'is', 'was', 'are', 'were', 'been', 'be', 'have', 'has', 'had',\n 'do', 'does', 'did', 'will', 'would', 'could', 'should', 'may', 'might', 'must',\n 'shall', 'can', 'need', 'dare', 'ought', 'used', 'it', 'its', 'this', 'that',\n 'these', 'those', 'i', 'you', 'he', 'she', 'we', 'they', 'what', 'which', 'who',\n 'whom', 'whose', 'where', 'when', 'why', 'how', 'all', 'each', 'every', 'both',\n 'few', 'more', 'most', 'other', 'some', 'such', 'no', 'nor', 'not', 'only', 'own',\n 'same', 'so', 'than', 'too', 'very', 'just', 'also', 'now', 'here', 'there', 'then',\n 'once', 'see', 'about', 'into', 'through', 'during', 'before', 'after', 'above',\n 'below', 'up', 'down', 'out', 'off', 'over', 'under', 'again', 'further',\n 'database', 'page', 'block', 'todo', 'task', 'project', 'll', 'you\\'ll', 've'\n]);\n\nfunction findOwnerMatches(searchName: string, ownerNames: PageNameEntry[]): PageNameEntry[] {\n const searchLower = searchName.toLowerCase().trim();\n if (!searchLower) return [];\n\n // Filter out stopwords only - keep short names like \"Al\" or \"Jo\"\n const searchWords = searchLower.split(/\\s+/).filter(w => !STOPWORDS.has(w));\n\n // If no meaningful words remain, no match\n if (searchWords.length === 0) return [];\n\n const matches: PageNameEntry[] = [];\n\n for (const entry of ownerNames) {\n // Exact match (highest confidence)\n if (entry.nameLower === searchLower) {\n matches.push(entry);\n continue;\n }\n\n // Word match: search word matches entry word exactly or entry word starts with search word\n // e.g., \"brad\" matches \"brad\" or \"bradley\", but NOT \"b\" matching \"brad\"\n const entryWords = entry.nameLower.split(/\\s+/);\n const hasMatch = searchWords.some(sw =>\n entryWords.some(ew => ew === sw || ew.startsWith(sw))\n );\n if (hasMatch) {\n matches.push(entry);\n }\n }\n\n return matches;\n}\n\n/**\n * Resolve owner matches to a single result, using AI disambiguation if needed\n *\n * @param matches - Array of matching entries\n * @param todoText - Todo text for AI disambiguation context\n * @returns Single resolved owner or null\n */\nasync function resolveOwnerMatches(\n matches: PageNameEntry[],\n todoText: string\n): Promise<{ id: string; name: string } | null> {\n if (matches.length === 0) {\n return null;\n }\n\n if (matches.length === 1) {\n return { id: matches[0].id, name: matches[0].name };\n }\n\n // Multiple matches - use AI disambiguation\n logger.debug(`resolveOwnerMatches: ${matches.length} matches, using AI disambiguation`);\n const disambiguatedId = await aiService.disambiguateOwnerMatch(\n todoText,\n matches.map(m => ({ id: m.id, name: m.name }))\n );\n\n if (disambiguatedId) {\n const match = matches.find(m => m.id === disambiguatedId);\n if (match) {\n return { id: match.id, name: match.name };\n }\n }\n\n // Fall back to first match if disambiguation fails\n logger.debug(`resolveOwnerMatches: disambiguation failed, using first match: ${matches[0].name}`);\n return { id: matches[0].id, name: matches[0].name };\n}\n\n/**\n * Try to resolve owner from heading text\n *\n * Uses flexible word matching and AI disambiguation when multiple contacts match.\n *\n * @param heading - The preceding heading text (e.g., \"### Alex\")\n * @param todoText - The todo text (for AI disambiguation context)\n * @param cache - Relation cache with owner property info\n * @returns Resolved owner or null if no match\n */\nasync function tryResolveOwnerFromHeading(\n heading: string | null,\n todoText: string,\n cache: RelationCache | null\n): Promise<{ resolved: ResolvedOwner; ownerName: string } | null> {\n if (!heading || !cache || !cache.ownerPropertyType) {\n return null;\n }\n\n // Clean heading (remove markdown heading markers)\n const cleanHeading = heading.replace(/^#+\\s*/, '').trim();\n if (!cleanHeading) {\n return null;\n }\n\n logger.debug(`tryResolveOwnerFromHeading: checking \"${cleanHeading}\" against ${cache.ownerPropertyType} type`);\n\n if (cache.ownerPropertyType === 'people') {\n // Check workspace users\n const userResult = await findUserByName(cleanHeading);\n if (userResult.success && userResult.user) {\n logger.debug(`tryResolveOwnerFromHeading: matched to user \"${userResult.user.name}\"`);\n return {\n resolved: { type: 'people', value: [{ id: userResult.user.id }] },\n ownerName: userResult.user.name\n };\n }\n } else if (cache.ownerPropertyType === 'relation' && cache.ownerTargetDbId) {\n const ownerNames = cache.relationPageNames.get(cache.ownerTargetDbId);\n if (!ownerNames || ownerNames.length === 0) {\n logger.debug(`tryResolveOwnerFromHeading: no owner names loaded`);\n return null;\n }\n\n // Find all matches using shared logic\n const matches = findOwnerMatches(cleanHeading, ownerNames);\n logger.debug(`tryResolveOwnerFromHeading: found ${matches.length} matches for \"${cleanHeading}\"`);\n\n // Resolve to single match (with AI disambiguation if needed)\n const resolved = await resolveOwnerMatches(matches, todoText);\n if (resolved) {\n logger.debug(`tryResolveOwnerFromHeading: resolved to \"${resolved.name}\"`);\n return {\n resolved: { type: 'relation', value: [{ id: resolved.id }] },\n ownerName: resolved.name\n };\n }\n }\n\n logger.debug(`tryResolveOwnerFromHeading: no match found for \"${cleanHeading}\"`);\n return null;\n}\n\n/**\n * Validate AI-extracted owner name against owner database\n *\n * Called when AI extracts an owner name from todo text (e.g., \"Brad to contact Kevin\").\n * Uses the same matching logic as heading resolution for consistency.\n *\n * @param ownerName - Name extracted by AI\n * @param todoText - Full todo text for AI disambiguation context\n * @param cache - Relation cache with owner property info\n * @returns Resolved owner or null if name not found in owner database\n */\nasync function validateAndResolveOwner(\n ownerName: string,\n todoText: string,\n cache: RelationCache | null\n): Promise<ResolvedOwner | null> {\n if (!ownerName || !cache || !cache.ownerPropertyType) {\n return null;\n }\n\n logger.debug(`validateAndResolveOwner: checking \"${ownerName}\" against ${cache.ownerPropertyType} type`);\n\n if (cache.ownerPropertyType === 'people') {\n // For people property type, try to find in workspace users\n const userResult = await findUserByName(ownerName);\n if (userResult.success && userResult.user) {\n logger.debug(`validateAndResolveOwner: matched to user \"${userResult.user.name}\"`);\n return {\n type: 'people',\n value: [{ id: userResult.user.id }]\n };\n }\n logger.debug(`validateAndResolveOwner: \"${ownerName}\" not found in workspace users`);\n return null;\n }\n\n if (cache.ownerPropertyType === 'relation' && cache.ownerTargetDbId) {\n const ownerNames = cache.relationPageNames.get(cache.ownerTargetDbId);\n if (!ownerNames || ownerNames.length === 0) {\n logger.debug('validateAndResolveOwner: no owner names loaded');\n return null;\n }\n\n // Use shared matching logic (same as heading resolution)\n const matches = findOwnerMatches(ownerName, ownerNames);\n logger.debug(`validateAndResolveOwner: found ${matches.length} matches for \"${ownerName}\"`);\n\n // Resolve to single match (with AI disambiguation if needed)\n const resolved = await resolveOwnerMatches(matches, todoText);\n if (resolved) {\n logger.debug(`validateAndResolveOwner: resolved to \"${resolved.name}\"`);\n return {\n type: 'relation',\n value: [{ id: resolved.id }]\n };\n }\n\n logger.debug(`validateAndResolveOwner: \"${ownerName}\" not found in owner database`);\n return null;\n }\n\n return null;\n}\n\n/**\n * Process todo save: Read blobs and sync to Notion database\n * Creates new pages or updates existing pages based on timestamp comparison\n *\n * Status handling:\n * - CREATE: Sets Status to \"Not started\" for new pages\n * - UPDATE: Does NOT modify Status (preserves user changes)\n */\nexport async function processTodoSave(): Promise<TodoSaveResult> {\n // Get database ID from environment\n const databaseId = Deno.env.get('TODOS_DB_ID');\n\n if (!databaseId) {\n return {\n success: false,\n totalBlobs: 0,\n pagesCreated: 0,\n pagesUpdated: 0,\n pagesSkipped: 0,\n pagesFailed: 0,\n failures: [],\n error: \"Missing TODOS_DB_ID environment variable\",\n details: \"Set TODOS_DB_ID to the Notion database ID where todos should be saved\"\n };\n }\n\n\n // Initialize relation cache - auto-discovers all relations from todos database schema\n // Also validates that required properties exist in the database\n const relationCache = await initializeRelationCache(databaseId);\n\n // Fail fast if property validation failed (required properties missing)\n if (!relationCache) {\n return {\n success: false,\n totalBlobs: 0,\n pagesCreated: 0,\n pagesUpdated: 0,\n pagesSkipped: 0,\n pagesFailed: 0,\n failures: [],\n error: \"Property validation failed\",\n details: \"Required properties not found in database. Run health check for details.\"\n };\n }\n\n // Get bot user ID for skip-update check\n // If a user (not this integration) last edited a todo page, skip updating to preserve their changes\n const botUserResult = await notionService.getBotUser();\n const botUserId = botUserResult.success ? botUserResult.data?.id : null;\n if (!botUserId) {\n logger.warn(`Could not get bot user ID: ${botUserResult.error}. Will update all pages.`);\n }\n\n // List all todo blobs\n const blobsResult = await blobService.listBlocksByCategory('todo');\n\n if (!blobsResult.success || !blobsResult.keys) {\n return {\n success: false,\n totalBlobs: 0,\n pagesCreated: 0,\n pagesUpdated: 0,\n pagesSkipped: 0,\n pagesFailed: 0,\n failures: [],\n error: \"Failed to list blobs\",\n details: blobsResult.error || undefined\n };\n }\n\n const blobKeys = blobsResult.keys;\n\n let pagesCreated = 0;\n let pagesUpdated = 0;\n let pagesSkipped = 0;\n let pagesFailed = 0;\n const failures: Array<{ blockId: string; error: string }> = [];\n\n // Process each blob\n for (const key of blobKeys) {\n const blockId = extractBlockIdFromKey(key);\n const projectName = key.split('--')[0];\n const fullKey = `${projectName}--todo--${blockId}`;\n let lockAcquired = false;\n\n try {\n // Get blob content\n const blobContentResult = await blobService.getBlockFromBlob('todo', blockId);\n\n if (!blobContentResult.success || !blobContentResult.data) {\n throw new Error(`Failed to get blob content: ${blobContentResult.error}`);\n }\n\n const reducedBlock: ReducedNotionBlock = blobContentResult.data;\n\n // OPTIMIZATION: Skip if already synced\n if (reducedBlock.sync_metadata?.synced === true) {\n pagesSkipped++;\n continue;\n }\n\n // RACE CONDITION GUARD: Skip if sync already in progress by another process\n // This prevents duplicate page creation when webhooks fire in rapid succession\n if (reducedBlock.sync_metadata?.syncing === true) {\n logger.debug(`Skipping block ${blockId} - sync already in progress`);\n pagesSkipped++;\n continue;\n }\n\n // ACQUIRE SYNC LOCK: Prevent concurrent syncs of same block\n // Uses optimistic locking - marks blob as \"syncing\" so other processes skip it\n const lockResult = await blobService.acquireSyncLock(fullKey);\n\n if (!lockResult.success) {\n // Lock already held or failed - skip this block (will be picked up next run)\n logger.debug(`Could not acquire sync lock for ${blockId}: ${lockResult.error}`);\n pagesSkipped++;\n continue;\n }\n\n lockAcquired = true;\n\n // NOTE: No validation needed here - all blobs are guaranteed valid\n // Validation happens during search phase (todoController.ts)\n // Only blocks with people_mentions AND date_mentions are saved to blob storage\n\n // Dynamic relation matching - handles all relations\n // Strategies: 1) Explicit mentions, 2) Contextual\n const relationMatches = await matchBlockToRelations(reducedBlock, relationCache);\n\n // Generate AI-interpreted v2 todo\n // Gather project names from relation matches for AI context\n const relatedProjectNames: string[] = [];\n\n if (relationCache && relationCache.relationPageNames.size > 0) {\n for (const match of relationMatches) {\n // Find the relation and its target database\n const relation = relationCache.relations.find(r => r.propertyName === match.propertyName);\n if (!relation) continue;\n\n // Skip Owner relation - owner resolution is handled separately\n if (relation.propertyName === TODOS_PROPERTIES.owner) continue;\n\n const names = relationCache.relationPageNames.get(relation.targetDatabaseId);\n if (!names) continue;\n\n for (const pageId of match.pageIds) {\n const page = names.find(n => n.id === pageId);\n if (page && !relatedProjectNames.includes(page.name)) {\n relatedProjectNames.push(page.name);\n }\n }\n }\n }\n\n // =======================================================================\n // OWNER RESOLUTION - Simplified Flow\n // =======================================================================\n // Priority: 1. Heading match → 2. @mention → 3. AI extraction (validated against owner DB)\n // No \"contact\" source - owner comes from explicit signals only.\n // =======================================================================\n\n logger.debug(`Owner resolution for block: heading=\"${reducedBlock.preceding_heading || 'none'}\", mentions=${reducedBlock.people_mentions?.length || 0}, projects=[${relatedProjectNames.join(', ')}]`);\n\n // Step 1: Try to resolve owner directly from heading\n const headingOwnerResult = await tryResolveOwnerFromHeading(\n reducedBlock.preceding_heading || null,\n reducedBlock.todo_string,\n relationCache\n );\n\n // Format dates for AI (human-readable instead of raw ISO)\n const formattedDates = (reducedBlock.date_mentions || []).map(formatDateString);\n\n // Calculate default due date BEFORE AI call (for AI to use if no date in content)\n const dueDateSetting = getDefaultDueDateSetting();\n const referenceDate = reducedBlock.created_time\n ? new Date(reducedBlock.created_time)\n : new Date();\n const defaultDueDate = calculateDueDate(dueDateSetting, referenceDate);\n const formattedDefaultDate = formatDateString(defaultDueDate);\n\n // Always call AI for summary generation (owner extraction happens here too)\n const summaryResult = await aiService.generateSummary({\n todoString: reducedBlock.todo_string,\n precedingHeading: reducedBlock.preceding_heading || null,\n peopleMentions: reducedBlock.people_mentions || [],\n dateMentions: formattedDates,\n matchedProject: relatedProjectNames[0] || null,\n sourcePage: null,\n defaultDueDate: formattedDefaultDate\n });\n\n // AI should always return a dueDate (from content or default)\n // Fallback to calculated default only if AI fails\n const finalDueDate = summaryResult.dueDate || defaultDueDate;\n logger.debug(`Due date: ${finalDueDate}${summaryResult.dueDate ? ' (from AI)' : ' (fallback)'}`);\n\n // Update date_mentions with final due date for property mapping\n reducedBlock.date_mentions = [finalDueDate];\n\n // Step 2: Resolve owner using priority order\n let resolvedOwner: ResolvedOwner | null = null;\n let resolvedOwnerName: string | null = null;\n\n if (headingOwnerResult) {\n // Heading matched an owner in the database - use it (highest priority, deterministic)\n resolvedOwner = headingOwnerResult.resolved;\n resolvedOwnerName = headingOwnerResult.ownerName;\n logger.debug(`Owner resolved from heading: \"${headingOwnerResult.ownerName}\"`);\n } else if (reducedBlock.people_mentions && reducedBlock.people_mentions.length > 0) {\n // Has @mentions - use first one (explicit, has user ID)\n const firstMention = reducedBlock.people_mentions[0];\n if (firstMention.id) {\n if (relationCache?.ownerPropertyType === 'people') {\n resolvedOwner = { type: 'people', value: [{ id: firstMention.id }] };\n resolvedOwnerName = firstMention.name;\n logger.debug(`Owner resolved from @mention: \"${firstMention.name}\"`);\n } else if (relationCache?.ownerPropertyType === 'relation' && firstMention.name) {\n // @mentions are explicit - try exact match first (no disambiguation needed)\n const ownerNames = relationCache.relationPageNames.get(relationCache.ownerTargetDbId || '');\n if (ownerNames) {\n const mentionNameLower = firstMention.name.toLowerCase();\n // Exact match takes priority - @Brad Noble should match \"Brad Noble\" not \"Brad Noble (personal)\"\n const exactMatch = ownerNames.find(n => n.nameLower === mentionNameLower);\n if (exactMatch) {\n resolvedOwner = { type: 'relation', value: [{ id: exactMatch.id }] };\n resolvedOwnerName = exactMatch.name;\n logger.debug(`Owner resolved from @mention (exact match): \"${exactMatch.name}\"`);\n } else {\n // No exact match - fall back to fuzzy matching with disambiguation\n const matches = findOwnerMatches(firstMention.name, ownerNames);\n const resolved = await resolveOwnerMatches(matches, reducedBlock.todo_string);\n if (resolved) {\n resolvedOwner = { type: 'relation', value: [{ id: resolved.id }] };\n resolvedOwnerName = resolved.name;\n logger.debug(`Owner resolved from @mention (fuzzy match): \"${resolved.name}\"`);\n }\n }\n }\n }\n }\n } else if (summaryResult.ownerName) {\n // AI extracted an owner name - validate against owner database\n resolvedOwner = await validateAndResolveOwner(\n summaryResult.ownerName,\n reducedBlock.todo_string,\n relationCache\n );\n if (resolvedOwner) {\n resolvedOwnerName = summaryResult.ownerName;\n logger.debug(`Owner resolved from AI extraction: \"${summaryResult.ownerName}\"`);\n } else {\n logger.debug(`AI suggested owner \"${summaryResult.ownerName}\" but not found in owner database`);\n }\n }\n\n // Log final owner result\n if (resolvedOwner) {\n logger.debug(`Final owner: type=${resolvedOwner.type}, id=${resolvedOwner.value[0]?.id}`);\n } else {\n logger.debug(`Final owner: null`);\n }\n\n // Add owner and @mentions to Contacts relation\n // This preserves connections even if Owner is later changed\n await addPeopleToContactsRelation(\n resolvedOwner,\n reducedBlock.people_mentions || [],\n relationMatches,\n relationCache\n );\n\n // Add AI-extracted mentioned people to Other People relation\n // Only works when otherPeople is configured as a relation type\n logger.debug(`AI mentionedPeople: [${summaryResult.mentionedPeople?.join(', ') || 'none'}]`);\n if (summaryResult.mentionedPeople && summaryResult.mentionedPeople.length > 0) {\n addAIMentionedPeopleToOtherPeople(\n summaryResult.mentionedPeople,\n resolvedOwnerName,\n reducedBlock.people_mentions || [],\n relationMatches,\n relationCache\n );\n }\n\n // Map to Notion properties with dynamic relation matches, resolved owner, and status info\n // Uses validated property names from cache to ensure exact column name match\n const properties = mapReducedBlockToProperties(\n reducedBlock,\n relationCache.validatedProperties,\n relationMatches,\n summaryResult.summary,\n resolvedOwner,\n relationCache.statusInfo\n );\n\n // Extract emoji from todo_string for page icon\n const emoji = extractFirstEmoji(reducedBlock.todo_string);\n const icon = emoji ? { type: \"emoji\", emoji: emoji } : undefined;\n\n // OPTIMIZATION: Use cached page ID if available (fallback to old field name for existing blobs)\n const cachedPageId = reducedBlock.sync_metadata?.target_page_id || reducedBlock.sync_metadata?.notion_page_id;\n let pageId: string | null = null;\n\n if (cachedPageId) {\n // We have a cached page ID - check if user edited before updating\n // Fetch page to check last_edited_by\n if (botUserId) {\n const pageCheckResult = await notionService.getNotionPage(cachedPageId);\n if (pageCheckResult.success && pageCheckResult.data) {\n const lastEditedBy = (pageCheckResult.data as any).last_edited_by?.id;\n if (lastEditedBy && lastEditedBy !== botUserId) {\n // User edited this page - skip update to preserve their changes\n logger.info(`Skipping update for block ${blockId}: page was edited by user (not bot)`);\n await blobService.updateBlobSyncMetadata(fullKey, { synced: true, syncing: false });\n pagesSkipped++;\n continue;\n }\n }\n }\n\n const updateResult = await notionService.updateNotionPage(cachedPageId, properties, icon);\n\n if (updateResult.success) {\n pagesUpdated++;\n pageId = cachedPageId;\n } else if (updateResult.error?.includes('archived')) {\n // Page was archived - clear cache and skip (user intentionally archived it)\n await blobService.updateBlobSyncMetadata(fullKey, {\n synced: true, // Mark synced to prevent retry\n syncing: false, // Release lock\n target_page_id: undefined, // Clear cached ID\n archived: true // Flag as archived\n });\n pagesSkipped++;\n continue;\n } else {\n throw new Error(`Failed to update page: ${enhanceNotionError(updateResult.error || 'Unknown error')}`);\n }\n } else {\n // No cached page ID - query database to check if page exists\n // Use validated property name from cache for consistency with how pages are created\n const blockIdPropertyName = relationCache.validatedProperties['sourceBlockId'] || undefined;\n const existingPageResult = await notionService.queryDatabaseByBlockId(databaseId, blockId, blockIdPropertyName);\n\n if (!existingPageResult.success) {\n throw new Error(`Failed to query database: ${existingPageResult.error}`);\n }\n\n if (existingPageResult.data) {\n // Page exists but we don't have cached ID - check if user edited before updating\n pageId = existingPageResult.data.id;\n\n // Check if user (not bot) last edited this page\n if (botUserId) {\n const lastEditedBy = (existingPageResult.data as any).last_edited_by?.id;\n if (lastEditedBy && lastEditedBy !== botUserId) {\n // User edited this page - skip update to preserve their changes\n // Cache the page ID for future reference\n logger.info(`Skipping update for block ${blockId}: page was edited by user (not bot)`);\n await blobService.updateBlobSyncMetadata(fullKey, {\n synced: true,\n syncing: false,\n target_page_id: pageId // Cache for future\n });\n pagesSkipped++;\n continue;\n }\n }\n\n const updateResult = await notionService.updateNotionPage(pageId, properties, icon);\n\n if (updateResult.success) {\n pagesUpdated++;\n } else {\n throw new Error(`Failed to update existing page: ${enhanceNotionError(updateResult.error || 'Unknown error')}`);\n }\n } else {\n // Page doesn't exist - create new\n const createResult = await notionService.createNotionPage(databaseId, properties, icon);\n\n if (createResult.success) {\n pagesCreated++;\n pageId = createResult.data?.id || null;\n\n // Add callout if source block has children\n if (pageId && reducedBlock.has_children && reducedBlock.block_url) {\n const calloutResult = await notionService.appendHasChildrenCallout(pageId, reducedBlock.block_url);\n if (!calloutResult.success) {\n logger.warn(`Failed to add children callout: ${calloutResult.error}`);\n }\n }\n\n // Set block background color on first sync\n // (Only on create, not update, to avoid infinite sync loop)\n const bgColor = getBlockBackgroundColor();\n if (bgColor) {\n const colorResult = await notionService.updateBlockColor(blockId, bgColor);\n if (!colorResult.success) {\n logger.warn(`Failed to set block color: ${colorResult.error}`);\n }\n }\n } else {\n throw new Error(`Failed to create page: ${enhanceNotionError(createResult.error || 'Unknown error')}`);\n }\n }\n }\n\n // Mark blob as synced on success and release lock\n if (pageId) {\n const updateMetadataResult = await blobService.updateBlobSyncMetadata(fullKey, {\n synced: true,\n syncing: false, // Release lock\n target_page_id: pageId\n });\n\n if (!updateMetadataResult.success) {\n logger.warn(`Failed to update sync metadata: ${updateMetadataResult.error}`);\n }\n } else {\n // No page created but lock was acquired - release it\n await blobService.releaseSyncLock(fullKey);\n }\n } catch (err: any) {\n // Release lock on failure so block can be retried on next run\n // Only release if we actually acquired the lock (error may have occurred before acquisition)\n if (lockAcquired) {\n await blobService.releaseSyncLock(fullKey);\n }\n\n pagesFailed++;\n failures.push({\n blockId: blockId,\n error: err.message\n });\n }\n }\n\n return {\n success: true,\n totalBlobs: blobKeys.length,\n pagesCreated,\n pagesUpdated,\n pagesSkipped,\n pagesFailed,\n failures\n };\n}\n",[],[412,713,704],"controllers","8ce47620-e999-11f0-9687-42dde27851f2","todoOrchestrationController.ts","fd176688-b4f0-11f0-8e48-0224a6c84d84",222,["D",1767550615134],"import * as pageController from './pageController.ts';\nimport * as todoController from './todoController.ts';\nimport * as todoSaveController from './todoSaveController.ts';\nimport { getRecentPagesLookbackHours } from '../utils/configUtils.ts';\n\n/**\n * Batch todo search result for a single page\n */\ninterface PageSearchResult {\n pageId: string;\n pageTitle: string;\n success: boolean;\n blocksFound: number;\n blocksNew?: number;\n blocksUpdated?: number;\n blocksSkipped?: number;\n blockIds: string[];\n error?: string;\n}\n\n/**\n * Batch search result (search and save to blobs only)\n */\nexport interface BatchSearchResult {\n success: boolean;\n pagesSearched: number;\n totalTodosFound: number;\n totalNew?: number;\n totalUpdated?: number;\n totalSkipped?: number;\n searchResults: PageSearchResult[];\n error?: string;\n details?: string;\n}\n\n/**\n * Batch todos processing result (search + sync to database)\n */\nexport interface BatchTodosResult {\n success: boolean;\n pagesSearched: number;\n totalTodosFound: number;\n searchResults: PageSearchResult[];\n saveResult: any;\n error?: string;\n details?: string;\n}\n\n/**\n * Orchestrate batch keyword search:\n * 1. Get recent pages from Notion\n * 2. Search each page for keywords and save to blobs\n *\n * Does NOT sync to Notion database - that's handled separately by processTodoSave()\n *\n * Keywords are read from SEARCH_KEYWORDS environment variable (comma-separated)\n * Defaults to \"todo\" if not set\n *\n * @param hoursAgo - Number of hours to look back for recent pages (default: RECENT_PAGES_LOOKBACK_HOURS env var, fallback 24)\n * @param applyStabilityDelay - Whether to apply stability check before saving blocks (default: true for cron usage)\n */\nexport async function processBatchSearch(hoursAgo?: number, applyStabilityDelay: boolean = true): Promise<BatchSearchResult> {\n const hours = hoursAgo ?? getRecentPagesLookbackHours();\n\n // Step 1: Get recent pages\n const pagesResult = await pageController.getRecentPages(hours);\n\n if (!pagesResult.success || !pagesResult.data?.pages) {\n return {\n success: false,\n pagesSearched: 0,\n totalTodosFound: 0,\n searchResults: [],\n error: \"Failed to fetch recent pages\",\n details: pagesResult.error || undefined\n };\n }\n\n const pages = pagesResult.data.pages;\n\n if (pages.length === 0) {\n return {\n success: true,\n pagesSearched: 0,\n totalTodosFound: 0,\n searchResults: []\n };\n }\n\n // Step 2: Search each page for keywords\n const searchResults: PageSearchResult[] = [];\n let totalTodosFound = 0;\n let totalNew = 0;\n let totalUpdated = 0;\n let totalSkipped = 0;\n\n for (let i = 0; i < pages.length; i++) {\n const page = pages[i];\n\n try {\n const searchResult = await todoController.processTodoSearch(page.id, applyStabilityDelay);\n\n searchResults.push({\n pageId: page.id,\n pageTitle: page.title,\n success: searchResult.success,\n blocksFound: searchResult.blocksFound,\n blocksNew: searchResult.blocksNew,\n blocksUpdated: searchResult.blocksUpdated,\n blocksSkipped: searchResult.blocksSkipped,\n blockIds: searchResult.blockIds,\n error: searchResult.error\n });\n\n if (searchResult.success) {\n totalTodosFound += searchResult.blocksFound;\n totalNew += searchResult.blocksNew || 0;\n totalUpdated += searchResult.blocksUpdated || 0;\n totalSkipped += searchResult.blocksSkipped || 0;\n }\n } catch (err: any) {\n searchResults.push({\n pageId: page.id,\n pageTitle: page.title,\n success: false,\n blocksFound: 0,\n blockIds: [],\n error: err.message\n });\n }\n }\n\n // Return comprehensive summary (no database sync)\n return {\n success: true,\n pagesSearched: pages.length,\n totalTodosFound,\n totalNew,\n totalUpdated,\n totalSkipped,\n searchResults\n };\n}\n\n/**\n * Orchestrate batch keyword search processing:\n * 1. Get recent pages from Notion\n * 2. Search each page for keywords and save to blobs\n * 3. Sync all todo blobs to Notion database\n *\n * Keywords are read from SEARCH_KEYWORDS environment variable (comma-separated)\n * Defaults to \"todo\" if not set\n *\n * @param hoursAgo - Number of hours to look back for recent pages (default: RECENT_PAGES_LOOKBACK_HOURS env var, fallback 24)\n * @param applyStabilityDelay - Whether to apply stability check before saving blocks (default: true, respects BLOCK_STABILITY_MINUTES)\n */\nexport async function processBatchTodos(hoursAgo?: number, applyStabilityDelay: boolean = true): Promise<BatchTodosResult> {\n const hours = hoursAgo ?? getRecentPagesLookbackHours();\n\n // Step 1: Get recent pages\n const pagesResult = await pageController.getRecentPages(hours);\n\n if (!pagesResult.success || !pagesResult.data?.pages) {\n return {\n success: false,\n pagesSearched: 0,\n totalTodosFound: 0,\n searchResults: [],\n saveResult: null,\n error: \"Failed to fetch recent pages\",\n details: pagesResult.error || undefined\n };\n }\n\n const pages = pagesResult.data.pages;\n\n if (pages.length === 0) {\n return {\n success: true,\n pagesSearched: 0,\n totalTodosFound: 0,\n searchResults: [],\n saveResult: { totalBlobs: 0, pagesCreated: 0, pagesUpdated: 0, pagesSkipped: 0, pagesFailed: 0 }\n };\n }\n\n // Step 2: Search each page for keywords\n const searchResults: PageSearchResult[] = [];\n let totalTodosFound = 0;\n let totalNew = 0;\n let totalUpdated = 0;\n let totalSkipped = 0;\n\n for (let i = 0; i < pages.length; i++) {\n const page = pages[i];\n\n try {\n const searchResult = await todoController.processTodoSearch(page.id, applyStabilityDelay);\n\n searchResults.push({\n pageId: page.id,\n pageTitle: page.title,\n success: searchResult.success,\n blocksFound: searchResult.blocksFound,\n blocksNew: searchResult.blocksNew,\n blocksUpdated: searchResult.blocksUpdated,\n blocksSkipped: searchResult.blocksSkipped,\n blockIds: searchResult.blockIds,\n error: searchResult.error\n });\n\n if (searchResult.success) {\n totalTodosFound += searchResult.blocksFound;\n totalNew += searchResult.blocksNew || 0;\n totalUpdated += searchResult.blocksUpdated || 0;\n totalSkipped += searchResult.blocksSkipped || 0;\n }\n } catch (err: any) {\n searchResults.push({\n pageId: page.id,\n pageTitle: page.title,\n success: false,\n blocksFound: 0,\n blockIds: [],\n error: err.message\n });\n }\n }\n\n // Step 3: Save all todos to Notion database\n let saveResult;\n\n try {\n saveResult = await todoSaveController.processTodoSave();\n } catch (err: any) {\n saveResult = {\n success: false,\n totalBlobs: 0,\n pagesCreated: 0,\n pagesUpdated: 0,\n pagesSkipped: 0,\n pagesFailed: 0,\n failures: [],\n error: err.message\n };\n }\n\n // Return comprehensive summary\n return {\n success: true,\n pagesSearched: pages.length,\n totalTodosFound,\n searchResults,\n saveResult\n };\n}\n",[],[412,713,715],"1e093f7c-03cc-11f1-a1cf-42dde27851f2","todoController.ts","15827462-b459-11f0-a09c-0224a6c84d84",["D",1770431063716],51,"import * as notionService from '../services/notion/index.ts';\nimport { TODO_CONTAINER_TYPES } from '../services/notion/pages.ts';\nimport { batchGetPageParentDatabaseIds, getPageParentDatabaseId } from '../services/notion/relations.ts';\nimport type { TodoSearchResult } from '../../shared/types.ts';\nimport { reduceNotionBlock, extractPageIdFromUrl } from '../utils/notionUtils.ts';\nimport * as blobService from '../services/blobService.ts';\nimport { logger } from '../utils/logger.ts';\nimport { getMinBlockWords, countWords } from '../utils/textUtils.ts';\nimport { getBlockStabilityMinutes, isBlockStable, getBlockAgeMinutes } from '../utils/blockUtils.ts';\n\n/**\n * Get search keywords from environment variable\n *\n * Reads SEARCH_KEYWORDS env var (comma-separated list)\n * Returns empty array if not set (keyword search disabled)\n *\n * @returns Array of keywords to search for (empty if not configured)\n */\nfunction getSearchKeywords(): string[] {\n const envKeywords = Deno.env.get('SEARCH_KEYWORDS');\n\n if (envKeywords) {\n return envKeywords\n .split(',')\n .map(k => k.trim())\n .filter(k => k.length > 0);\n }\n\n return []; // No default - keyword search only if explicitly configured\n}\n\n/**\n * Search configuration interface\n *\n * Block type search is always active (primary).\n * Keyword search is additive (only if SEARCH_KEYWORDS is set).\n */\ninterface SearchConfig {\n blockType: string; // Always set (defaults to 'to_do')\n keywords: string[]; // Empty array if not configured\n}\n\n/**\n * Get search configuration from environment variables\n *\n * - SEARCH_BLOCK_TYPE: Block type to search for (defaults to 'to_do')\n * - SEARCH_KEYWORDS: Optional comma-separated keywords for additive matching\n *\n * Block type search is always active. Keyword search is additive.\n *\n * @returns Search configuration\n */\nfunction getSearchConfig(): SearchConfig {\n // Block type is always active (defaults to 'to_do')\n const blockTypeEnv = Deno.env.get('SEARCH_BLOCK_TYPE');\n const blockType = blockTypeEnv !== undefined\n ? (blockTypeEnv.trim() || 'to_do')\n : 'to_do';\n\n // Keywords are additive (only if explicitly set)\n const keywords = getSearchKeywords();\n\n return { blockType, keywords };\n}\n\n/**\n * Extract creator information from Notion block\n * Returns user object in people_mentions format\n *\n * @param block - Notion block object\n * @returns Creator user object or null\n */\nfunction getBlockCreator(block: any): { id: string; name: string | null; email: string | null } | null {\n if (!block.created_by || !block.created_by.id) {\n return null;\n }\n\n return {\n id: block.created_by.id,\n name: null, // Not available without additional API call\n email: null\n };\n}\n\n/**\n * Search for blocks in a page\n * Logs entire block to console when found\n *\n * Two search modes:\n * 1. Keyword mode: Uses SEARCH_KEYWORDS env var (comma-separated)\n * 2. Block type mode: Uses SEARCH_BLOCK_TYPE env var (e.g., \"to_do\")\n *\n * Block type mode takes precedence if both are set.\n *\n * Validation:\n * - Blocks must have MIN_BLOCK_WORDS words (default: 5) to be saved\n * - Short blocks lack context needed for meaningful todos\n *\n * Auto-assignment:\n * - If no date_mentions: auto-assigns due date based on DEFAULT_DUE_DATE env var (defaults to \"today\")\n * - If no people_mentions: auto-assigns creator as owner\n *\n * Stability check:\n * - If applyStabilityDelay is true: blocks must be older than BLOCK_STABILITY_MINUTES (default: 2)\n * - If applyStabilityDelay is false: blocks are saved immediately (webhook/manual triggers)\n *\n * @param pageId - Notion page ID to search\n * @param applyStabilityDelay - Whether to check block age before saving (default: false for webhook compatibility)\n */\nexport async function processTodoSearch(pageId: string, applyStabilityDelay: boolean = false): Promise<TodoSearchResult> {\n // Validate input\n if (!pageId || typeof pageId !== 'string') {\n return {\n success: false,\n blocksFound: 0,\n blockIds: [],\n error: \"Invalid pageId\",\n details: \"pageId must be a non-empty string\"\n };\n }\n\n // Get search configuration\n const searchConfig = getSearchConfig();\n\n try {\n // Fetch page to get URL (single API call for all blocks in this page)\n const pageResult = await notionService.getNotionPage(pageId);\n const pageUrl = pageResult.success ? (pageResult.data as any)?.url : null;\n\n // Get source page's parent database ID for contextual mapping\n let sourcePageDatabaseId: string | null = null;\n // Extract relation property values from source page for contextual matching\n // Format: { propertyName -> [pageId, ...] } — only relation-type properties with values\n let sourcePageRelations: Record<string, string[]> | null = null;\n if (pageResult.success && pageResult.data) {\n const parent = (pageResult.data as any).parent;\n if (parent?.type === 'database_id') {\n sourcePageDatabaseId = parent.database_id.replace(/-/g, '');\n }\n\n const props = (pageResult.data as any).properties;\n if (props) {\n const relations: Record<string, string[]> = {};\n for (const [propName, propValue] of Object.entries(props)) {\n const prop = propValue as any;\n if (prop.type === 'relation' && Array.isArray(prop.relation) && prop.relation.length > 0) {\n relations[propName] = prop.relation.map((r: any) => r.id.replace(/-/g, ''));\n }\n }\n if (Object.keys(relations).length > 0) {\n sourcePageRelations = relations;\n }\n }\n }\n\n // Get all blocks recursively from the page\n // Optimize recursion when only searching for block types (no keywords)\n const containerFilter = searchConfig.keywords.length === 0 ? TODO_CONTAINER_TYPES : undefined;\n const allBlocks = await notionService.getPageBlocksRecursive(pageId, containerFilter);\n\n if (!allBlocks || allBlocks.length === 0) {\n return {\n success: true,\n blocksFound: 0,\n blockIds: []\n };\n }\n\n const blocksMatched: string[] = [];\n let blocksNew = 0;\n let blocksUpdated = 0;\n let blocksSkipped = 0;\n\n // Track preceding headings for context matching\n let currentHeading: string | null = null;\n\n // Search based on mode\n for (const block of allBlocks) {\n // Update current heading when we see one (for context in owner matching)\n // Reset heading when any non-heading, non-todo block is encountered\n // This ensures headings only apply to uninterrupted stacks of todos directly beneath them\n if (block.type === 'heading_1' || block.type === 'heading_2' || block.type === 'heading_3') {\n const headingContent = block[block.type]?.rich_text;\n if (headingContent && Array.isArray(headingContent)) {\n currentHeading = headingContent.map((rt: any) => rt.plain_text || '').join('');\n }\n } else if (block.type !== 'to_do') {\n // Any other block type (paragraph, list item, empty, etc.) resets heading\n currentHeading = null;\n }\n // Note: to_do blocks inherit currentHeading without changing it\n\n let matched = false;\n let matchInfo = '';\n\n // Block type match (primary - always active)\n if (block.type === searchConfig.blockType) {\n matched = true;\n matchInfo = `block type \"${searchConfig.blockType}\"`;\n }\n\n // Keyword match (additive - only if keywords configured)\n if (!matched && searchConfig.keywords.length > 0) {\n const matchedKeyword = searchConfig.keywords.find(kw => searchBlockForKeyword(block, kw));\n if (matchedKeyword) {\n matched = true;\n matchInfo = `keyword \"${matchedKeyword}\"`;\n }\n }\n\n if (matched) {\n const reducedBlock = reduceNotionBlock(block);\n\n // ENRICH PAGE TITLES: Replace \"Untitled\" with actual page titles\n if (reducedBlock.link_mentions && reducedBlock.link_mentions.length > 0) {\n for (const linkMention of reducedBlock.link_mentions) {\n // Only fetch title if it's currently \"Untitled\"\n if (linkMention.text === 'Untitled') {\n // Extract page ID from URL\n const mentionPageId = extractPageIdFromUrl(linkMention.url);\n\n if (mentionPageId) {\n // Fetch page title from Notion API\n const titleResult = await notionService.getPageTitle(mentionPageId);\n\n if (titleResult.success && titleResult.title) {\n // Replace \"Untitled\" with actual title\n linkMention.text = titleResult.title;\n }\n }\n }\n }\n\n // ENRICH PARENT DATABASE IDs: Batch lookup for dynamic relation mapping\n const mentionPageIds = reducedBlock.link_mentions\n .map((lm: any) => lm.pageId)\n .filter((id: string | null): id is string => id !== null);\n\n if (mentionPageIds.length > 0) {\n const parentDbResult = await batchGetPageParentDatabaseIds(mentionPageIds);\n\n if (parentDbResult.success) {\n for (const linkMention of reducedBlock.link_mentions) {\n if (linkMention.pageId) {\n linkMention.parentDatabaseId = parentDbResult.mapping.get(linkMention.pageId) || null;\n }\n }\n }\n }\n }\n\n // STABILITY CHECK: Only apply if requested (cron mode)\n if (applyStabilityDelay) {\n const stabilityMinutes = getBlockStabilityMinutes();\n\n if (stabilityMinutes > 0 && !isBlockStable(reducedBlock.last_edited_time, stabilityMinutes)) {\n continue; // Block too recent, skip\n }\n }\n\n // VALIDATION: Only save blocks that meet sync criteria\n\n // Check minimum word count\n const minWords = getMinBlockWords();\n const wordCount = countWords(reducedBlock.todo_string);\n\n if (wordCount < minWords) {\n continue; // Block too short, skip\n }\n\n // Note: Due date is determined by AI during save phase, with DEFAULT_DUE_DATE fallback\n // date_mentions now contains ALL dates from the block (no by/due filtering)\n\n // Capture block creator as author (separate from @mentions)\n // Author = block's created_by; Owner will be derived from AI Summary later\n const creator = getBlockCreator(block);\n (reducedBlock as any).author = creator; // May be null if no creator\n\n // Check if blob already exists\n const existingBlobResult = await blobService.getBlockFromBlob('todo', block.id);\n\n let shouldSave = false;\n let saveReason = '';\n let existingSyncMetadata: { synced?: boolean; syncing?: boolean; target_page_id?: string; archived?: boolean } = {};\n\n if (!existingBlobResult.success || !existingBlobResult.data) {\n // No existing blob, save as new\n shouldSave = true;\n saveReason = 'new';\n } else {\n // Preserve entire sync_metadata to avoid race conditions\n // A concurrent sync may have written target_page_id after we read but before we write\n existingSyncMetadata = existingBlobResult.data.sync_metadata || {};\n\n // Compare timestamps to detect block changes\n const existingTime = new Date(existingBlobResult.data.last_edited_time);\n const currentTime = new Date(reducedBlock.last_edited_time);\n\n // MINUTE-PRECISION WORKAROUND: Notion's last_edited_time is rounded to the minute.\n // If a checkbox is toggled within the same minute as the last edit, the timestamp\n // appears unchanged and the update wouldn't be detected. We explicitly compare\n // the checked state for to_do blocks to catch same-minute checkbox toggles.\n // See: https://developers.notion.com/reference/block (last_edited_time is minute-precision)\n const checkedChanged = block.type === 'to_do' &&\n existingBlobResult.data.checked !== reducedBlock.checked;\n\n if (currentTime > existingTime || checkedChanged) {\n shouldSave = true;\n saveReason = checkedChanged ? 'checked_changed' : 'updated';\n } else {\n shouldSave = false;\n }\n }\n\n // Save to blob storage if needed\n if (shouldSave) {\n // Extract parent block ID (null if parent is a page, not a block)\n const parentId = block.parent?.type === 'block_id' ? block.parent.block_id : null;\n\n // RE-READ blob to get latest sync_metadata (prevents race condition)\n // A sync may have completed and written target_page_id since we first read\n const latestBlobResult = await blobService.getBlockFromBlob('todo', block.id);\n const latestSyncMetadata = latestBlobResult.success && latestBlobResult.data\n ? (latestBlobResult.data.sync_metadata || {})\n : existingSyncMetadata;\n\n // Preserve target_page_id from latest sync_metadata (fallback to old field name)\n const preservedPageId = latestSyncMetadata.target_page_id || latestSyncMetadata.notion_page_id;\n\n // Add sync metadata - mark as needing sync, preserve fields from latest read\n // IMPORTANT: Preserve syncing flag to avoid breaking concurrent sync locks\n const blobData: any = {\n ...reducedBlock,\n page_url: pageUrl, // Source page URL\n parent_id: parentId, // Parent block ID for traversal\n preceding_heading: currentHeading, // Closest heading for context matching\n source_page_database_id: sourcePageDatabaseId, // For contextual relation mapping\n source_page_relations: sourcePageRelations, // Relation values from source page\n sync_metadata: {\n ...latestSyncMetadata, // Preserve all existing flags (syncing, archived, etc.)\n synced: false, // Mark as needing sync since block changed\n target_page_id: preservedPageId // Preserve page ID\n }\n };\n\n // Flag if block has children (for callout in todo page)\n if (block.has_children) {\n blobData.has_children = true;\n }\n\n const blobResult = await blobService.saveBlockToBlob(\n 'todo',\n block.id,\n blobData\n );\n\n if (!blobResult.success) {\n logger.error(`Failed to save blob: ${blobResult.error}`);\n }\n\n // Track counters based on save reason\n if (saveReason === 'new') {\n blocksNew++;\n } else {\n blocksUpdated++;\n }\n blocksMatched.push(block.id);\n } else {\n // Block unchanged, skip\n blocksSkipped++;\n }\n }\n }\n\n return {\n success: true,\n blocksFound: blocksNew + blocksUpdated,\n blocksNew,\n blocksUpdated,\n blocksSkipped,\n blockIds: blocksMatched\n };\n\n } catch (err: any) {\n return {\n success: false,\n blocksFound: 0,\n blockIds: [],\n error: `Failed to search for keywords`,\n details: err.message\n };\n }\n}\n\n/**\n * Helper function to escape special regex characters\n */\nfunction escapeRegex(str: string): string {\n return str.replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\$&');\n}\n\n/**\n * Helper function to search a block for a keyword\n *\n * For text keywords: Uses word boundary regex for exact word match, case-insensitive\n * For non-alphanumeric keywords (like emojis): Uses exact match\n *\n * Searches within:\n * - Standard rich_text blocks (paragraphs, headings, lists, toggles, etc.)\n * - Blocks inside columns (automatically handled by recursive fetching)\n *\n * Skips:\n * - Unsupported blocks\n * - Button blocks\n * - Table blocks\n * - Table row blocks\n *\n * @param block - Notion block object\n * @param keyword - Keyword to search for (default: \"todo\")\n */\nfunction searchBlockForKeyword(block: any, keyword: string = 'todo'): boolean {\n const blockType = block.type;\n\n // Skip if block type is not defined\n if (!blockType || !block[blockType]) {\n return false;\n }\n\n // Explicitly skip unsupported, button, table, and table_row blocks\n if (blockType === 'unsupported' || blockType === 'button' || blockType === 'table' || blockType === 'table_row') {\n return false;\n }\n\n let textContent = '';\n\n // Extract text from standard rich_text array (most block types)\n if (block[blockType]?.rich_text && Array.isArray(block[blockType].rich_text)) {\n textContent = block[blockType].rich_text\n .map((rt: any) => rt.plain_text || '')\n .join(' ');\n }\n\n // If no text content extracted, return false\n if (!textContent.trim()) {\n return false;\n }\n\n // Escape special regex characters in the keyword\n const escapedKeyword = escapeRegex(keyword);\n\n // Check if keyword contains only alphanumeric characters (and common word chars)\n const isAlphanumeric = /^[\\w\\s-]+$/.test(keyword);\n\n if (isAlphanumeric) {\n // Use word boundary for text keywords (case-insensitive)\n const regex = new RegExp(`\\\\b${escapedKeyword}\\\\b`, 'i');\n return regex.test(textContent);\n } else {\n // For emojis and special characters, use exact match (case-sensitive)\n return textContent.includes(keyword);\n }\n}\n",[],[412,713,723],"43d11f54-b962-11f0-8efb-0224a6c84d84","pageController.ts","019a26dc-99bf-7269-a8f5-f4ae4994f9d7",["D",1762249214392],"import * as notionService from '../services/notion/index.ts';\nimport { getRecentPagesLookbackHours } from '../utils/configUtils.ts';\n\n/**\n * Get a Notion page by ID\n * Generic controller that works with any Notion page type\n */\nexport async function getPage(pageId: string) {\n // Validate input\n if (!pageId || typeof pageId !== 'string') {\n return {\n success: false,\n data: null,\n error: \"Invalid pageId\",\n details: \"pageId must be a non-empty string\"\n };\n }\n\n // Call service\n const result = await notionService.getNotionPage(pageId);\n\n if (!result.success) {\n return {\n success: false,\n data: null,\n error: \"Failed to fetch page\",\n details: result.error\n };\n }\n\n // Filter sensitive data if needed\n // Example: Remove button properties\n const filteredData = filterSensitiveProperties(result.data);\n\n return {\n success: true,\n data: filteredData,\n error: null\n };\n}\n\n/**\n * Update a Notion page's properties\n * Generic controller that works with any Notion page type\n */\nexport async function updatePage(pageId: string, properties: any) {\n // Validate input\n if (!pageId || typeof pageId !== 'string') {\n return {\n success: false,\n data: null,\n error: \"Invalid pageId\",\n details: \"pageId must be a non-empty string\"\n };\n }\n\n if (!properties || typeof properties !== 'object') {\n return {\n success: false,\n data: null,\n error: \"Invalid properties\",\n details: \"properties must be a non-empty object\"\n };\n }\n\n // Call service\n const result = await notionService.updateNotionPage(pageId, properties);\n\n if (!result.success) {\n return {\n success: false,\n data: null,\n error: \"Failed to update page\",\n details: result.error\n };\n }\n\n // Filter sensitive data if needed\n const filteredData = filterSensitiveProperties(result.data);\n\n return {\n success: true,\n data: filteredData,\n error: null\n };\n}\n\n/**\n * Create a new page in a database\n */\nexport async function createPage(databaseId: string, properties: any) {\n // Validate input\n if (!databaseId || typeof databaseId !== 'string') {\n return {\n success: false,\n data: null,\n error: \"Invalid databaseId\",\n details: \"databaseId must be a non-empty string\"\n };\n }\n\n if (!properties || typeof properties !== 'object') {\n return {\n success: false,\n data: null,\n error: \"Invalid properties\",\n details: \"properties must be a non-empty object\"\n };\n }\n\n // Call service\n const result = await notionService.createNotionPage(databaseId, properties);\n\n if (!result.success) {\n return {\n success: false,\n data: null,\n error: \"Failed to create page\",\n details: result.error\n };\n }\n\n // Filter sensitive data if needed\n const filteredData = filterSensitiveProperties(result.data);\n\n return {\n success: true,\n data: filteredData,\n error: null\n };\n}\n\n/**\n * Get page content (blocks)\n */\nexport async function getPageContent(pageId: string) {\n // Validate input\n if (!pageId || typeof pageId !== 'string') {\n return {\n success: false,\n data: null,\n error: \"Invalid pageId\",\n details: \"pageId must be a non-empty string\"\n };\n }\n\n // Call service\n const result = await notionService.getPageBlocks(pageId);\n\n if (!result.success) {\n return {\n success: false,\n data: null,\n error: \"Failed to fetch page content\",\n details: result.error\n };\n }\n\n return {\n success: true,\n data: result.data,\n error: null\n };\n}\n\n/**\n * Get recent pages edited within the last X hours\n * Returns simplified page objects with id, title, url, last_edited_time, parent\n * Filters out:\n * - Archived pages\n * - Database objects\n * - Pages inside TODOS_DB_ID database\n *\n * @param hoursAgo - Number of hours to look back (default: RECENT_PAGES_LOOKBACK_HOURS env var, fallback 24)\n */\nexport async function getRecentPages(hoursAgo?: number) {\n // Validate input - use env var default if not provided\n const hours = hoursAgo ?? getRecentPagesLookbackHours();\n\n if (typeof hours !== 'number' || hours <= 0) {\n return {\n success: false,\n data: null,\n error: \"Invalid hours parameter\",\n details: \"hours must be a positive number\"\n };\n }\n\n // Call service\n const result = await notionService.searchRecentPages(hours);\n\n if (!result.success) {\n return {\n success: false,\n data: null,\n error: \"Failed to fetch recent pages\",\n details: result.error\n };\n }\n\n return {\n success: true,\n data: {\n pages: result.data,\n count: result.data?.length || 0,\n timeRange: `${hours} hours`\n },\n error: null\n };\n}\n\n/**\n * Filter sensitive properties from Notion data\n * Example: Remove button properties\n */\nfunction filterSensitiveProperties(data: any) {\n if (!data || !data.properties) {\n return data;\n }\n\n const filtered = {\n ...data,\n properties: Object.entries(data.properties)\n .filter(([_key, value]: [string, any]) => value.type !== 'button')\n .reduce((acc, [key, value]) => ({ ...acc, [key]: value }), {})\n };\n\n return filtered;\n}\n",[],[412,713,731],"84e31ca2-02ca-11f1-9771-42dde27851f2","healthController.ts","43d0e1d8-b962-11f0-8efb-0224a6c84d84",421,["D",1770320426113],"import * as notionService from '../services/notion/index.ts';\nimport * as blobService from '../services/blobService.ts';\nimport * as aiService from '../services/aiService.ts';\nimport {\n getDatabaseSchema,\n findTitleProperty,\n findPropertiesByType,\n findPropertyByName,\n findCandidateProperties,\n findPropertyByNamePatterns,\n type SchemaProperty\n} from '../services/notion/databases.ts';\nimport { discoverRelationProperties } from '../services/notion/relations.ts';\nimport { getRecentPagesLookbackHours } from '../utils/configUtils.ts';\nimport { getBlockStabilityMinutes } from '../utils/blockUtils.ts';\nimport { getMinBlockWords } from '../utils/textUtils.ts';\nimport { getDefaultDueDateSetting } from '../utils/dateUtils.ts';\nimport {\n TODOS_PROPERTIES,\n PROPERTY_METADATA,\n CATEGORY_METADATA,\n getEnvVarName,\n getEnvVarValue,\n getPropertyKeys,\n getDefaultPropertyName\n} from '../utils/notionConfig.ts';\nimport { getProjectInfo, type ProjectInfo } from '../utils/valUtils.ts';\nimport { logger } from '../utils/logger.ts';\nimport type {\n PropertyStatus,\n PropertyMetadata,\n HealthCheckResponse\n} from '../../shared/types.ts';\n\n// Cache configuration\nconst HEALTH_CACHE_KEY = 'health';\nconst HEALTH_CACHE_TTL_MS = 60_000; // 1 minute\n\n/**\n * Try to autodiscover a property using name pattern matching\n *\n * Conservative approach - only autodiscover when confident:\n * 1. Title property (unique per database) - safe, unambiguous\n * 2. Name pattern matching (+ correct type) - reasonable confidence\n *\n * Type-only matching is NOT used for autodiscovery (too risky for false positives).\n * Instead, type-compatible properties are shown as suggestions in `candidates`.\n */\nfunction tryAutodiscover(\n schemaProperties: SchemaProperty[],\n metadata: PropertyMetadata\n): SchemaProperty | null {\n const expectedTypes = metadata.expectedTypes;\n\n // Title property - safe because there's exactly one per database\n if (expectedTypes.includes('title')) {\n const titleProp = findTitleProperty(schemaProperties);\n if (titleProp) return titleProp;\n }\n\n // Name pattern matching - autodiscover if name matches expected patterns\n if (metadata.namePatterns && metadata.namePatterns.length > 0) {\n const matched = findPropertyByNamePatterns(\n schemaProperties,\n metadata.namePatterns,\n expectedTypes\n );\n if (matched) return matched;\n }\n\n // No type-only fallback - let candidates show suggestions instead\n return null;\n}\n\n/**\n * Build property status for a single expected property\n * Property-centric: starts from what we expect, checks against what exists\n *\n * Logic:\n * - Explicit env var → respect it, no autodiscovery fallback\n * - Default name → try exact match first, then autodiscovery if not found\n * - No default (null) → try autodiscovery\n */\nfunction buildPropertyStatus(\n key: string,\n schemaProperties: SchemaProperty[] | null\n): PropertyStatus {\n const metadata = PROPERTY_METADATA[key as keyof typeof PROPERTY_METADATA];\n const configuredName = (TODOS_PROPERTIES as Record<string, string | null>)[key];\n const envValue = getEnvVarValue(key);\n\n // Base status object\n const status: PropertyStatus = {\n key,\n displayName: metadata?.displayName || key,\n purpose: metadata?.purpose || 'Unknown',\n category: metadata?.category || 'enrichment',\n required: metadata?.required || false,\n envVar: getEnvVarName(key),\n envValue,\n defaultName: getDefaultPropertyName(key),\n foundName: null,\n foundType: null\n };\n\n // No database schema available - can't determine beyond config\n if (!schemaProperties) {\n return status;\n }\n\n // Case 1: User EXPLICITLY set an env var (envValue is not null)\n // Respect their choice - no autodiscovery fallback\n if (envValue !== null) {\n if (configuredName) {\n const found = findPropertyByName(schemaProperties, configuredName);\n if (found) {\n status.foundName = found.name;\n status.foundType = found.type;\n } else {\n status.candidates = findCandidateProperties(schemaProperties, configuredName).slice(0, 3);\n }\n }\n return status;\n }\n\n // Case 2: Using DEFAULT name (env var not set)\n if (configuredName) {\n const found = findPropertyByName(schemaProperties, configuredName);\n if (found) {\n status.foundName = found.name;\n status.foundType = found.type;\n return status;\n }\n\n // Default name not found - try autodiscovery if available\n if (metadata?.autodiscoverable) {\n const autodiscovered = tryAutodiscover(schemaProperties, metadata);\n if (autodiscovered) {\n status.foundName = autodiscovered.name;\n status.foundType = autodiscovered.type;\n return status;\n }\n }\n\n // Default not found and autodiscovery failed\n status.candidates = findCandidateProperties(schemaProperties, configuredName).slice(0, 3);\n return status;\n }\n\n // Case 3: No default (null) but autodiscoverable\n if (metadata?.autodiscoverable) {\n const autodiscovered = tryAutodiscover(schemaProperties, metadata);\n if (autodiscovered) {\n status.foundName = autodiscovered.name;\n status.foundType = autodiscovered.type;\n return status;\n }\n }\n\n // Case 4: Not configured and not autodiscoverable\n // Suggest candidates if we have expected types\n if (metadata?.expectedTypes && metadata.expectedTypes.length > 0) {\n const candidates = findPropertiesByType(schemaProperties, metadata.expectedTypes);\n if (candidates.length > 0) {\n status.candidates = candidates.map(c => c.name).slice(0, 3);\n }\n }\n\n return status;\n}\n\n/**\n * Build property statuses for all expected properties\n * Property-centric approach: iterate over expected properties, not configured values\n */\nfunction buildAllPropertyStatuses(\n schemaProperties: SchemaProperty[] | null\n): PropertyStatus[] {\n const keys = getPropertyKeys();\n return keys.map(key => buildPropertyStatus(key, schemaProperties));\n}\n\n/**\n * Generate a Notion formula for display string composition\n * Format: \"[Owner] to [task] by [due date]\"\n *\n * Uses format() which works universally for people and relation types.\n * Returns null if Name property is not mapped (required).\n */\nfunction generateDisplayFormula(properties: PropertyStatus[]): string | null {\n const nameProp = properties.find(p => p.key === 'name');\n const ownerProp = properties.find(p => p.key === 'owner');\n const dueDateProp = properties.find(p => p.key === 'dueDate');\n\n // Name is required\n if (!nameProp?.foundName) return null;\n\n const parts: string[] = [];\n\n // Owner part (optional)\n if (ownerProp?.foundName) {\n parts.push(`if(empty(prop(\"${ownerProp.foundName}\")), \"\", format(prop(\"${ownerProp.foundName}\")) + \" to \")`);\n }\n\n // Name part (required) - lowercase first letter\n parts.push(`if(empty(prop(\"${nameProp.foundName}\")), \"\", lower(substring(format(prop(\"${nameProp.foundName}\")), 0, 1)) + substring(format(prop(\"${nameProp.foundName}\")), 1))`);\n\n // Due date part (optional)\n if (dueDateProp?.foundName) {\n parts.push(`if(empty(prop(\"${dueDateProp.foundName}\")), \"\", \" by \" + formatDate(prop(\"${dueDateProp.foundName}\"), \"MMMM D, YYYY\"))`);\n }\n\n return parts.join(' + ');\n}\n\n/**\n * Setting configuration for health check\n */\ninterface SettingConfig {\n key: string;\n envVar: string;\n description: string;\n defaultValue: string | null;\n getValue: () => string | null;\n}\n\n/**\n * Get all settings configurations\n */\nfunction getSettingsConfigs(): SettingConfig[] {\n return [\n {\n key: 'searchBlockType',\n envVar: 'SEARCH_BLOCK_TYPE',\n description: 'Block type to search for (e.g., to_do for checkboxes)',\n defaultValue: 'to_do',\n getValue: () => {\n const val = Deno.env.get('SEARCH_BLOCK_TYPE');\n return val !== undefined ? (val.trim() || 'to_do') : null;\n }\n },\n {\n key: 'searchKeywords',\n envVar: 'SEARCH_KEYWORDS',\n description: 'Additional keywords to search for (comma-separated, optional)',\n defaultValue: null,\n getValue: () => Deno.env.get('SEARCH_KEYWORDS') || null\n },\n {\n key: 'minBlockWords',\n envVar: 'MIN_BLOCK_WORDS',\n description: 'Minimum word count for a block to be saved',\n defaultValue: String(getMinBlockWords()),\n getValue: () => Deno.env.get('MIN_BLOCK_WORDS') || null\n },\n {\n key: 'defaultDueDate',\n envVar: 'DEFAULT_DUE_DATE',\n description: 'Default due date (today, tomorrow, one_week, end_of_week, next_business_day)',\n defaultValue: 'today',\n getValue: () => Deno.env.get('DEFAULT_DUE_DATE') || null\n },\n {\n key: 'blockStabilityMinutes',\n envVar: 'BLOCK_STABILITY_MINUTES',\n description: 'Minutes to wait before syncing (prevents syncing while editing)',\n defaultValue: '0',\n getValue: () => Deno.env.get('BLOCK_STABILITY_MINUTES') || null\n },\n {\n key: 'blockBackgroundColor',\n envVar: 'BLOCK_BACKGROUND_COLOR',\n description: 'Background color for synced blocks (e.g., gray_background)',\n defaultValue: null,\n getValue: () => Deno.env.get('BLOCK_BACKGROUND_COLOR') || null\n },\n {\n key: 'recentPagesLookbackHours',\n envVar: 'RECENT_PAGES_LOOKBACK_HOURS',\n description: 'Hours to look back when searching for recent pages',\n defaultValue: '24',\n getValue: () => Deno.env.get('RECENT_PAGES_LOOKBACK_HOURS') || null\n },\n {\n key: 'cronsDisabled',\n envVar: 'CRONS_DISABLED',\n description: 'Set to \"true\" to disable scheduled cron jobs',\n defaultValue: 'false',\n getValue: () => Deno.env.get('CRONS_DISABLED') || null\n }\n ];\n}\n\n\n/**\n * Internal function to build health response (no caching)\n */\nasync function buildHealthResponse(): Promise<HealthCheckResponse> {\n const timestamp = new Date().toISOString();\n const project = getProjectInfo();\n\n const notionApiKey = Deno.env.get('NOTION_API_KEY');\n const todosDbId = Deno.env.get('TODOS_DB_ID');\n\n // Initialize status variables\n let connectionStatus: 'connected' | 'error' | 'not_configured' = 'not_configured';\n let connectionError: string | undefined;\n let integrationName: string | undefined;\n let databaseStatus: 'connected' | 'error' | 'not_configured' = 'not_configured';\n let databaseError: string | undefined;\n let databaseName: string | null = null;\n let databaseUrl: string | null = null;\n let schemaProperties: SchemaProperty[] | null = null;\n\n // Wave 1: Parallel fetch of bot user and database schema\n // These are independent and can run concurrently\n const [botResult, schemaResult] = await Promise.all([\n notionApiKey ? notionService.getBotUser() : Promise.resolve(null),\n todosDbId ? getDatabaseSchema(todosDbId) : Promise.resolve(null)\n ]);\n\n // Process bot result\n if (!notionApiKey) {\n connectionStatus = 'not_configured';\n } else if (botResult?.success && botResult.data) {\n connectionStatus = 'connected';\n integrationName = botResult.data.name;\n } else {\n connectionStatus = 'error';\n connectionError = botResult?.error || 'Failed to connect to Notion';\n }\n\n // Process schema result\n if (!todosDbId) {\n databaseStatus = 'not_configured';\n } else if (schemaResult?.success) {\n databaseStatus = 'connected';\n connectionStatus = 'connected'; // If DB works, connection works\n databaseName = schemaResult.databaseName;\n schemaProperties = schemaResult.properties;\n } else {\n databaseStatus = 'error';\n databaseError = schemaResult?.error || 'Failed to connect to database';\n }\n\n // Wave 2: Parallel fetch of database URL and relations (if database is connected)\n let relationsResult: Awaited<ReturnType<typeof discoverRelationProperties>> | null = null;\n\n if (todosDbId && databaseStatus === 'connected') {\n const [dbResult, relResult] = await Promise.all([\n notionService.getNotionDatabase(todosDbId),\n discoverRelationProperties(todosDbId)\n ]);\n\n if (dbResult.success && dbResult.data) {\n databaseUrl = (dbResult.data as any).url || null;\n }\n relationsResult = relResult;\n }\n\n // 3. Build property statuses (property-centric)\n const properties = buildAllPropertyStatuses(schemaProperties);\n\n // Wave 3: Parallel fetch of page counts AND database names for mapped relations\n // Only relations with TODOS_PROP_xxx mappings are active\n const relations: HealthCheckResponse['relations'] = [];\n // Map to merge relation metadata into properties later\n const relationMetadataByPropName = new Map<string, {\n targetDbId: string;\n targetDbName: string | null;\n pageCount: number;\n }>();\n\n if (relationsResult?.success && relationsResult.relations.length > 0) {\n // Build set of mapped relation property names from TODOS_PROPERTIES\n const mappedRelationNames = new Set(\n Object.values(TODOS_PROPERTIES).filter((v): v is string => typeof v === 'string' && v !== '')\n );\n\n // Filter to only mapped relations\n const mappedRelations = relationsResult.relations.filter(\n rel => mappedRelationNames.has(rel.propertyName)\n );\n\n // Fetch page counts AND database schemas in parallel for mapped relations\n const [pageCountResults, schemaResults] = await Promise.all([\n Promise.all(mappedRelations.map(rel =>\n notionService.getDatabasePageNames(rel.targetDatabaseId)\n )),\n Promise.all(mappedRelations.map(rel =>\n getDatabaseSchema(rel.targetDatabaseId)\n ))\n ]);\n\n // Build relations array and metadata map from parallel results\n mappedRelations.forEach((rel, index) => {\n const pageNamesResult = pageCountResults[index];\n const schemaResult = schemaResults[index];\n const pageCount = pageNamesResult.success ? pageNamesResult.pages.length : 0;\n const targetDbName = schemaResult.success ? schemaResult.databaseName : null;\n\n relations.push({\n propertyName: rel.propertyName,\n targetDatabaseId: rel.targetDatabaseId,\n pageCount\n });\n\n // Store metadata for merging into properties\n relationMetadataByPropName.set(rel.propertyName, {\n targetDbId: rel.targetDatabaseId,\n targetDbName,\n pageCount\n });\n });\n }\n\n // Merge relation metadata into properties\n for (const prop of properties) {\n if (prop.foundType === 'relation' && prop.foundName) {\n const metadata = relationMetadataByPropName.get(prop.foundName);\n if (metadata) {\n prop.relationTargetDbId = metadata.targetDbId;\n prop.relationTargetDbName = metadata.targetDbName;\n prop.relationPageCount = metadata.pageCount;\n }\n }\n }\n\n // 5. Build settings array\n const settingsConfigs = getSettingsConfigs();\n const settings: HealthCheckResponse['settings'] = settingsConfigs.map(config => ({\n key: config.key,\n envVar: config.envVar,\n value: config.getValue(),\n defaultValue: config.defaultValue,\n description: config.description\n }));\n\n // 6. Check OpenAI connection (validate API key with actual API call)\n const openaiApiKey = Deno.env.get('OPENAI_API_KEY');\n let openaiStatus: 'connected' | 'not_configured' | 'error' = 'not_configured';\n let openaiError: string | null = null;\n\n if (openaiApiKey) {\n const openaiResult = await aiService.validateOpenAIConnection();\n if (openaiResult.success) {\n openaiStatus = 'connected';\n } else {\n openaiStatus = 'error';\n openaiError = openaiResult.error || 'Failed to connect to OpenAI';\n }\n }\n\n // 7. Generate display formula from property mappings\n const displayFormula = generateDisplayFormula(properties);\n\n return {\n project,\n timestamp,\n connections: {\n notion: {\n envVar: 'NOTION_API_KEY',\n status: connectionStatus,\n ...(integrationName ? { integrationName } : {}),\n ...(connectionError ? { error: connectionError } : {})\n },\n database: {\n envVar: 'TODOS_DB_ID',\n status: databaseStatus,\n id: todosDbId || null,\n name: databaseName,\n url: databaseUrl,\n ...(databaseError ? { error: databaseError } : {})\n },\n openai: {\n envVar: 'OPENAI_API_KEY',\n status: openaiStatus,\n error: openaiError\n }\n },\n properties,\n categories: CATEGORY_METADATA as HealthCheckResponse['categories'],\n relations,\n settings,\n displayFormula\n };\n}\n\n/**\n * Health response with cache metadata\n */\nexport interface CachedHealthCheckResponse extends HealthCheckResponse {\n cached?: boolean;\n cacheAge?: number;\n}\n\n/**\n * Get system health status (with caching)\n *\n * Property-centric approach:\n * - Shows all expected properties and their state\n * - Autodiscovers properties where possible\n * - Suggests fixes for missing properties\n *\n * Caching:\n * - Caches response in blob storage for 1 minute\n * - Returns cached response if available and fresh\n * - Fetches fresh data if cache is stale or missing\n *\n * Returns structured health data for monitoring/debugging\n */\nexport async function getSystemHealth(forceRefresh: boolean = false): Promise<CachedHealthCheckResponse> {\n // Check cache first (unless force refresh requested)\n if (!forceRefresh) {\n const cached = await blobService.getCache<HealthCheckResponse>(HEALTH_CACHE_KEY);\n\n if (cached && Date.now() - cached.timestamp < HEALTH_CACHE_TTL_MS) {\n const cacheAge = Date.now() - cached.timestamp;\n logger.debug(`Health check: using cached response (age: ${Math.round(cacheAge / 1000)}s)`);\n return {\n ...cached.data,\n cached: true,\n cacheAge\n };\n }\n }\n\n // Cache miss, stale, or force refresh - fetch fresh data\n logger.debug('Health check: fetching fresh data');\n const freshData = await buildHealthResponse();\n\n // Save to cache (non-blocking)\n await blobService.setCache(HEALTH_CACHE_KEY, freshData);\n\n return freshData;\n}\n",[],[412,713,738],"71e9f9e6-b369-11f0-bc36-0224a6c84d84","faviconController.ts","b7798a9c-b361-11f0-9afd-0224a6c84d84",14,["D",1761592591230],"import * as notionService from '../services/notion/index.ts';\nimport * as faviconizerService from '../services/faviconizerService.ts';\nimport type { FaviconUpdateResult } from '../../shared/types.ts';\n\n/**\n * Process favicon update for a Notion page\n * Orchestrates: Get page → Extract Domain → Fetch favicon → Update icon\n * On error: Append error note to page\n */\nexport async function processFaviconUpdate(pageId: string): Promise<FaviconUpdateResult> {\n // Validate input\n if (!pageId || typeof pageId !== 'string') {\n return {\n success: false,\n error: \"Invalid pageId\",\n details: \"pageId must be a non-empty string\"\n };\n }\n\n // Step 1: Get the Notion page\n const pageResult = await notionService.getNotionPage(pageId);\n\n if (!pageResult.success) {\n return {\n success: false,\n error: \"Failed to fetch Notion page\",\n details: pageResult.error\n };\n }\n\n // Step 2: Extract domain from Name property (title type)\n const domain = pageResult.data?.properties?.Name?.title?.[0]?.plain_text;\n\n if (!domain || typeof domain !== 'string' || domain.trim() === '') {\n const errorMessage = \"⚠️ Favicon update failed: Name property not found or empty\";\n\n // Append error note to page\n await notionService.appendBlockToPage(pageId, errorMessage);\n\n return {\n success: false,\n iconUpdated: false,\n errorNoteAdded: true,\n error: \"Name property not found or empty\"\n };\n }\n\n // Step 3: Fetch favicon from Faviconizer\n const faviconResult = await faviconizerService.getFaviconForDomain(domain);\n\n if (!faviconResult.success || !faviconResult.data?.faviconUrl) {\n const errorMessage = `⚠️ Favicon update failed: Could not fetch favicon for domain '${domain}'. Error: ${faviconResult.error}`;\n\n // Append error note to page\n await notionService.appendBlockToPage(pageId, errorMessage);\n\n return {\n success: false,\n iconUpdated: false,\n errorNoteAdded: true,\n error: \"Failed to fetch favicon\",\n details: faviconResult.error\n };\n }\n\n // Step 4: Update page icon\n const iconUpdateResult = await notionService.updatePageIcon(pageId, faviconResult.data.faviconUrl);\n\n if (!iconUpdateResult.success) {\n const errorMessage = `⚠️ Favicon update failed for domain '${domain}': Could not update page icon. Error: ${iconUpdateResult.error}`;\n\n // Append error note to page\n await notionService.appendBlockToPage(pageId, errorMessage);\n\n return {\n success: false,\n iconUpdated: false,\n errorNoteAdded: true,\n error: \"Failed to update page icon\",\n details: iconUpdateResult.error\n };\n }\n\n // Success! Append success note to page\n const successMessage = `✅ Favicon successfully updated for domain: ${domain}`;\n const appendResult = await notionService.appendBlockToPage(pageId, successMessage);\n\n return {\n success: true,\n iconUpdated: true,\n errorNoteAdded: false,\n successNoteAdded: appendResult.success\n };\n}\n",[],[412,713,746],"1cb94fca-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[182,183,184,185,186,187],[412,713],"1cb966d6-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[181,188,192,204,216,227],[412],"f2153128-fc5a-11f0-b13e-42dde27851f2","settings.local.json","019a26dc-99bf-7269-a8f5-ec22817c2d0b","019a26dc-99bf-7269-a8f5-f075c4855e23",403,["D",1769612798707],7,"{\n \"permissions\": {\n \"allow\": [\n \"Bash(cat:*)\",\n \"mcp__ide__getDiagnostics\",\n \"WebFetch(domain:docs.val.town)\",\n \"WebFetch(domain:developers.notion.com)\",\n \"Bash(find:*)\",\n \"Bash(deno check:*)\",\n \"WebFetch(domain:platform.openai.com)\",\n \"WebSearch\"\n ],\n \"deny\": [],\n \"ask\": []\n }\n}\n",[],[771,762],".claude","1cb9476e-b360-11f0-bc6f-0224a6c84d84",["D",1761588582800],[179],[771],["D",1761588582800],["D",1789440062937],{"_11":12,"_129":74},{"_11":12,"_125":-5,"_73":74,"_30":72,"_126":127,"_128":24},{"_38":39,"_13":16,"_32":35,"_113":76,"_114":115,"_116":117,"_118":119,"_120":121,"_122":41,"_123":14,"_124":12},"ctx",{"_52":-5,"_149":814,"_815":816,"_817":818,"_819":24,"_820":24,"_821":24,"_822":87,"_823":-5,"_824":-7,"_825":-7,"_826":26,"_827":24,"_828":24,"_829":-5,"_830":24,"_831":24,"_832":24,"_833":-5},"ENV",{"_785":786,"_787":788,"_789":16,"_790":791,"_792":793,"_794":795,"_796":797,"_798":799,"_800":801,"_802":803,"_804":805,"_806":807,"_808":809,"_810":811,"_812":813},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","e87eb04e707cf64e7e532c8e8ec395f209e17f15","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"] </article></body></html>

History · LICENSE