Lorraine

[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_333":334,"_335":336},"routes/_app.x.$handle.$name",{"_95":94,"_31":43,"_103":85,"_26":51,"_13":14,"_104":-5,"_11":105,"_106":107,"_108":109,"_110":111,"_112":113,"_20":-5,"_114":-5,"_115":-5,"_116":24,"_117":118,"_119":24,"_120":-5,"_121":-5,"_122":123},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_41":42,"_31":43,"_36":44,"_45":-5,"_46":24,"_47":48,"_49":48,"_50":51,"_52":-5,"_18":53,"_54":55,"_56":57,"_58":57,"_59":-5,"_60":61,"_62":63,"_64":-5,"_65":-5,"_66":-5,"_67":-5,"_68":69,"_70":69,"_71":72,"_73":72,"_74":72,"_75":69,"_76":69,"_77":78,"_79":80,"_81":82,"_83":24},"version",19,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":-7,"_23":24,"_25":24,"_26":27,"_28":27},"file",{"_29":30,"_31":32,"_33":20,"_34":35,"_36":37,"_38":14,"_39":40,"_23":24},"previous","deleted",false,"deletedHere","path","schemas/response.schema.json","latestPath","fileId","76d573c4-7618-11f0-9291-0224a6c84d84","name","response.schema.json","fileType","content","{\n \"$schema\": \"https://json-schema.org/draft/2020-12/schema\",\n \"$id\": \"https://example.com/schemas/response.schema.json\",\n \"title\": \"User Response (Minimal)\",\n \"type\": \"object\",\n \"additionalProperties\": false,\n \"required\": [\"question_id\", \"raw_response\", \"created_at\"],\n \"properties\": {\n \"question_id\": { \"type\": \"string\", \"minLength\": 1 },\n \"raw_response\": { \"type\": \"string\", \"minLength\": 1 },\n \"created_at\": { \"type\": \"string\", \"format\": \"date-time\" },\n \"neutral_distance_note\": { \"type\": [\"string\", \"null\"] },\n \"response_latency_ms\": { \"type\": \"integer\", \"minimum\": 0 }\n }\n}\n\n\n","createdAt",["D",1754850789337],"branchVersion","branchId","621a44f0-7618-11f0-b48b-0224a6c84d84","id","62116484-7618-11f0-b48b-0224a6c84d84","lorraine",["D",1754850754170],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","","imageUrl","a8270e74-6479-475a-8025-9b69544ee0e7","pinnedFileId","76dc3d3a-7618-11f0-93b6-0224a6c84d84","pinPreview",true,"pinMigrated","screenshotVersion","likeCount",0,"orgId","8319defa-6e1a-46d4-ba66-2114ffd3360b","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_95":94,"_102":98},"org",{"_95":94,"_96":-5,"_97":98,"_18":53,"_99":100,"_101":24},"branch",{"_13":84,"_31":85,"_41":40,"_86":42,"_87":-5,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":43,"_93":94},"isOwned",70,"main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","yawnxyz","handle","displayName","avatar","https://images.clerk.dev/oauth_github/img_2NnaHhpxNuH1xWRIRjQNoo16TVc.jpeg","tier","pro","canRestrictVals","profileImageUrl","branchName","adoptOrgId",{"_41":42,"_31":43,"_36":329,"_45":-5,"_46":24,"_47":48,"_49":48,"_50":51,"_52":-5,"_18":53,"_54":55,"_56":57,"_58":57,"_59":-5,"_60":61,"_62":63,"_64":-5,"_65":-5,"_66":-5,"_67":-5,"_68":69,"_70":69,"_71":72,"_73":72,"_74":72,"_75":69,"_76":69,"_77":330,"_79":331,"_81":332,"_83":24},"tree",[213,216,218,222,224,227,228,229,230,231],"files",[213,214,215,216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231],"counts",{"_203":61,"_204":194,"_205":61,"_206":207,"_208":209},"contents",[124,125,126,127,128,129,130,131,132,133],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},{"_41":197,"_31":198,"_29":199,"_137":-5,"_39":40,"_38":200,"_139":57,"_36":201,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_26":202},{"_41":191,"_31":192,"_29":193,"_137":-5,"_39":40,"_38":194,"_139":57,"_36":195,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_26":196},{"_41":185,"_31":186,"_29":187,"_137":-5,"_39":40,"_38":188,"_139":57,"_36":189,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_26":190},{"_41":180,"_31":43,"_29":181,"_137":-5,"_39":40,"_38":182,"_139":57,"_36":183,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_26":184},{"_41":173,"_31":174,"_29":175,"_137":-5,"_39":40,"_38":176,"_139":57,"_36":177,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_26":179},{"_41":166,"_31":167,"_29":168,"_137":-5,"_39":40,"_38":169,"_139":57,"_36":170,"_13":61,"_33":163,"_34":171,"_23":24,"_142":-5,"_26":172},{"_41":158,"_31":159,"_29":160,"_137":-5,"_39":40,"_38":161,"_139":57,"_36":162,"_13":61,"_33":163,"_34":164,"_23":24,"_142":-5,"_26":165},{"_41":151,"_31":152,"_29":153,"_137":-5,"_39":40,"_38":154,"_139":24,"_36":155,"_13":61,"_33":20,"_34":156,"_23":24,"_142":-5,"_26":157},{"_41":144,"_31":145,"_29":146,"_137":-5,"_39":40,"_38":147,"_139":24,"_36":148,"_13":61,"_33":20,"_34":149,"_23":24,"_142":-5,"_26":150},{"_41":134,"_31":135,"_29":136,"_137":-5,"_39":40,"_38":138,"_139":57,"_36":140,"_13":61,"_33":20,"_34":141,"_23":24,"_142":-5,"_26":143},"76d2dbfa-7618-11f0-afda-0224a6c84d84","readme.md","76d2dc9a-7618-11f0-afda-0224a6c84d84","parentId",18,"latest",["D",1754850789295],"### (re)Search — Groq demo\n\nTiny, no-nonsense demo for Groq: search the web and papers, then get clean summaries.\n\n- Search with optional source/date filters\n- Streams model reasoning/tools (SSE)\n- Returns tight JSON (deep links + one-line descriptions)\n- One‑click page summaries in your language\n\n### Run\n\n```sh\ndeno run -A smallweb/re-search/main.js \n```\n\nOpen http://localhost:8000 and paste your API key if the server doesn’t have one.\n\nPowered by Groq — get a key at [groq.com](https://groq.com).\n","label",[135],"76adfd58-7618-11f0-aa22-0224a6c84d84","index.html","76adfe0c-7618-11f0-aa22-0224a6c84d84",12,["D",1754850789070],"\n\n\n \n \n Lorraine\n \n \n \n \n \n \n\n \n \n \n \n \n \n \n \n \n \n \n \n\n \n\n\n

\n\n \n\n \n\n \n

\n \n

\n\n

\n\n \n

\n \n \n

\n \n

\n\n \n

\n \n \n

\n \n

\n\n \n

\n \n set\n

\n\n

\n \n \n \n \n

\n\n \n

\n \n

\n max\n \n tags\n \n \n \n \n

\n \n \n

\n \n\n

Loading questions…

\n\n \n

\n \n \n \n \n

\n \n

\n\n \n

0 || isStreaming\">\n

\n \n

\n\n \n

\n \n \n \n\n \n

\n \n

\n \n \n

\n \n

\n\n \n

\n \n \n \n github\n

\n \n\n \n\n \n\n \n \n \n\n",[145],"76a2f50c-7618-11f0-8008-0224a6c84d84","design_notes.md","76a2f638-7618-11f0-8008-0224a6c84d84",10,["D",1754850788963],"## Lorraine — Personal Voice Dataset Builder\n\nPurpose: Replace the existing research search UI with a simple, guided Q&A flow that displays many source wordings (formal, business, casual, social, etc.) and captures how the user would naturally say the same thing. Collected answers form a small dataset that can be used to fine‑tune a LoRA on the user's casual style.\n\nGoals\n- Present prompts from a JSON file (no model required to ask questions)\n- Capture answers via text input and optional voice dictation (Groq Whisper)\n- Keep the clean, minimal design from the current app (same fonts, spacing, Tailwind) and keep the API key UX\n- Export answers as JSON for downstream fine‑tuning\n\nData Model\n- Prompts JSON: array of items\n - id: string\n - label: short descriptor\n - instruction: what to do (e.g., \"Rewrite this in your style\")\n - source_text: the wording to adapt\n - tags: string[] (optional)\n- Answers (client-side):\n - question_id: string\n - answer_text: string\n - transcript_text?: string (if captured via Whisper)\n - created_at: ISO timestamp\n\nFrontend Flow\n- Load prompts from /data/questions.json\n- Show current prompt card (instruction + source text)\n- Provide a text area for \"Your version\"\n- Optional mic button to record and transcribe with Whisper; user can insert the transcript as their answer\n- Navigation: Previous / Next; progress indicator\n- Export: Download JSON of all answers\n- Persist API key (Groq) in localStorage; server key supported if present\n\nBackend Endpoints (Hono)\n- GET /data/questions.json → serve local JSON file (Val Town compatible loader)\n- POST /api/transcribe → forwards uploaded audio to Groq Whisper (`model: whisper-large-v3`) and returns `{ text }`\n- Existing /api/check-key retained for client key handling\n\nExport Format\n```json\n{\n \"app\": \"lorraine\",\n \"version\": 1,\n \"answers\": [\n {\n \"question_id\": \"q1\",\n \"answer_text\": \"...\",\n \"transcript_text\": \"...\",\n \"created_at\": \"2025-01-01T00:00:00.000Z\"\n }\n ]\n}\n```\n\nNotes / Future\n- Add prompt categories and filters\n- Add session save/restore in localStorage\n- Add CSV export and metadata like language/setting\n- Add batch microphone capture (hands‑free), and per‑answer audio blob export if needed\n\n\n",[152],"76bcd40e-7618-11f0-afda-0224a6c84d84","chatStreaming.js","76bcd4e0-7618-11f0-afda-0224a6c84d84",15,["D",1754850789182],"script","// Deno script and module: streaming chat completion test with SSE parsing and aggregation\n// - Reads GROQ_API_KEY from environment (when run as CLI)\n// - Exports streamChatAndAggregate(apiKey, payload, { logSse }) to allow server reuse\n// - CLI: logs raw SSE lines and prints aggregated JSON at the end\n\nimport \"jsr:@std/dotenv/load\";\n\nconst API_URL = \"https://api.groq.com/openai/v1/chat/completions\";\n\nfunction getApiKey() {\n const key = (typeof Deno !== \"undefined\") ? Deno.env.get(\"GROQ_API_KEY\") : null;\n if (!key) {\n console.error(\"Missing GROQ_API_KEY in environment.\");\n }\n return key;\n}\n\nfunction buildPayload(userQuery) {\n const prompt = typeof userQuery === \"string\" && userQuery.trim().length > 0\n ? userQuery\n : \"Stream test: please emit a short response and demonstrate reasoning and any tool usage.\";\n\n return {\n messages: [\n {\n role: \"user\",\n content: prompt,\n },\n ],\n model: \"openai/gpt-oss-120b\",\n stream: true,\n reasoning_effort: \"medium\",\n tools: [\n {\n type: \"browser_search\",\n },\n ],\n };\n}\n\n// Utility: Parse one Groq SSE JSON line (string without the leading \"data: \")\nfunction parseGroqSseJsonLine(jsonStr) {\n try {\n const obj = JSON.parse(jsonStr);\n return obj;\n } catch (_err) {\n return null;\n }\n}\n\n// Utility: Normalize a parsed chunk into events and collapsed accumulators\nfunction applyChunkToAggregates(parsed, aggregates) {\n const { events, collapsed } = aggregates;\n\n const timestamp = Date.now();\n const baseInfo = {\n id: parsed.id,\n model: parsed.model,\n created: parsed.created,\n instance_id: parsed.instance_id,\n system_fingerprint: parsed.system_fingerprint,\n timestamp,\n };\n\n if (Array.isArray(parsed.choices) && parsed.choices.length > 0) {\n const choice = parsed.choices[0];\n const delta = choice?.delta || {};\n\n if (delta.role) {\n events.push({ type: \"role\", role: delta.role, ...baseInfo });\n }\n\n if (delta.reasoning) {\n events.push({ type: \"reasoning\", content: delta.reasoning, channel: delta.channel, ...baseInfo });\n collapsed.reasoning += delta.reasoning;\n }\n\n if (delta.content) {\n events.push({ type: \"content\", content: delta.content, ...baseInfo });\n collapsed.content += delta.content;\n }\n\n if (delta.executed_tools) {\n events.push({ type: \"executed_tools\", executed_tools: delta.executed_tools, ...baseInfo });\n for (const t of delta.executed_tools) {\n collapsed.tools.push({ kind: \"executed\", ...t, timestamp });\n }\n }\n\n if (delta.tool_calls) {\n events.push({ type: \"tool_calls\", tool_calls: delta.tool_calls, ...baseInfo });\n for (const t of delta.tool_calls) {\n collapsed.tools.push({ kind: \"call\", ...t, timestamp });\n }\n }\n }\n}\n\nexport async function streamChatAndAggregate(apiKey, payload, options = {}) {\n const { logSse = false } = options;\n\n const response = await fetch(API_URL, {\n method: \"POST\",\n headers: {\n \"Content-Type\": \"application/json\",\n Authorization: `Bearer ${apiKey}`,\n },\n body: JSON.stringify(payload),\n });\n\n if (!response.ok) {\n const text = await response.text();\n throw new Error(`HTTP ${response.status} — ${text}`);\n }\n\n const reader = response.body?.getReader();\n const aggregates = {\n events: [],\n collapsed: { reasoning: \"\", content: \"\", tools: [] },\n };\n\n if (!reader) {\n try {\n const nonStream = await response.json();\n const content = nonStream?.choices?.[0]?.message?.content || \"\";\n if (content) {\n aggregates.events.push({ type: \"content\", content, timestamp: Date.now() });\n aggregates.collapsed.content = content;\n }\n } catch (_e) {\n // ignore\n }\n console.log(\"No readable stream; returning aggregates:\");\n console.log(JSON.stringify(aggregates, null, 2));\n return;\n }\n\n const decoder = new TextDecoder();\n let buffer = \"\";\n if (logSse) console.log(\"--- BEGIN SSE ---\");\n while (true) {\n const { done, value } = await reader.read();\n if (done) break;\n buffer += decoder.decode(value, { stream: true });\n\n // Split Server-Sent Events by double newlines\n const events = buffer.split(\"\\n\\n\");\n buffer = events.pop() || \"\";\n\n for (const evt of events) {\n if (!evt.trim()) continue;\n // Extract `data:` line(s)\n const lines = evt.split(\"\\n\");\n for (const line of lines) {\n if (line.startsWith(\"data: \")) {\n const data = line.slice(6);\n if (data === \"[DONE]\") {\n if (logSse) console.log(\"[DONE]\");\n break;\n }\n if (logSse) {\n // Log raw SSE packet\n console.log(\"data:\", data);\n }\n // Parse and aggregate\n const parsed = parseGroqSseJsonLine(data);\n if (!parsed) continue;\n applyChunkToAggregates(parsed, aggregates);\n }\n }\n }\n }\n if (logSse) {\n console.log(\"--- END SSE ---\");\n console.log(\"Aggregates:\");\n console.log(JSON.stringify(aggregates, null, 2));\n }\n return aggregates;\n}\n\n// Entry point\nif (import.meta.main) {\n const apiKey = getApiKey();\n if (!apiKey) Deno.exit(1);\n\n const userQuery = Deno.args?.[0] || \"\";\n const payload = buildPayload(userQuery);\n\n streamChatAndAggregate(apiKey, payload, { logSse: true })\n .then(() => console.log(\"Stream complete.\"))\n .catch((err) => {\n console.error(\"Stream failed:\", err);\n Deno.exit(1);\n });\n}\n\n\n",[159],"76948b52-7618-11f0-b5b6-0224a6c84d84","chatCompletion.js","76948c74-7618-11f0-b5b6-0224a6c84d84",8,["D",1754850788916],"// Minimal Groq Chat Completions client using fetch\nimport { extractOrRepairJsonResults, extractStructuredSummary, renderStructuredSummaryToHtml } from \"./jsonUtils.js\";\nexport async function groqChatCompletion(apiKey, payload) {\n console.log('>>> [groqChatCompletion] Payload:', payload);\n const response = await fetch('https://api.groq.com/openai/v1/chat/completions', {\n method: 'POST',\n headers: {\n 'Content-Type': 'application/json',\n 'Authorization': `Bearer ${apiKey}`,\n },\n body: JSON.stringify(payload),\n });\n if (!response.ok) {\n const text = await response.text();\n throw new Error(`Groq API error: ${response.status} ${text}`);\n }\n console.log('>>> [groqChatCompletion] Response:', response);\n return await response.json();\n}\n\n// Use LLM to normalize any freeform output into strict JSON that matches our schema.\n// extractOrRepairJsonResults moved to ./jsonUtils.js\n\nexport function buildSearchPrompt(query, source, dateRange, language, offset) {\n // Always search in English; present in the selected language\n const presentationLanguage = (typeof language === 'string' && language.trim()) ? language : 'english';\n const languageInstruction = `IMPORTANT: Always conduct searches in English. Workflow: 1) If the user query is not in English, translate \"${query}\" into English search terms and use ONLY the English translation for searching. 2) After finding results, present all titles and descriptions in ${presentationLanguage}. Always search in English but present results in ${presentationLanguage}.`;\n\n const base = source && source.trim()\n ? `Search ${source} for: \"${query}\"`\n : `\"${query}\"`;\n\n let dateRangeText = '';\n if (dateRange) {\n dateRangeText = ({\n '1month': 'in the last month',\n '3months': 'in the last 3 months',\n '6months': 'in the last 6 months',\n '1year': 'in the last year',\n })[dateRange] || '';\n }\n\n const startResult = offset + 1;\n const endResult = offset + 5;\n const offsetInstruction = offset > 0 ? ` Skip the first ${offset} results and` : '';\n\n const searchPrompt = `${base}${dateRangeText ? ' ' + dateRangeText : ''}.${offsetInstruction} Return results ${startResult}-${endResult} (max 5 new results). Return ONLY JSON with a top-level \\\\\"results\\\\\" array. Each result must include: title and url. Optionally include: description (one concise sentence only if confidently grounded by the target page). Do NOT add or guess fields like source, published_at, comments, pmid, or doi — omit anything unknown.\\n\\nURL RULES (critical):\\n- Always provide a direct deep link to the specific item/resource — never a homepage, category page, or search results page.\\n- If the listing points to another resource, set url to the canonical content page, not the aggregator:\\n - Hacker News: prefer the external article URL if present; otherwise use the exact item permalink (https://news.ycombinator.com/item?id=...). Do not invent dates or comment counts.\\n - PubMed: use the PMID page (https://pubmed.ncbi.nlm.nih.gov/PMID/) or the publisher DOI landing page if available. Do not link to the PubMed homepage.\\n - arXiv: use https://arxiv.org/abs/ (or the PDF if explicitly asked for PDFs).\\n - DOI: use the https://doi.org/... resolver link.\\n - Reddit/Forums: use the exact thread permalink if the discussion is the result.\\n - GitHub: link to the exact repository, release, issue, or file path, not github.com root.\\n- Never return root domains (e.g., https://pubmed.ncbi.nlm.nih.gov/, https://news.ycombinator.com/). Ensure url resolves to the specific item. ${languageInstruction}`;\n return searchPrompt;\n}\n\n// Lightweight router: decide between 'links' (JSON list) and 'text' (short answer)\nexport async function routeQueryMode(apiKey, query) {\n try {\n const res = await groqChatCompletion(apiKey, {\n model: 'openai/gpt-oss-120b',\n messages: [\n { role: 'system', content: 'Classify the user request as either links or text. Respond with a single token: links or text. Use links if the user appears to want a search results list of sources; use text if the user is asking for an explanation/summary/definition.' },\n { role: 'user', content: String(query || '') }\n ],\n stream: false,\n });\n const out = (res?.choices?.[0]?.message?.content || '').trim().toLowerCase();\n if (out.includes('links')) return 'links';\n if (out.includes('text')) return 'text';\n } catch (_) { /* fall back to heuristic */ }\n // Heuristic fallback\n const q = String(query || '').trim().toLowerCase();\n const chatStarts = ['explain', 'what is', 'what are', 'how does', 'how do', 'summarize', 'overview', 'define', 'in simple terms', 'briefly', 'why is', 'compare', 'contrast'];\n if (chatStarts.some((s) => q.startsWith(s))) return 'text';\n if (q.endsWith('?') && q.length <= 200) return 'text';\n return 'links';\n}\n\nexport async function performSearch({ apiKey, query, dateRange, source, language, offset = 0, model, stream, reasoningEffort, forceLinksRoute = false }) {\n const prompt = buildSearchPrompt(query, source, dateRange, language, offset);\n const start = Date.now();\n const route = forceLinksRoute ? 'links' : await routeQueryMode(apiKey, query);\n const response = await groqChatCompletion(apiKey, {\n model,\n messages: route === 'text'\n ? [\n {\n role: 'system',\n content: `If the user is asking for an explanation or summary rather than a list of links, respond with a concise plain-text answer in ${language || 'english'} (no JSON). If the request appears to be a search for sources, return ONLY JSON with a top-level results array as previously specified. ALWAYS conduct searches in English when needed.`\n },\n { role: 'user', content: query }\n ]\n : [\n {\n role: 'system',\n content: `Return ONLY JSON matching the schema; no markdown or commentary. For every result, url MUST be a direct deep link to the specific item/resource (not a homepage, category page, or search results page). Prefer canonical targets: external article for HN listings, PubMed PMID page or publisher DOI for papers, arXiv abs page, DOI resolver, exact thread permalinks for discussions, and precise GitHub paths. Never output root domains as url. Additionally, ALWAYS conduct searches in English. If the user query is not in English, translate it to English first and use ONLY the English translation for searching. Present titles and descriptions in ${language || 'english'}.`\n },\n { role: 'user', content: prompt },\n ],\n stream: false,\n reasoning_effort: reasoningEffort,\n tools: [ { type: 'browser_search' } ],\n });\n const end = Date.now();\n const searchTime = (end - start) / 1000;\n const contentText = response.choices?.[0]?.message?.content || '';\n\n console.log('>> contentText', contentText);\n let results = [];\n if (route === 'links') {\n try {\n // Try direct parse first (in case the model already returned valid JSON)\n const stripped = String(contentText).replace(/^```json\\s*/i, '').replace(/```\\s*$/i, '');\n const parsed = JSON.parse(stripped);\n if (Array.isArray(parsed?.results)) results = parsed.results;\n } catch (_) {\n // as a fallback, attempt repair pass\n results = await extractOrRepairJsonResults(contentText, apiKey, language);\n }\n } else {\n // text mode → do not attempt to coerce content into results\n results = [];\n }\n // Also return assistant content and full response for upstream logging / fallback display\n return { results, searchTime, content: contentText, full: response, route };\n}\n\nimport { marked } from \"npm:marked\";\n\nfunction buildSummaryPrompt(url, title, language) {\n const languageInstruction = language && language !== 'english'\n ? `Please provide the summary in ${language}. `\n : '';\n\n return `Visit this URL: ${url}\n \nTitle: \"${title}\"\n\nCRITICAL INSTRUCTIONS:\n- Read the page content and produce a clean, professional summary.\n- Ignore navigation, comments, ads, headers/footers, and numerical artifacts.\n- If this is a research paper, use the RESEARCH format below.\n- If this is a news/blog/listing/discussion page (e.g., HN, Reddit, news site), use the GENERAL format below.\n- DO NOT include any citations, reference numbers, or raw scraping artifacts.\n- Write in clear prose without formatting artifacts.\n- The opening line of the summary must not simply repeat the title; avoid mirroring the title.\n\nReturn only one of the two structures below in markdown (choose based on page type). ${languageInstruction}\n\nRESEARCH FORMAT:\n

\n**Research Question:** [brief description of what the study investigates]\n\n**Methods & Findings:** [1-2 paragraphs covering methodology and key results - NO CITATIONS]\n\n**Significance:** [why this research matters and potential applications - NO CITATIONS]\n

\n\nGENERAL FORMAT:\n

\n**What it is:** [brief description of the page]\n\n**Key points:** [bullet list or short paragraphs with the main takeaways]\n\n**Details:** [1–2 short paragraphs with notable specifics]\n

`;\n}\n\nexport async function summarizeUrl({ apiKey, url, title, language, model, stream, reasoningEffort }) {\n const summaryPrompt = buildSummaryPrompt(url, title, language);\n const start = Date.now();\n const response = await groqChatCompletion(apiKey, {\n model,\n messages: [\n {\n role: 'system',\n content: 'You are a concise web content summarization expert. CRITICAL REQUIREMENTS: 1) Return ONLY the summary content within

tags - no other text, explanations, or metadata. 2) Use clean, professional markdown formatting. 3) NEVER include citations, reference numbers, line numbers, or any numerical markers like [1†, L0:, L1:, etc. 4) Write in clear prose without any formatting artifacts from source documents. 5) Adapt structure to page type and, if it is a research paper, strictly use the sections: Research Question, Methods & Findings, Significance; otherwise use the GENERAL structure.'\n },\n {\n role: 'user',\n content: summaryPrompt\n }\n ],\n stream,\n reasoning_effort: reasoningEffort,\n tools: [ { type: 'browser_search' } ],\n });\n const end = Date.now();\n const summaryTime = (end - start) / 1000;\n const responseText = response.choices[0]?.message?.content || '';\n\n // Prefer structured, JSON-cleaned summary to avoid leaking citation junk\n let cleanedHtml = '';\n try {\n const inner = (responseText.match(/

([\\s\\S]*?)<\\/summary>/) || [null, responseText])[1];\n const structured = await extractStructuredSummary(inner, apiKey);\n if (structured) {\n cleanedHtml = renderStructuredSummaryToHtml(structured);\n }\n } catch (_) {\n // ignore and fallback\n }\n if (!cleanedHtml) {\n // Final fallback: render markdown, but still try to strip outer

tags\n const match = responseText.match(/

([\\s\\S]*?)<\\/summary>/);\n const raw = match ? match[1].trim() : responseText;\n cleanedHtml = marked(raw);\n }\n\n return { summary: cleanedHtml, summaryTime };\n}\n\n\n",[167],"7643624a-7618-11f0-9f35-0224a6c84d84","schemas","76436362-7618-11f0-9f35-0224a6c84d84",4,["D",1754850788382],"directory",[174],"76311d7e-7618-11f0-9d27-0224a6c84d84","76311e46-7618-11f0-9d27-0224a6c84d84",3,["D",1754850788267],[43],"7624a31e-7618-11f0-9d27-0224a6c84d84","frontend","7624a3f0-7618-11f0-9d27-0224a6c84d84",2,["D",1754850788185],[186],"7616c442-7618-11f0-9f35-0224a6c84d84","design","7616c514-7618-11f0-9f35-0224a6c84d84",1,["D",1754850788090],[192],"7651ff6c-7618-11f0-afda-0224a6c84d84","data","7652003e-7618-11f0-afda-0224a6c84d84",5,["D",1754850788481],[198],"pulls","branches","forks","envVars",{"_210":61,"_212":61},"connections",{"_210":61,"_211":61},"count","unfilledCount","emptyCount",{"_41":197,"_31":198,"_29":199,"_137":-5,"_39":40,"_38":200,"_139":57,"_36":326,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_233":327,"_26":328},{"_41":321,"_31":320,"_29":315,"_137":199,"_39":40,"_38":322,"_139":57,"_36":323,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_233":324,"_26":325},{"_41":312,"_31":313,"_29":314,"_137":315,"_39":40,"_38":316,"_139":24,"_36":317,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_233":318,"_26":319},{"_41":191,"_31":192,"_29":193,"_137":-5,"_39":40,"_38":194,"_139":57,"_36":309,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_233":310,"_26":311},{"_41":301,"_31":302,"_29":303,"_137":193,"_39":40,"_38":304,"_139":24,"_36":305,"_13":61,"_33":20,"_34":306,"_23":24,"_142":-5,"_233":307,"_26":308},{"_41":185,"_31":186,"_29":187,"_137":-5,"_39":40,"_38":188,"_139":57,"_36":298,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_233":299,"_26":300},{"_41":290,"_31":291,"_29":292,"_137":187,"_39":40,"_38":293,"_139":24,"_36":294,"_13":61,"_33":163,"_34":295,"_23":24,"_142":-5,"_233":296,"_26":297},{"_41":282,"_31":283,"_29":284,"_137":187,"_39":40,"_38":285,"_139":24,"_36":286,"_13":61,"_33":163,"_34":287,"_23":24,"_142":-5,"_233":288,"_26":289},{"_41":274,"_31":275,"_29":276,"_137":187,"_39":40,"_38":277,"_139":57,"_36":278,"_13":61,"_33":163,"_34":279,"_23":24,"_142":-5,"_233":280,"_26":281},{"_41":180,"_31":43,"_29":181,"_137":-5,"_39":40,"_38":182,"_139":57,"_36":271,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_233":272,"_26":273},{"_41":263,"_31":264,"_29":265,"_137":181,"_39":40,"_38":266,"_139":57,"_36":267,"_13":61,"_33":20,"_34":268,"_23":24,"_142":-5,"_233":269,"_26":270},{"_41":173,"_31":174,"_29":175,"_137":-5,"_39":40,"_38":176,"_139":57,"_36":260,"_13":61,"_33":178,"_34":-5,"_23":24,"_142":-5,"_233":261,"_26":262},{"_41":252,"_31":253,"_29":254,"_137":175,"_39":40,"_38":255,"_139":57,"_36":256,"_13":61,"_33":20,"_34":257,"_23":24,"_142":-5,"_233":258,"_26":259},{"_41":248,"_31":32,"_29":30,"_137":175,"_39":40,"_38":14,"_139":57,"_36":249,"_13":61,"_33":20,"_34":35,"_23":24,"_142":-5,"_233":250,"_26":251},{"_41":166,"_31":167,"_29":168,"_137":-5,"_39":40,"_38":169,"_139":57,"_36":245,"_13":61,"_33":163,"_34":171,"_23":24,"_142":-5,"_233":246,"_26":247},{"_41":158,"_31":159,"_29":160,"_137":-5,"_39":40,"_38":161,"_139":57,"_36":242,"_13":61,"_33":163,"_34":164,"_23":24,"_142":-5,"_233":243,"_26":244},{"_41":151,"_31":152,"_29":153,"_137":-5,"_39":40,"_38":154,"_139":24,"_36":239,"_13":61,"_33":20,"_34":156,"_23":24,"_142":-5,"_233":240,"_26":241},{"_41":144,"_31":145,"_29":146,"_137":-5,"_39":40,"_38":147,"_139":24,"_36":236,"_13":61,"_33":20,"_34":149,"_23":24,"_142":-5,"_233":237,"_26":238},{"_41":134,"_31":135,"_29":136,"_137":-5,"_39":40,"_38":138,"_139":57,"_36":232,"_13":61,"_33":20,"_34":141,"_23":24,"_142":-5,"_233":234,"_26":235},["D",1754850789295],"children",[],[135],["D",1754850789070],[],[145],["D",1754850788963],[],[152],["D",1754850789182],[],[159],["D",1754850788916],[],[167],"76d572e8-7618-11f0-9291-0224a6c84d84",["D",1754850789337],[],[174,32],"76c5f908-7618-11f0-b7e2-0224a6c84d84","question.schema.json","76c5f9d0-7618-11f0-b7e2-0224a6c84d84",17,["D",1754850789241],"{\n \"$schema\": \"https://json-schema.org/draft/2020-12/schema\",\n \"$id\": \"https://example.com/schemas/question.schema.json\",\n \"title\": \"Question Item (Minimal)\",\n \"type\": \"object\",\n \"additionalProperties\": false,\n \"required\": [\"id\", \"label\", \"instruction\", \"tags\"],\n \"properties\": {\n \"id\": { \"type\": \"string\", \"minLength\": 1 },\n \"label\": { \"type\": \"string\", \"minLength\": 1 },\n \"instruction\": { \"type\": \"string\", \"minLength\": 1 },\n \"tags\": {\n \"type\": \"array\",\n \"items\": { \"type\": \"string\" },\n \"minItems\": 0,\n \"uniqueItems\": false\n },\n \"neutral_sample\": { \"type\": [\"string\", \"null\"] },\n \"example\": { \"type\": [\"string\", \"null\"] }\n }\n}\n\n\n",[],[174,253],["D",1754850788382],[225,226],[174],"76c02b54-7618-11f0-aa22-0224a6c84d84","deno.json","76c02c80-7618-11f0-aa22-0224a6c84d84",16,["D",1754850789124],"{\n \"$schema\": \"https://raw.githubusercontent.com/denoland/deno/348900b8b79f4a434cab4c74b3bc8d4d2fa8ee74/cli/schemas/config-file.v1.json\",\n \"lock\": false,\n \"compilerOptions\": {\n \"noImplicitAny\": false,\n \"strict\": false,\n \"types\": [\n \"https://www.val.town/types/valtown.d.ts\"\n ],\n \"lib\": [\n \"dom\",\n \"dom.iterable\",\n \"dom.asynciterable\",\n \"deno.ns\",\n \"deno.unstable\"\n ]\n },\n \"lint\": {\n \"files\": {\n \"include\": [\n \"deno:/https/esm.town/**/*\"\n ]\n },\n \"rules\": {\n \"exclude\": [\n \"no-explicit-any\"\n ]\n }\n },\n \"node_modules_dir\": false,\n \"experimental\": {\n \"unstable-node-globals\": true,\n \"unstable-temporal\": true,\n \"unstable-worker-options\": true,\n \"unstable-sloppy-imports\": true\n }\n}",[],[43,264],["D",1754850788267],[223],[43],"76ab52ba-7618-11f0-bde4-0224a6c84d84","researchApp.js","76ab535a-7618-11f0-bde4-0224a6c84d84",11,["D",1754850789066],"export default function researchApp() {\n return {\n searchQuery: '',\n dateRange: '',\n selectedSource: '',\n customSource: '',\n selectedLanguage: 'english',\n results: [],\n assistantContent: '',\n route: 'links',\n isSearching: false,\n isLoadingMore: false,\n hasSearched: false,\n currentOffset: 0,\n hasMoreResults: true,\n errorMessage: '',\n summaryStates: {},\n hasServerKey: true,\n userApiKey: '',\n apiKeyInput: '',\n // Stream test state\n streamPrompt: '',\n isStreaming: false,\n streamError: '',\n streamAggregates: null,\n streamOpen: true,\n eventsOpen: false,\n // Trace log state\n traceEvents: [],\n traceExpanded: false,\n expandedToolMap: {},\n copyEventsText: 'copy json',\n copyEventsTimeoutId: null,\n \n // Timer state\n timerStartMs: null,\n timerIntervalId: null,\n timerFrozenText: '',\n timerElapsedMs: 0,\n \n // Typewriter effect variables\n hasUserInteracted: false,\n typewriterActive: false,\n typewriterTimeouts: [],\n currentLanguageIndex: 0,\n currentCharIndex: 0,\n isTyping: false,\n lastTypewriterText: '',\n \n // Testing override - if set to a string, will replace typewriter effect\n testOverride: '', // Set to null or '' to disable override\n\n // Sidebar + saved searches UI state (UI-first)\n sidebarOpen: false, // mobile overlay\n sidebarCollapsed: false, // desktop collapse\n sidebarTransitioning: false,\n sidebarAnimMs: 300,\n activeSearchId: null,\n savedSearches: [\n { id: 'init-1', title: 'machine learning' },\n { id: 'init-2', title: 'LLM evaluation benchmarks' },\n { id: 'init-3', title: 'gene therapy delivery vectors' }\n ],\n\n async init() {\n console.log('🔧 Research app initializing...');\n \n // Load API key from localStorage\n const storedApiKey = localStorage.getItem('groq_api_key');\n if (storedApiKey) {\n this.userApiKey = storedApiKey;\n }\n \n // Check if server has API key\n await this.checkServerKey();\n \n // Use test override if defined, otherwise start typewriter effect\n if (this.testOverride && this.testOverride.trim()) {\n this.searchQuery = this.testOverride;\n this.hasUserInteracted = true; // Prevent typewriter from starting\n console.log('🧪 Using test override query:', this.testOverride);\n } else {\n // Start typewriter effect\n this.startTypewriter();\n }\n \n console.log('✅ Research app initialized');\n },\n\n // Sidebar helpers\n toggleSidebar() {\n this.sidebarOpen = !this.sidebarOpen;\n },\n collapseSidebar() {\n // Trigger width animation immediately but delay the expand-control reveal\n this.sidebarTransitioning = true;\n this.sidebarCollapsed = true;\n setTimeout(() => { this.sidebarTransitioning = false; }, this.sidebarAnimMs);\n },\n expandSidebar() {\n // Trigger width animation immediately and hide expand-control during transition\n this.sidebarTransitioning = true;\n this.sidebarCollapsed = false;\n setTimeout(() => { this.sidebarTransitioning = false; }, this.sidebarAnimMs);\n },\n createNewSavedSearch() {\n const title = (this.searchQuery && this.searchQuery.trim()) || 'Untitled search';\n const id = 's-' + Date.now();\n this.savedSearches = [{ id, title }, ...this.savedSearches];\n this.activeSearchId = id;\n },\n selectSavedSearch(id) {\n this.activeSearchId = id;\n const s = this.savedSearches.find(x => x.id === id);\n if (s) {\n this.searchQuery = s.title;\n }\n // Close sidebar on mobile\n this.sidebarOpen = false;\n },\n\n async checkServerKey() {\n try {\n const response = await fetch('/api/check-key');\n const data = await response.json();\n this.hasServerKey = data.hasServerKey;\n } catch (error) {\n console.error('Error checking server key:', error);\n this.hasServerKey = false;\n }\n },\n\n setApiKey() {\n if (this.apiKeyInput.trim()) {\n this.userApiKey = this.apiKeyInput.trim();\n localStorage.setItem('groq_api_key', this.userApiKey);\n this.apiKeyInput = '';\n }\n },\n\n changeApiKey() {\n this.userApiKey = '';\n },\n\n clearApiKey() {\n this.userApiKey = '';\n localStorage.removeItem('groq_api_key');\n },\n\n get maskedApiKey() {\n if (!this.userApiKey) return '';\n const key = this.userApiKey;\n return key.substring(0, 8) + '*'.repeat(Math.max(0, key.length - 12)) + key.substring(key.length - 4);\n },\n\n get displaySourceName() {\n const sourceMap = {\n 'pubmed': 'PubMed',\n 'biorxiv': 'bioRxiv',\n 'arxiv': 'arXiv',\n 'medrxiv': 'medRxiv',\n 'researchgate': 'ResearchGate',\n 'google scholar': 'Google Scholar',\n 'hacker news': 'Hacker News',\n 'reddit': 'Reddit',\n 'news': 'News'\n };\n \n if (this.selectedSource === 'custom' && this.customSource) {\n return this.customSource;\n }\n \n return sourceMap[this.selectedSource] || this.selectedSource;\n },\n\n // Internationalization system\n get i18n() {\n const translations = {\n 'english': {\n // Main headers\n title: '(re)Search',\n subtitle: 'Search the web and academic sources with multilingual summaries',\n \n // API Key section\n apiKey: 'API Key:',\n change: 'Change',\n clear: 'Clear',\n getKeyHere: 'Get your free key here',\n \n // Form labels\n language: 'language',\n researchQuery: 'search query',\n dateRange: 'Choose a date range (optional)',\n source: 'Choose a source (optional)',\n \n // Form placeholders\n enterTopic: 'Enter your research topic...',\n customSource: 'e.g., Nature, Science, Cell, Hacker News...',\n customSourceOption: 'Custom source...',\n chooseDateRange: 'Choose a date range (optional)',\n chooseSource: 'Choose a source (optional)',\n \n // Date range options\n lastMonth: 'Last month',\n last3Months: 'Last 3 months',\n last6Months: 'Last 6 months',\n lastYear: 'Last year',\n allTime: 'All time',\n \n // Buttons\n search: 'Search',\n searching: 'Searching',\n streaming: 'Streaming...',\n searchMore: '(Re)(Re)Search',\n summarize: 'Summarize',\n analyzing: 'Analyzing...',\n openPaper: 'Open',\n \n // Messages\n foundResults: 'Found',\n results: 'results',\n loading: 'Searching',\n loadingMore: 'Loading more results',\n aiAnalyzing: 'AI is analyzing the paper...',\n analysisTime: 'This may take 10-30 seconds depending on paper length',\n noResultsOrError: 'No results found or a server error occurred.',\n \n // Loading message templates\n searchingTemplate: (source, query, dateRange) => {\n let msg = `Searching ${source}`;\n if (query) msg += ` for \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n loadingMoreTemplate: (source, query, dateRange) => {\n let msg = `Loading more results from ${source}`;\n if (query) msg += ` for \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n \n // Error messages\n pleaseAddApiKey: 'please add api key',\n pleaseEnterQuery: 'please enter a search query',\n \n // Summary\n summary: 'Summary:'\n },\n 'french': {\n title: '(re)Recherche',\n subtitle: 'Rechercher dans plusieurs bases de données académiques avec des résumés multilingues',\n \n apiKey: 'Clé API :',\n change: 'Modifier',\n clear: 'Effacer',\n getKeyHere: 'Obtenez votre clé gratuite ici',\n \n language: 'langue',\n researchQuery: 'requête de recherche',\n dateRange: 'Choisir une plage de dates (facultative)',\n source: 'Choisir une source (facultative)',\n \n enterTopic: 'Entrez votre sujet de recherche...',\n customSource: 'ex. Nature, Science, Cell...',\n customSourceOption: 'Source personnalisée...',\n chooseDateRange: 'Choisir une plage de dates (facultative)',\n chooseSource: 'Choisir une source (facultative)',\n \n lastMonth: 'Le mois dernier',\n last3Months: 'Les 3 derniers mois',\n last6Months: 'Les 6 derniers mois',\n lastYear: 'La dernière année',\n allTime: 'Tout le temps',\n \n search: '(re)Rechercher',\n searching: '(re)Recherche',\n streaming: 'Diffusion...',\n searchMore: '(Re)(Re)Recherche',\n summarize: 'Résumer',\n analyzing: 'Analyse...',\n openPaper: 'Ouvrir',\n \n foundResults: 'Trouvé',\n results: 'résultats',\n loading: 'Recherche',\n loadingMore: 'Chargement de plus de résultats',\n aiAnalyzing: 'L\\'IA analyse l\\'article...',\n analysisTime: 'Cela peut prendre 10 à 30 secondes selon la longueur de l\\'article',\n noResultsOrError: 'Aucun résultat trouvé ou une erreur du serveur s\\'est produite.',\n \n // Loading message templates\n searchingTemplate: (source, query, dateRange) => {\n let msg = `Recherche sur ${source}`;\n if (query) msg += ` pour \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n loadingMoreTemplate: (source, query, dateRange) => {\n let msg = `Chargement de plus de résultats de ${source}`;\n if (query) msg += ` pour \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n \n pleaseAddApiKey: 'veuillez ajouter la clé api',\n pleaseEnterQuery: 'veuillez entrer une requête de recherche',\n \n summary: 'Résumé :'\n },\n 'spanish': {\n title: '(re)Buscar',\n subtitle: 'Buscar en múltiples bases de datos académicas con resúmenes multilingües',\n \n apiKey: 'Clave API:',\n change: 'Cambiar',\n clear: 'Limpiar',\n getKeyHere: 'Obtén tu clave gratuita aquí',\n \n language: 'idioma',\n researchQuery: 'consulta de investigación',\n dateRange: 'Elegir un rango de fechas (opcional)',\n source: 'Elegir una fuente (opcional)',\n \n enterTopic: 'Ingresa tu tema de investigación...',\n customSource: 'ej. Nature, Science, Cell...',\n customSourceOption: 'Fuente personalizada...',\n chooseDateRange: 'Elegir un rango de fechas (opcional)',\n chooseSource: 'Elegir una fuente (opcional)',\n \n lastMonth: 'Último mes',\n last3Months: 'Últimos 3 meses',\n last6Months: 'Últimos 6 meses',\n lastYear: 'Último año',\n allTime: 'Todo el tiempo',\n \n search: '(re)Buscar',\n searching: '(re)Buscando',\n streaming: 'Transmitiendo...',\n searchMore: '(Re)(Re)Buscar',\n summarize: 'Resumir',\n analyzing: 'Analizando...',\n openPaper: 'Abrir',\n \n foundResults: 'Encontrado',\n results: 'resultados',\n loading: 'Buscando',\n loadingMore: 'Cargando más resultados',\n aiAnalyzing: 'La IA está analizando el artículo...',\n analysisTime: 'Esto puede tomar 10-30 segundos dependiendo de la longitud del artículo',\n noResultsOrError: 'No se encontraron resultados o ocurrió un error del servidor.',\n \n // Loading message templates\n searchingTemplate: (source, query, dateRange) => {\n let msg = `Buscando en ${source}`;\n if (query) msg += ` por \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n loadingMoreTemplate: (source, query, dateRange) => {\n let msg = `Cargando más resultados de ${source}`;\n if (query) msg += ` por \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n \n pleaseAddApiKey: 'por favor agrega clave api',\n pleaseEnterQuery: 'por favor ingresa una consulta de búsqueda',\n \n summary: 'Resumen:'\n },\n 'german': {\n title: '(re)Suchen',\n subtitle: 'Durchsuchen Sie mehrere akademische Datenbanken mit mehrsprachigen Zusammenfassungen',\n \n apiKey: 'API-Schlüssel:',\n change: 'Ändern',\n clear: 'Löschen',\n getKeyHere: 'Holen Sie sich Ihren kostenlosen Schlüssel hier',\n \n language: 'sprache',\n researchQuery: 'suchanfrage',\n dateRange: 'Datumsbereich wählen (optional)',\n source: 'Quelle wählen (optional)',\n \n enterTopic: 'Geben Sie Ihr Forschungsthema ein...',\n customSource: 'z.B. Nature, Science, Cell...',\n customSourceOption: 'Benutzerdefinierte Quelle...',\n chooseDateRange: 'Datumsbereich wählen (optional)',\n chooseSource: 'Quelle wählen (optional)',\n \n lastMonth: 'Letzter Monat',\n last3Months: 'Letzte 3 Monate',\n last6Months: 'Letzte 6 Monate',\n lastYear: 'Letztes Jahr',\n allTime: 'Alle Zeit',\n \n search: '(re)Suchen',\n searching: '(re)Suchen',\n streaming: 'Streaming...',\n searchMore: '(Re)(Re)Suchen',\n summarize: 'Zusammenfassen',\n analyzing: 'Analysiere...',\n openPaper: 'Öffnen',\n \n foundResults: 'Gefunden',\n results: 'ergebnisse',\n loading: 'Suchen',\n loadingMore: 'Lade weitere Ergebnisse',\n aiAnalyzing: 'KI analysiert den Artikel...',\n analysisTime: 'Dies kann 10-30 Sekunden dauern, abhängig von der Artikellänge',\n noResultsOrError: 'Keine Ergebnisse gefunden oder ein Serverfehler ist aufgetreten.',\n \n // Loading message templates\n searchingTemplate: (source, query, dateRange) => {\n let msg = `Suche in ${source}`;\n if (query) msg += ` nach \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n loadingMoreTemplate: (source, query, dateRange) => {\n let msg = `Lade weitere Ergebnisse von ${source}`;\n if (query) msg += ` nach \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n \n pleaseAddApiKey: 'bitte api-schlüssel hinzufügen',\n pleaseEnterQuery: 'bitte suchanfrage eingeben',\n \n summary: 'Zusammenfassung:'\n },\n 'japanese': {\n title: '(re)検索',\n subtitle: '多言語要約付きの複数の学術データベースを検索',\n \n apiKey: 'APIキー:',\n change: '変更',\n clear: 'クリア',\n getKeyHere: '無料キーを取得する',\n \n language: '言語',\n researchQuery: '研究クエリ',\n dateRange: '日付範囲を選択 (任意)',\n source: 'ソースを選択 (任意)',\n \n enterTopic: '研究トピックを入力...',\n customSource: '例：Nature、Science、Cell...',\n customSourceOption: 'カスタムソース...',\n chooseDateRange: '日付範囲を選択 (任意)',\n chooseSource: 'ソースを選択 (任意)',\n \n lastMonth: '先月',\n last3Months: '過去3ヶ月',\n last6Months: '過去6ヶ月',\n lastYear: '昨年',\n allTime: '全期間',\n \n search: '(re)検索',\n searching: '(re)検索中',\n streaming: 'ストリーミング中...',\n searchMore: '(再)(再)検索',\n summarize: '要約',\n analyzing: '分析中...',\n openPaper: '開く',\n \n foundResults: '見つかった',\n results: '件の結果',\n loading: '検索中',\n loadingMore: 'より多くの結果を読み込み中',\n aiAnalyzing: 'AIが論文を分析中...',\n analysisTime: '論文の長さによって10〜30秒かかることがあります',\n noResultsOrError: '結果が見つからないか、サーバーエラーが発生しました。',\n \n // Loading message templates\n searchingTemplate: (source, query, dateRange) => {\n let msg = `${source}で検索中`;\n if (query) msg += `「${query}」`;\n if (dateRange) msg += `(${dateRange})`;\n return msg + '...';\n },\n loadingMoreTemplate: (source, query, dateRange) => {\n let msg = `${source}からさらに結果を読み込み中`;\n if (query) msg += `「${query}」`;\n if (dateRange) msg += `(${dateRange})`;\n return msg + '...';\n },\n \n pleaseAddApiKey: 'apiキーを追加してください',\n pleaseEnterQuery: '検索クエリを入力してください',\n \n summary: '要約:'\n },\n 'korean': {\n title: '(re)검색',\n subtitle: '다국어 요약과 함께 여러 학술 데이터베이스 검색',\n \n apiKey: 'API 키:',\n change: '변경',\n clear: '지우기',\n getKeyHere: '여기서 무료 키 받기',\n \n language: '언어',\n researchQuery: '연구 쿼리',\n dateRange: '날짜 범위 선택 (선택 사항)',\n source: '소스 선택 (선택 사항)',\n \n enterTopic: '연구 주제를 입력하세요...',\n customSource: '예: Nature, Science, Cell...',\n customSourceOption: '사용자 정의 소스...',\n chooseDateRange: '날짜 범위 선택 (선택 사항)',\n chooseSource: '소스 선택 (선택 사항)',\n \n lastMonth: '지난 달',\n last3Months: '지난 3개월',\n last6Months: '지난 6개월',\n lastYear: '작년',\n allTime: '전체 기간',\n \n search: '(re)검색',\n searching: '(re)검색 중',\n streaming: '스트리밍 중...',\n searchMore: '(재)(재)검색',\n summarize: '요약',\n analyzing: '분석 중...',\n openPaper: '열기',\n \n foundResults: '발견된',\n results: '개 결과',\n loading: '검색 중',\n loadingMore: '더 많은 결과 로딩 중',\n aiAnalyzing: 'AI가 논문을 분석 중...',\n analysisTime: '논문 길이에 따라 10-30초가 걸릴 수 있습니다',\n noResultsOrError: '결과가 없거나 서버 오류가 발생했습니다.',\n \n // Loading message templates\n searchingTemplate: (source, query, dateRange) => {\n let msg = `${source}에서 검색 중`;\n if (query) msg += ` \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n loadingMoreTemplate: (source, query, dateRange) => {\n let msg = `${source}에서 더 많은 결과 로딩 중`;\n if (query) msg += ` \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n \n pleaseAddApiKey: 'api 키를 추가해 주세요',\n pleaseEnterQuery: '검색 쿼리를 입력해 주세요',\n \n summary: '요약:'\n },\n 'chinese': {\n title: '(re)搜索',\n subtitle: '搜索多个学术数据库并提供多语言摘要',\n \n apiKey: 'API密钥:',\n change: '更改',\n clear: '清除',\n getKeyHere: '在此获取免费密钥',\n \n language: '语言',\n researchQuery: '研究查询',\n dateRange: '选择日期范围 (可选)',\n source: '选择来源 (可选)',\n \n enterTopic: '输入您的研究主题...',\n customSource: '例如：Nature、Science、Cell...',\n customSourceOption: '自定义来源...',\n chooseDateRange: '选择日期范围 (可选)',\n chooseSource: '选择来源 (可选)',\n \n lastMonth: '上个月',\n last3Months: '最近3个月',\n last6Months: '最近6个月',\n lastYear: '去年',\n allTime: '全部时间',\n \n search: '(re)搜索',\n searching: '(re)搜索中',\n streaming: '流式传输中...',\n searchMore: '(再)(再)搜索',\n summarize: '总结',\n analyzing: '分析中...',\n openPaper: '打开',\n \n foundResults: '找到',\n results: '个结果',\n loading: '搜索中',\n loadingMore: '加载更多结果',\n aiAnalyzing: 'AI正在分析论文...',\n analysisTime: '根据论文长度，这可能需要10-30秒',\n noResultsOrError: '未找到结果或发生服务器错误。',\n \n // Loading message templates\n searchingTemplate: (source, query, dateRange) => {\n let msg = `正在${source}中搜索`;\n if (query) msg += `“${query}”`;\n if (dateRange) msg += `(${dateRange})`;\n return msg + '...';\n },\n loadingMoreTemplate: (source, query, dateRange) => {\n let msg = `正在从${source}加载更多结果`;\n if (query) msg += `“${query}”`;\n if (dateRange) msg += `(${dateRange})`;\n return msg + '...';\n },\n \n pleaseAddApiKey: '请添加api密钥',\n pleaseEnterQuery: '请输入搜索查询',\n \n summary: '摘要:'\n },\n 'portuguese': {\n title: '(re)Pesquisar',\n subtitle: 'Pesquise em múltiplas bases de dados acadêmicas com resumos multilíngues',\n \n apiKey: 'Chave API:',\n change: 'Alterar',\n clear: 'Limpar',\n getKeyHere: 'Obtenha sua chave gratuita aqui',\n \n language: 'idioma',\n researchQuery: 'consulta de pesquisa',\n dateRange: 'Escolher intervalo de datas (opcional)',\n source: 'Escolher fonte (opcional)',\n \n enterTopic: 'Digite seu tópico de pesquisa...',\n customSource: 'ex. Nature, Science, Cell...',\n customSourceOption: 'Fonte personalizada...',\n chooseDateRange: 'Escolher intervalo de datas (opcional)',\n chooseSource: 'Escolher fonte (opcional)',\n \n lastMonth: 'Último mês',\n last3Months: 'Últimos 3 meses',\n last6Months: 'Últimos 6 meses',\n lastYear: 'Último ano',\n allTime: 'Todo o tempo',\n \n search: '(re)Pesquisar',\n searching: '(re)Pesquisando',\n streaming: 'Transmitindo...',\n searchMore: '(Re)(Re)Pesquisar',\n summarize: 'Resumir',\n analyzing: 'Analisando...',\n openPaper: 'Abrir',\n \n foundResults: 'Encontrado',\n results: 'resultados',\n loading: 'Pesquisando',\n loadingMore: 'Carregando mais resultados',\n aiAnalyzing: 'IA está analisando o artigo...',\n analysisTime: 'Isso pode levar 10-30 segundos dependendo do comprimento do artigo',\n noResultsOrError: 'Nenhum resultado encontrado ou ocorreu um erro no servidor.',\n \n // Loading message templates\n searchingTemplate: (source, query, dateRange) => {\n let msg = `Pesquisando no ${source}`;\n if (query) msg += ` por \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n loadingMoreTemplate: (source, query, dateRange) => {\n let msg = `Carregando mais resultados do ${source}`;\n if (query) msg += ` por \"${query}\"`;\n if (dateRange) msg += ` (${dateRange})`;\n return msg + '...';\n },\n \n pleaseAddApiKey: 'por favor adicione chave api',\n pleaseEnterQuery: 'por favor digite uma consulta de pesquisa',\n \n summary: 'Resumo:'\n }\n };\n \n return translations[this.selectedLanguage] || translations['english'];\n },\n\n getSummaryTitle() {\n return this.i18n.summary;\n },\n\n getSearchButtonText() {\n return `${this.i18n.search} ${this.displaySourceName}`;\n },\n\n getSearchingButtonText() {\n return `${this.i18n.searching} ${this.displaySourceName}...`;\n },\n\n get loadingMessage() {\n // Use the last typewriter text if available, otherwise fall back to current searchQuery\n const queryText = this.lastTypewriterText || this.searchQuery;\n const dateRangeText = this.dateRange ? this.getDateRangeText() : '';\n \n // Get the actual source being searched (matches backend logic)\n const actualSource = this.getActualSearchSourceName();\n \n // Use language-specific template\n const template = this.isLoadingMore ? this.i18n.loadingMoreTemplate : this.i18n.searchingTemplate;\n return template(actualSource, queryText, dateRangeText);\n },\n\n // Timer computed display\n get displayTimerText() {\n if (this.timerFrozenText) return this.timerFrozenText;\n if (!this.timerStartMs) return '';\n const elapsed = this.timerElapsedMs / 1000;\n const seconds = Math.floor(elapsed);\n const hundredths = Math.floor((elapsed - seconds) * 100);\n const s = String(seconds).padStart(2, '0');\n const h = String(hundredths).padStart(2, '0');\n return `${s}.${h}s`;\n },\n\n getDateRangeText() {\n const dateRangeMap = {\n '1month': this.i18n.lastMonth.toLowerCase(),\n '3months': this.i18n.last3Months.toLowerCase(),\n '6months': this.i18n.last6Months.toLowerCase(),\n '1year': this.i18n.lastYear.toLowerCase()\n };\n return dateRangeMap[this.dateRange] || '';\n },\n\n getActualSearchSource() {\n if (this.selectedSource === 'custom' && this.customSource) {\n return `on ${this.customSource}`;\n } else if (this.selectedSource === 'pubmed') {\n return 'PubMed';\n } else if (this.selectedSource === 'biorxiv') {\n return 'bioRxiv';\n } else if (this.selectedSource === 'arxiv') {\n return 'arXiv';\n } else if (this.selectedSource === 'medrxiv') {\n return 'medRxiv';\n } else if (this.selectedSource === 'researchgate') {\n return 'ResearchGate';\n } else if (this.selectedSource === 'google scholar') {\n return 'Google Scholar';\n } else {\n return '';\n }\n },\n\n getActualSearchSourceName() {\n if (this.selectedSource === 'custom' && this.customSource) {\n return this.customSource;\n } else if (this.selectedSource === 'pubmed') {\n return 'PubMed';\n } else if (this.selectedSource === 'biorxiv') {\n return 'bioRxiv';\n } else if (this.selectedSource === 'arxiv') {\n return 'arXiv';\n } else if (this.selectedSource === 'medrxiv') {\n return 'medRxiv';\n } else if (this.selectedSource === 'researchgate') {\n return 'ResearchGate';\n } else if (this.selectedSource === 'google scholar') {\n return 'Google Scholar';\n } else if (this.selectedSource === 'hacker news') {\n return 'Hacker News';\n } else if (this.selectedSource === 'reddit') {\n return 'Reddit';\n } else if (this.selectedSource === 'news') {\n return 'News';\n } else {\n return '';\n }\n },\n\n handleSourceChange() {\n // Clear custom source when switching away from custom\n if (this.selectedSource !== 'custom') {\n this.customSource = '';\n }\n },\n\n startTypewriter() {\n if (this.hasUserInteracted) return;\n \n this.typewriterActive = true;\n const translations = [\n 'machine learning', // English\n 'apprentissage automatique', // French\n 'maschinelles lernen', // German\n 'aprendizaje automático', // Spanish\n '機械学習', // Japanese\n '기계 학습', // Korean\n '机器学习', // Chinese\n 'aprendizado de máquina' // Portuguese\n ];\n \n this.typewriterCycle(translations);\n },\n\n typewriterCycle(translations) {\n if (this.hasUserInteracted || !this.typewriterActive) return;\n \n // Clear any existing timeouts first\n this.clearAllTimeouts();\n \n const currentText = translations[this.currentLanguageIndex];\n this.typeText(currentText, () => {\n // Wait 2 seconds after completing the word, then move to next language\n const timeout = setTimeout(() => {\n if (this.hasUserInteracted || !this.typewriterActive) return;\n \n this.currentLanguageIndex = (this.currentLanguageIndex + 1) % translations.length;\n this.typewriterCycle(translations);\n }, 2000); // Reduced from 4000ms to 2000ms\n this.typewriterTimeouts.push(timeout);\n });\n },\n\n typeText(text, onComplete) {\n if (this.hasUserInteracted || !this.typewriterActive) return;\n \n // Clear any existing timeouts and reset\n this.clearAllTimeouts();\n this.isTyping = true;\n this.currentCharIndex = 0;\n this.searchQuery = '';\n \n const typeNextChar = () => {\n if (this.hasUserInteracted || !this.typewriterActive) {\n this.isTyping = false;\n return;\n }\n \n if (this.currentCharIndex < text.length) {\n this.searchQuery = text.substring(0, this.currentCharIndex + 1);\n this.currentCharIndex++;\n const timeout = setTimeout(typeNextChar, 80); // Slightly faster typing\n this.typewriterTimeouts.push(timeout);\n } else {\n this.isTyping = false;\n if (onComplete) onComplete();\n }\n };\n \n typeNextChar();\n },\n\n clearAllTimeouts() {\n this.typewriterTimeouts.forEach(timeout => clearTimeout(timeout));\n this.typewriterTimeouts = [];\n },\n\n stopTypewriter() {\n this.hasUserInteracted = true;\n this.typewriterActive = false;\n this.clearAllTimeouts();\n this.isTyping = false;\n // Save the current typewriter text for search messages\n this.lastTypewriterText = this.searchQuery;\n },\n\n // Timer controls\n startTimer() {\n this.stopTimer();\n this.timerFrozenText = '';\n this.timerStartMs = Date.now();\n this.timerElapsedMs = 0;\n // Update every 30ms for smooth hundredths\n this.timerIntervalId = setInterval(() => {\n this.timerElapsedMs = Date.now() - this.timerStartMs;\n }, 30);\n },\n stopTimer() {\n if (this.timerIntervalId) {\n clearInterval(this.timerIntervalId);\n this.timerIntervalId = null;\n }\n if (this.timerStartMs) {\n const elapsed = (this.timerElapsedMs || (Date.now() - this.timerStartMs)) / 1000;\n const seconds = Math.floor(elapsed);\n const hundredths = Math.floor((elapsed - seconds) * 100);\n const s = String(seconds).padStart(2, '0');\n const h = String(hundredths).padStart(2, '0');\n this.timerFrozenText = `${s}.${h}s...done`;\n }\n this.timerStartMs = null;\n this.timerElapsedMs = 0;\n },\n\n handleSearchClick() {\n // Clear previous results immediately when user initiates a new search\n this.results = [];\n this.assistantContent = '';\n // Stop typewriter immediately to freeze current text\n this.stopTypewriter();\n \n // Clear any existing error message\n this.errorMessage = '';\n \n // Check if button should be disabled and show appropriate error\n if (!this.searchQuery.trim()) {\n this.errorMessage = this.i18n.pleaseEnterQuery;\n return;\n }\n \n if (!this.hasServerKey && !this.userApiKey) {\n this.errorMessage = this.i18n.pleaseAddApiKey;\n return;\n }\n \n // If all checks pass, proceed with search\n this.searchResearch();\n },\n \n async searchResearch(loadMore = false) {\n console.log('🔍 Search triggered with query:', this.searchQuery, 'loadMore:', loadMore);\n if (!this.searchQuery.trim()) return;\n \n // Stop typewriter immediately and freeze the search text\n this.stopTypewriter();\n \n // Check if API key is available\n if (!this.hasServerKey && !this.userApiKey) {\n this.errorMessage = 'Please provide a Groq API key to search.';\n return;\n }\n \n if (loadMore) {\n this.isLoadingMore = true;\n } else {\n this.isSearching = true;\n this.errorMessage = '';\n this.results = [];\n this.assistantContent = '';\n this.currentOffset = 0;\n this.hasMoreResults = true;\n this.hasSearched = false;\n\n // Reset thinking trace for a brand new search\n this.traceEvents = [];\n this.expandedToolMap = {};\n this.traceExpanded = false;\n this.streamAggregates = { events: [], collapsed: { reasoning: '', content: '', tools: [] } };\n this.copyEventsText = 'copy json';\n if (this.copyEventsTimeoutId) {\n clearTimeout(this.copyEventsTimeoutId);\n this.copyEventsTimeoutId = null;\n }\n this.queueScrollToBottom();\n }\n\n // Start the elapsed timer\n this.startTimer();\n\n try {\n const response = await fetch('/api/search', {\n method: 'POST',\n headers: {\n 'Content-Type': 'application/json',\n },\n body: JSON.stringify({\n query: this.searchQuery,\n dateRange: this.dateRange,\n source: this.selectedSource === 'custom' ? this.customSource : this.selectedSource,\n language: this.selectedLanguage,\n userApiKey: this.userApiKey,\n offset: this.currentOffset\n })\n });\n\n if (!response.ok) throw new Error('Search failed: ' + response.statusText);\n \n const data = await response.json();\n \n if (data.error) {\n this.errorMessage = data.error;\n } else {\n const newResults = data.results || [];\n // Save assistant content and route for fallback/display\n this.assistantContent = (data.assistantContent && typeof data.assistantContent === 'string') ? data.assistantContent : '';\n this.route = data.route || 'links';\n if (loadMore) {\n this.results = [...this.results, ...newResults];\n } else {\n this.results = newResults;\n }\n \n // Update pagination state\n this.currentOffset += newResults.length;\n this.hasMoreResults = newResults.length >= 5; // Assume no more results if we got less than 5\n }\n } catch (error) {\n this.errorMessage = 'Error searching research databases: ' + error.message;\n } finally {\n this.isSearching = false;\n this.isLoadingMore = false;\n // Stop and freeze the timer\n this.stopTimer();\n this.hasSearched = true;\n }\n },\n\n async summarizePaper(url, title, index) {\n console.log('📄 Summary triggered for:', title, 'at index:', index);\n // Initialize summary state for this paper\n if (!this.summaryStates[index]) {\n this.summaryStates[index] = {};\n }\n \n this.summaryStates[index].loading = true;\n this.summaryStates[index].summary = '';\n\n try {\n const response = await fetch('/api/search', {\n method: 'POST',\n headers: {\n 'Content-Type': 'application/json',\n },\n body: JSON.stringify({\n url,\n title,\n language: this.selectedLanguage,\n userApiKey: this.userApiKey\n })\n });\n\n if (!response.ok) throw new Error('Summary failed: ' + response.statusText);\n \n const data = await response.json();\n \n if (data.error) {\n this.summaryStates[index].summary = 'Error: ' + data.error;\n } else {\n this.summaryStates[index].summary = data.summary;\n }\n } catch (error) {\n this.summaryStates[index].summary = 'Error analyzing paper: ' + error.message;\n } finally {\n this.summaryStates[index].loading = false;\n }\n },\n async runStreamTest() {\n // Clear previous results immediately when user initiates streaming\n this.results = [];\n this.streamError = '';\n this.isStreaming = true;\n this.hasSearched = false;\n this.streamAggregates = { events: [], collapsed: { reasoning: '', content: '', tools: [] } };\n this.traceEvents = [];\n this.expandedToolMap = {};\n this.assistantContent = '';\n this.queueScrollToBottom();\n // Start elapsed timer for streaming\n this.startTimer();\n try {\n const response = await fetch('/api/chatStreamSSE', {\n method: 'POST',\n headers: { 'Content-Type': 'application/json' },\n body: JSON.stringify({\n userApiKey: this.userApiKey,\n // Send structured search params; server will build the prompt for streaming\n query: this.searchQuery && this.searchQuery.trim() ? this.searchQuery.trim() : undefined,\n dateRange: this.dateRange,\n source: this.selectedSource === 'custom' ? this.customSource : this.selectedSource,\n language: this.selectedLanguage,\n offset: this.currentOffset\n })\n });\n\n if (!response.ok || !response.body) {\n let errText = '';\n try { errText = await response.text(); } catch (_) {}\n throw new Error(errText || response.statusText);\n }\n\n const decoder = new TextDecoder();\n const reader = response.body.getReader();\n let buffer = '';\n\n const pushOrAppend = (kind, text) => {\n if (!text) return;\n const last = this.traceEvents[this.traceEvents.length - 1];\n if (last && last.kind === kind) {\n last.text += text;\n } else {\n this.traceEvents.push({ kind, text });\n }\n this.queueScrollToBottom();\n };\n const pushTool = (tool) => {\n const name = tool.name || tool.tool || tool.function?.name || tool.id || 'tool';\n this.traceEvents.push({ kind: 'tool', name, id: tool.id, data: tool });\n this.queueScrollToBottom();\n };\n\n while (true) {\n const { done, value } = await reader.read();\n if (done) break;\n buffer += decoder.decode(value, { stream: true });\n const events = buffer.split('\\n\\n');\n buffer = events.pop() || '';\n for (const evt of events) {\n if (!evt.trim()) continue;\n const lines = evt.split('\\n');\n let currentEvent = 'message';\n for (const line of lines) {\n if (line.startsWith('event: ')) {\n currentEvent = line.slice(7).trim();\n continue;\n }\n if (!line.startsWith('data: ')) continue;\n const data = line.slice(6);\n if (data === '[DONE]') {\n break;\n }\n let parsed = null;\n try { parsed = JSON.parse(data); } catch (_) { parsed = null; }\n\n if (currentEvent === 'results') {\n if (parsed && Array.isArray(parsed.results)) {\n this.results = parsed.results;\n this.currentOffset = this.results.length;\n this.hasMoreResults = false;\n }\n continue;\n }\n\n // Fallback: accept arrays or `{ results: [...] }` without explicit event name\n if (parsed && Array.isArray(parsed.results)) {\n this.results = parsed.results;\n this.currentOffset = this.results.length;\n this.hasMoreResults = false;\n continue;\n }\n if (Array.isArray(parsed)) {\n this.results = parsed;\n this.currentOffset = this.results.length;\n this.hasMoreResults = false;\n continue;\n }\n\n if (!parsed) continue;\n const choice = Array.isArray(parsed.choices) ? parsed.choices[0] : null;\n const delta = choice?.delta || {};\n if (delta.reasoning && delta.channel) {\n pushOrAppend('reasoning', delta.reasoning);\n this.streamAggregates.collapsed.reasoning += delta.reasoning;\n }\n if (delta.content) {\n pushOrAppend('content', delta.content);\n this.streamAggregates.collapsed.content += delta.content;\n }\n if (Array.isArray(delta.executed_tools)) {\n for (const t of delta.executed_tools) {\n pushTool(t);\n this.streamAggregates.collapsed.tools.push({ kind: 'executed', ...t, timestamp: Date.now() });\n }\n }\n if (Array.isArray(delta.tool_calls)) {\n for (const t of delta.tool_calls) {\n pushTool(t);\n this.streamAggregates.collapsed.tools.push({ kind: 'call', ...t, timestamp: Date.now() });\n }\n }\n }\n }\n }\n\n // Flush any remaining buffered event at EOF (e.g., final 'results')\n try {\n const leftover = (buffer || '').trim();\n if (leftover) {\n const events = (leftover + '\\n\\n').split('\\n\\n');\n for (const evt of events) {\n if (!evt || !evt.trim()) continue;\n const lines = evt.split('\\n');\n let currentEvent = 'message';\n for (const line of lines) {\n if (line.startsWith('event: ')) {\n currentEvent = line.slice(7).trim();\n continue;\n }\n if (!line.startsWith('data: ')) continue;\n const data = line.slice(6);\n if (data === '[DONE]') continue;\n let parsed = null;\n try { parsed = JSON.parse(data); } catch (_) { parsed = null; }\n if (currentEvent === 'results') {\n if (parsed && Array.isArray(parsed.results)) {\n this.results = parsed.results;\n this.currentOffset = this.results.length;\n this.hasMoreResults = false;\n }\n continue;\n }\n if (parsed && Array.isArray(parsed.results)) {\n this.results = parsed.results;\n this.currentOffset = this.results.length;\n this.hasMoreResults = false;\n continue;\n }\n if (Array.isArray(parsed)) {\n this.results = parsed;\n this.currentOffset = this.results.length;\n this.hasMoreResults = false;\n continue;\n }\n if (!parsed) continue;\n const choice = Array.isArray(parsed.choices) ? parsed.choices[0] : null;\n const delta = choice?.delta || {};\n if (delta.reasoning) {\n pushOrAppend('reasoning', delta.reasoning);\n this.streamAggregates.collapsed.reasoning += delta.reasoning;\n }\n if (delta.content) {\n pushOrAppend('content', delta.content);\n this.streamAggregates.collapsed.content += delta.content;\n }\n if (Array.isArray(delta.executed_tools)) {\n for (const t of delta.executed_tools) {\n pushTool(t);\n this.streamAggregates.collapsed.tools.push({ kind: 'executed', ...t, timestamp: Date.now() });\n }\n }\n if (Array.isArray(delta.tool_calls)) {\n for (const t of delta.tool_calls) {\n pushTool(t);\n this.streamAggregates.collapsed.tools.push({ kind: 'call', ...t, timestamp: Date.now() });\n }\n }\n }\n }\n }\n } catch (_) { /* no-op */ }\n\n this.hasMoreResults = false;\n } catch (e) {\n this.streamError = e.message || String(e);\n } finally {\n this.isStreaming = false;\n this.hasSearched = true;\n // Stop and freeze the timer at end of streaming\n this.stopTimer();\n // Fallback: if no results and we have assistant content from stream, show it\n try {\n if ((this.results || []).length === 0) {\n const c = this.streamAggregates?.collapsed?.content || '';\n if (c && c.trim()) this.assistantContent = c;\n // If assistant content is actually JSON with results, adopt it\n const maybe = (this.assistantContent || '').trim();\n if (maybe) {\n const stripped = maybe.replace(/^```json\\s*/i, '').replace(/```\\s*$/i, '');\n try {\n const parsed = JSON.parse(stripped);\n if (parsed && Array.isArray(parsed.results) && parsed.results.length > 0) {\n this.results = parsed.results;\n this.currentOffset = this.results.length;\n this.hasMoreResults = false;\n this.assistantContent = '';\n }\n } catch (_) { /* ignore */ }\n }\n }\n } catch (_) {}\n }\n },\n // Pretty-print assistant content if it's valid JSON, else return as-is\n formatAssistantContent() {\n const text = this.assistantContent || '';\n if (!text) return '';\n const trimmed = text.trim();\n if ((trimmed.startsWith('{') || trimmed.startsWith('['))) {\n try {\n const obj = JSON.parse(trimmed);\n return JSON.stringify(obj, null, 2);\n } catch (_) {\n return text;\n }\n }\n return text;\n },\n // Render assistant content as HTML using marked + DOMPurify (safe markdown rendering)\n renderAssistantContent() {\n const raw = this.formatAssistantContent();\n try {\n // If it is JSON (from formatAssistantContent), keep as a code block\n if (raw && (raw.trim().startsWith('{') || raw.trim().startsWith('['))) {\n const escaped = raw.replace(/&/g, '&').replace(//g, '>');\n return `

${escaped}

`;\n }\n const html = (window.marked ? window.marked.parse(raw) : raw);\n return (window.DOMPurify ? window.DOMPurify.sanitize(html) : html);\n } catch (_) {\n return raw;\n }\n },\n // Trace helpers\n computeTraceMaxHeight() {\n return this.traceExpanded ? '' : 'max-height: 100px';\n },\n toggleTool(idx) {\n this.expandedToolMap[idx] = !this.expandedToolMap[idx];\n this.queueScrollToBottom();\n },\n prettify(obj) {\n try { return JSON.stringify(obj, null, 2); } catch (_) { return String(obj); }\n },\n async copyEventsJson() {\n try {\n const json = JSON.stringify(this.traceEvents || [], null, 2);\n await navigator.clipboard.writeText(json);\n this.copyEventsText = 'copied';\n if (this.copyEventsTimeoutId) clearTimeout(this.copyEventsTimeoutId);\n this.copyEventsTimeoutId = setTimeout(() => {\n this.copyEventsText = 'copy json';\n this.copyEventsTimeoutId = null;\n }, 3000);\n } catch (_) {\n // no-op\n }\n },\n // Smoothly ensure the trace scroller is at bottom after DOM updates\n queueScrollToBottom() {\n try {\n const el = this.$refs && this.$refs.traceContainer;\n if (!el) return;\n const doScroll = () => { el.scrollTop = el.scrollHeight; };\n if (typeof this.$nextTick === 'function') {\n this.$nextTick(doScroll);\n } else {\n requestAnimationFrame(doScroll);\n }\n } catch (_) {\n // no-op\n }\n }\n };\n}\n\n\n",[],[186,275],"76b2e89a-7618-11f0-8008-0224a6c84d84","lorraineApp.js","76b2e976-7618-11f0-8008-0224a6c84d84",13,["D",1754850789104],"export default function lorraineApp() {\n return {\n // Shared/global\n hasServerKey: true,\n userApiKey: '',\n apiKeyInput: '',\n selectedLanguage: 'english',\n // Job naming\n jobName: '',\n jobInput: '',\n\n // UI state (reuse layout from research app)\n sidebarOpen: false,\n sidebarCollapsed: false,\n sidebarTransitioning: false,\n sidebarAnimMs: 300,\n\n // Lorraine state\n questions: [],\n currentIndex: 0,\n answers: {}, // { [id]: { answer_text, transcript_text, created_at } }\n showNeutral: false,\n answeredMap: {}, // { [id]: true }\n loadingQuestions: false,\n errorMessage: '',\n isTranscribing: false,\n recorder: null,\n mediaStream: null,\n chunks: [],\n answerBuffer: '',\n // Import/Export & Storage\n importJsonOpen: false,\n importJsonBuffer: '',\n importError: '',\n importSuccess: '',\n\n // Inline JSON editor for the full questions set\n showJsonEditor: false,\n jsonEditBuffer: '',\n jsonEditError: '',\n jsonEditSuccess: '',\n\n // Inline editor for user responses (answers)\n showAnswersEditor: false,\n answersEditBuffer: '',\n answersEditError: '',\n answersEditSuccess: '',\n\n // Corpus → Questions tool\n showCorpusTool: false,\n corpusInput: '',\n corpusMaxQuestions: 4,\n corpusTags: '',\n corpusGenerating: false,\n corpusStatus: '',\n corpusError: '',\n\n // Minimal compatibility fields for legacy research UI fragments in index.html\n isSearching: false,\n isLoadingMore: false,\n isStreaming: false,\n hasSearched: false,\n results: [],\n assistantContent: '',\n summaryStates: {},\n copyEventsText: 'copy json',\n traceEvents: [],\n expandedToolMap: {},\n traceExpanded: false,\n streamAggregates: { events: [], collapsed: { reasoning: '', content: '', tools: [] } },\n currentOffset: 0,\n hasMoreResults: false,\n // Search/typewriter/timer placeholders (not actively used in Lorraine mode)\n selectedSource: '',\n customSource: '',\n searchQuery: '',\n dateRange: '',\n lastTypewriterText: '',\n typewriterActive: false,\n hasUserInteracted: true,\n typewriterTimeouts: [],\n isTyping: false,\n currentLanguageIndex: 0,\n currentCharIndex: 0,\n timerStartMs: null,\n timerElapsedMs: 0,\n timerFrozenText: '',\n timerIntervalId: null,\n\n async init() {\n // Load API key from localStorage\n const storedApiKey = localStorage.getItem('groq_api_key');\n if (storedApiKey) this.userApiKey = storedApiKey;\n // Load job name from localStorage\n try {\n const storedJob = localStorage.getItem('lorraine_job_name');\n if (storedJob) { this.jobName = storedJob; this.jobInput = storedJob; }\n } catch (_) {}\n await this.checkServerKey();\n await this.loadQuestions();\n // sync buffer when index changes\n try { this.$watch('currentIndex', () => { this.answerBuffer = this.currentAnswer; }); } catch (_) {}\n this.answerBuffer = this.currentAnswer;\n // Attempt to load saved answers (session first, then local)\n try { this.loadFromSessionStorage(); } catch (_) {}\n if (Object.keys(this.answers || {}).length === 0) {\n try { this.loadFromLocalStorage(); } catch (_) {}\n }\n // build answered map and move to a random question\n this.rebuildAnsweredMap();\n this.jumpToRandomUnanswered();\n },\n\n async checkServerKey() {\n try {\n const r = await fetch('/api/check-key');\n const j = await r.json();\n this.hasServerKey = j.hasServerKey;\n } catch (_) {\n this.hasServerKey = false;\n }\n },\n\n setApiKey() {\n if (this.apiKeyInput.trim()) {\n this.userApiKey = this.apiKeyInput.trim();\n localStorage.setItem('groq_api_key', this.userApiKey);\n this.apiKeyInput = '';\n }\n },\n setJobName() {\n const trimmed = (this.jobInput || '').trim();\n this.jobName = trimmed;\n try { localStorage.setItem('lorraine_job_name', this.jobName || ''); } catch (_) {}\n // If the inline responses editor is visible, refresh its buffer to reflect the new job\n try {\n if (this.showAnswersEditor) {\n this.answersEditBuffer = this.getExportJsonPretty();\n if (typeof this.$nextTick === 'function') {\n this.$nextTick(() => this.autoSizeAnswersTa());\n } else {\n setTimeout(() => this.autoSizeAnswersTa(), 0);\n }\n }\n } catch (_) {}\n // Persist updated export with job name\n try { this.saveToSessionStorage(); } catch (_) {}\n try { this.saveToLocalStorage(); } catch (_) {}\n },\n changeApiKey() { this.userApiKey = ''; },\n clearApiKey() { this.userApiKey = ''; localStorage.removeItem('groq_api_key'); },\n get maskedApiKey() {\n if (!this.userApiKey) return '';\n const k = this.userApiKey;\n return k.substring(0, 8) + '*'.repeat(Math.max(0, k.length - 12)) + k.substring(k.length - 4);\n },\n\n // Load questions JSON\n async loadQuestions() {\n this.loadingQuestions = true;\n this.errorMessage = '';\n try {\n // Local override: use saved custom questions if present\n try {\n const override = localStorage.getItem('lorraine_custom_questions_v1');\n if (override) {\n const a = JSON.parse(override);\n if (Array.isArray(a) && a.length > 0) {\n this.questions = a;\n if (!this.questions.length) this.errorMessage = 'No questions available.';\n return;\n }\n }\n } catch (_) {}\n // Support selecting groups via URL: ?groups=work,friends\n const url = new URL(window.location.href);\n const groupParam = (url.searchParams.get('groups') || '').trim();\n let arr = [];\n // Try selected groups (simple schema) first\n if (groupParam) {\n try {\n const rGroups = await fetch('/data/questions/simple/by-groups.json?groups=' + encodeURIComponent(groupParam));\n if (rGroups.ok) {\n const a = await rGroups.json();\n if (Array.isArray(a)) arr = a;\n }\n } catch (_) {}\n }\n // Try aggregated next\n if (!Array.isArray(arr) || arr.length === 0) {\n try {\n const rAll = await fetch('/data/questions/simple.json');\n if (rAll.ok) {\n const a = await rAll.json();\n if (Array.isArray(a)) arr = a;\n }\n } catch (_) {}\n }\n // Fallback to base file\n if (!Array.isArray(arr) || arr.length === 0) {\n const r = await fetch('/data/questions.json');\n if (!r.ok) throw new Error('failed to load questions');\n const a = await r.json();\n if (Array.isArray(a)) arr = a;\n }\n if (Array.isArray(arr)) this.questions = arr;\n if (!this.questions.length) this.errorMessage = 'No questions available.';\n } catch (e) {\n this.errorMessage = 'Error loading questions: ' + (e?.message || String(e));\n } finally {\n this.loadingQuestions = false;\n }\n },\n\n // Toggle and manage inline JSON editor\n toggleJsonEditor() {\n this.showJsonEditor = !this.showJsonEditor;\n this.jsonEditError = '';\n this.jsonEditSuccess = '';\n if (this.showJsonEditor) {\n try { this.jsonEditBuffer = JSON.stringify(this.questions || [], null, 2); } catch (_) { this.jsonEditBuffer = '[]'; }\n }\n },\n saveJsonEdits() {\n this.jsonEditError = '';\n this.jsonEditSuccess = '';\n let parsed = null;\n try { parsed = JSON.parse(this.jsonEditBuffer || ''); } catch (e) {\n this.jsonEditError = 'invalid JSON';\n return;\n }\n if (!Array.isArray(parsed)) {\n this.jsonEditError = 'root must be an array';\n return;\n }\n for (const [i, q] of parsed.entries()) {\n if (!q || typeof q !== 'object') { this.jsonEditError = `item ${i+1} is not an object`; return; }\n if (!q.id || !q.label || !q.instruction || !Array.isArray(q.tags)) { this.jsonEditError = `item ${i+1} missing id/label/instruction/tags`; return; }\n }\n this.questions = parsed;\n try { localStorage.setItem('lorraine_custom_questions_v1', JSON.stringify(parsed)); } catch (_) {}\n this.rebuildAnsweredMap();\n this.currentIndex = 0;\n this.answerBuffer = this.currentAnswer;\n this.jsonEditSuccess = 'saved';\n try { this.jsonEditBuffer = JSON.stringify(this.questions || [], null, 2); } catch (_) {}\n },\n\n // Toggle and save the answers JSON editor\n toggleAnswersEditor() {\n this.showAnswersEditor = !this.showAnswersEditor;\n this.answersEditError = '';\n this.answersEditSuccess = '';\n if (this.showAnswersEditor) {\n try { this.answersEditBuffer = this.getExportJsonPretty(); } catch (_) { this.answersEditBuffer = '{\"answers\": []}'; }\n // Ensure autosize runs after the textarea becomes visible\n try {\n if (typeof this.$nextTick === 'function') {\n this.$nextTick(() => { this.autoSizeAnswersTa(); });\n } else {\n setTimeout(() => this.autoSizeAnswersTa(), 0);\n }\n } catch (_) {}\n }\n },\n saveAnswersEdits() {\n this.answersEditError = '';\n this.answersEditSuccess = '';\n let parsed = null;\n try { parsed = JSON.parse(this.answersEditBuffer || ''); } catch (e) {\n this.answersEditError = 'invalid JSON';\n return;\n }\n let list = [];\n if (parsed && Array.isArray(parsed)) list = parsed;\n else if (parsed && Array.isArray(parsed.answers)) list = parsed.answers;\n if (!Array.isArray(list)) { this.answersEditError = 'root must be an array or { answers: [...] }'; return; }\n\n const next = { ...this.answers };\n for (const item of list) {\n const qid = item?.question_id || item?.id;\n const text = item?.raw_response || item?.answer_text || item?.answer || '';\n if (!qid || !String(qid).trim()) continue;\n const existed = next[qid] || {};\n next[qid] = {\n ...existed,\n answer_text: text,\n transcript_text: item?.transcript_text || existed.transcript_text || null,\n created_at: existed.created_at || item?.created_at || new Date().toISOString()\n };\n }\n this.answers = next;\n this.rebuildAnsweredMap();\n // persist\n try { this.saveToSessionStorage(); } catch (_) {}\n try { this.saveToLocalStorage(); } catch (_) {}\n this.answersEditSuccess = 'saved';\n try { this.answersEditBuffer = this.getExportJsonPretty(); } catch (_) {}\n // Re-autosize after updating content\n try { this.autoSizeAnswersTa(); } catch (_) {}\n },\n\n // Autosize responses textarea to content height\n autoSizeAnswersTa() {\n try {\n const ta = this.$refs && this.$refs.answersEditorTa;\n if (!ta) return;\n ta.style.height = 'auto';\n ta.style.height = (ta.scrollHeight + 2) + 'px';\n } catch (_) {}\n },\n\n // Navigation\n // Non-linear navigation\n get remainingIndices() {\n if (!Array.isArray(this.questions)) return [];\n const remain = [];\n for (let i = 0; i < this.questions.length; i++) {\n const q = this.questions[i];\n if (!q || !q.id) continue;\n if (!this.answeredMap[q.id]) remain.push(i);\n }\n return remain;\n },\n allDone() { return this.questions.length > 0 && this.remainingIndices.length === 0; },\n pickRandomIndex() {\n const pool = this.remainingIndices.length > 0 ? this.remainingIndices : (this.questions.map((_, i) => i));\n if (pool.length === 0) return 0;\n const r = Math.floor(Math.random() * pool.length);\n return pool[r];\n },\n jumpToRandomUnanswered() {\n const idx = this.pickRandomIndex();\n this.currentIndex = idx;\n this.answerBuffer = this.currentAnswer;\n },\n skipQuestion() {\n this.jumpToRandomUnanswered();\n },\n continueToNext() {\n // Ensure current answer is captured and marked\n const id = this.currentId;\n if (id) {\n const txt = (this.answerBuffer || '').trim();\n const prev = this.answers[id] || {};\n this.answers[id] = { ...prev, answer_text: this.answerBuffer || '', created_at: prev.created_at || new Date().toISOString() };\n if (txt) this.answeredMap[id] = true;\n else delete this.answeredMap[id];\n try { this.saveToSessionStorage(); } catch (_) {}\n try { this.saveToLocalStorage(); } catch (_) {}\n }\n this.jumpToRandomUnanswered();\n },\n\n // Current question/answer\n get currentQuestion() { return this.questions[this.currentIndex] || null; },\n get currentId() { return this.currentQuestion?.id || null; },\n get currentAnswer() {\n const id = this.currentId; if (!id) return '';\n return this.answers[id]?.answer_text || '';\n },\n setCurrentAnswer(v) {\n const id = this.currentId; if (!id) return;\n const prev = this.answers[id] || {};\n const trimmed = (v || '').trim();\n this.answers[id] = { ...prev, answer_text: v, created_at: prev.created_at || new Date().toISOString() };\n this.answerBuffer = v;\n // mark answered if non-empty, else unmark\n if (trimmed) this.answeredMap[id] = true; else delete this.answeredMap[id];\n // autosave to both storages\n try { this.saveToSessionStorage(); } catch (_) {}\n try { this.saveToLocalStorage(); } catch (_) {}\n },\n\n insertTranscriptIntoAnswer(text) {\n const id = this.currentId; if (!id) return;\n const prev = this.answers[id] || {};\n const merged = (prev.answer_text && prev.answer_text.trim())\n ? (prev.answer_text + '\\n' + text)\n : text;\n this.answers[id] = { ...prev, answer_text: merged, transcript_text: text, created_at: prev.created_at || new Date().toISOString() };\n this.answeredMap[id] = true;\n // autosave\n try { this.saveToSessionStorage(); } catch (_) {}\n try { this.saveToLocalStorage(); } catch (_) {}\n },\n\n // Export answers (minimal response schema)\n getExportPayload() {\n const questionById = Object.fromEntries((this.questions || []).map(q => [q.id, q]));\n const answers = Object.entries(this.answers || {}).map(([qid, a]) => {\n const q = questionById[qid] || {};\n return {\n question_id: qid,\n label: q.label || null,\n instruction: q.instruction || null,\n tags: Array.isArray(q.tags) ? q.tags : null,\n raw_response: a?.answer_text || '',\n created_at: a?.created_at || new Date().toISOString()\n };\n });\n return { app: 'lorraine', version: 2, language: this.selectedLanguage, job: this.jobName || null, answers };\n },\n\n getExportJsonPretty() {\n try { return JSON.stringify(this.getExportPayload(), null, 2); } catch (_) { return '{}'; }\n },\n\n downloadAnswers() {\n const payload = this.getExportPayload();\n const blob = new Blob([JSON.stringify(payload, null, 2)], { type: 'application/json' });\n const url = URL.createObjectURL(blob);\n const a = document.createElement('a');\n a.href = url; a.download = 'lorraine_answers.json';\n document.body.appendChild(a); a.click(); a.remove();\n URL.revokeObjectURL(url);\n },\n\n // Download JSONL (prompt/completion schema)\n downloadAnswersJsonl() {\n try {\n const questionById = Object.fromEntries((this.questions || []).map(q => [q.id, q]));\n const lines = [];\n for (const [qid, a] of Object.entries(this.answers || {})) {\n const txt = (a && a.answer_text) ? String(a.answer_text).trim() : '';\n if (!txt) continue;\n const q = questionById[qid] || {};\n const instr = (q && q.instruction) ? String(q.instruction) : '';\n const ex = (q && q.example) ? String(q.example) : '';\n let prompt = 'Instructions: ' + instr;\n if (ex && ex.trim()) {\n prompt += '\\n\\nExample: ' + ex;\n }\n const obj = { prompt: prompt, completion: txt };\n lines.push(JSON.stringify(obj));\n }\n const data = lines.join('\\n');\n const blob = new Blob([data], { type: 'text/plain' });\n const url = URL.createObjectURL(blob);\n const a = document.createElement('a');\n a.href = url; a.download = 'lorraine_answers.jsonl';\n document.body.appendChild(a); a.click(); a.remove();\n URL.revokeObjectURL(url);\n } catch (_) {\n // no-op\n }\n },\n\n // Corpus tool helpers\n toggleCorpusTool() {\n this.showCorpusTool = !this.showCorpusTool;\n if (this.showCorpusTool) {\n this.corpusStatus = '';\n this.corpusError = '';\n }\n },\n parseTagsInput(text) {\n const raw = String(text || '').trim();\n if (!raw) return [];\n return raw.split(/[;,\\s]+/).map(s => s.trim()).filter(Boolean);\n },\n async generateFromCorpus() {\n try {\n this.corpusError = '';\n this.corpusStatus = '';\n if (!this.hasServerKey && !this.userApiKey) {\n this.corpusError = 'please add api key';\n return;\n }\n const maxQ = Math.max(1, Math.min(50, Number(this.corpusMaxQuestions) || 4));\n const body = {\n userApiKey: this.userApiKey || undefined,\n corpusText: this.corpusInput || '',\n max_questions: maxQ,\n tags: this.parseTagsInput(this.corpusTags),\n };\n this.corpusGenerating = true;\n const resp = await fetch('/api/corpus-to-questions', {\n method: 'POST',\n headers: { 'Content-Type': 'application/json' },\n body: JSON.stringify(body)\n });\n if (!resp.ok) {\n const txt = await resp.text().catch(() => '');\n throw new Error(txt || ('HTTP ' + resp.status));\n }\n const data = await resp.json();\n const questions = Array.isArray(data?.questions) ? data.questions : [];\n const answers = Array.isArray(data?.answers) ? data.answers : [];\n if (questions.length === 0) {\n this.corpusError = 'no questions generated';\n return;\n }\n // Load questions into app and persist as custom set\n this.questions = questions;\n try { localStorage.setItem('lorraine_custom_questions_v1', JSON.stringify(questions)); } catch (_) {}\n // Import answers using existing importer\n try {\n this.importJsonBuffer = JSON.stringify({ answers }, null, 2);\n this.importAnswersFromJson();\n } catch (_) {}\n this.rebuildAnsweredMap();\n this.jumpToRandomUnanswered();\n this.corpusStatus = `generated ${questions.length} questions`;\n } catch (e) {\n this.corpusError = e?.message || String(e);\n } finally {\n this.corpusGenerating = false;\n }\n },\n\n async copyAnswersJson() {\n try {\n const json = this.getExportJsonPretty();\n await navigator.clipboard.writeText(json);\n this.importSuccess = 'copied';\n setTimeout(() => { this.importSuccess = ''; }, 2000);\n } catch (e) {\n this.importError = 'copy failed';\n setTimeout(() => { this.importError = ''; }, 3000);\n }\n },\n\n toggleImportArea() {\n this.importJsonOpen = !this.importJsonOpen;\n if (this.importJsonOpen) {\n this.importJsonBuffer = '';\n this.importError = '';\n this.importSuccess = '';\n }\n },\n\n importAnswersFromJson() {\n this.importError = '';\n this.importSuccess = '';\n let parsed = null;\n try {\n parsed = JSON.parse(this.importJsonBuffer || '');\n } catch (e) {\n this.importError = 'invalid JSON';\n return;\n }\n let list = [];\n if (parsed && Array.isArray(parsed)) {\n list = parsed;\n } else if (parsed && Array.isArray(parsed.answers)) {\n list = parsed.answers;\n }\n if (!Array.isArray(list) || list.length === 0) {\n this.importError = 'no answers found';\n return;\n }\n const next = { ...this.answers };\n for (const item of list) {\n const qid = item?.question_id || item?.id;\n const text = item?.raw_response || item?.answer_text || item?.answer || '';\n if (!qid || !String(qid).trim()) continue;\n if (!text || !String(text).trim()) continue;\n const existed = next[qid] || {};\n next[qid] = {\n ...existed,\n answer_text: text,\n transcript_text: item?.transcript_text || existed.transcript_text || null,\n created_at: existed.created_at || item?.created_at || new Date().toISOString()\n };\n }\n this.answers = next;\n this.rebuildAnsweredMap();\n // keep UI buffer in sync\n this.answerBuffer = this.currentAnswer;\n // save after import\n try { this.saveToSessionStorage(); } catch (_) {}\n try { this.saveToLocalStorage(); } catch (_) {}\n this.importSuccess = 'loaded';\n setTimeout(() => { this.importSuccess = ''; }, 2000);\n },\n\n clearAnswers() {\n this.answers = {};\n this.answerBuffer = this.currentAnswer;\n try { sessionStorage.removeItem('lorraine_answers_v1'); } catch (_) {}\n try { localStorage.removeItem('lorraine_answers_v1'); } catch (_) {}\n },\n\n saveToLocalStorage() {\n try { localStorage.setItem('lorraine_answers_v1', this.getExportJsonPretty()); } catch (_) {}\n },\n loadFromLocalStorage() {\n try {\n const raw = localStorage.getItem('lorraine_answers_v1');\n if (!raw) return;\n this.importJsonBuffer = raw;\n this.importAnswersFromJson();\n } catch (_) {}\n },\n saveToSessionStorage() {\n try { sessionStorage.setItem('lorraine_answers_v1', this.getExportJsonPretty()); } catch (_) {}\n },\n loadFromSessionStorage() {\n try {\n const raw = sessionStorage.getItem('lorraine_answers_v1');\n if (!raw) return;\n this.importJsonBuffer = raw;\n this.importAnswersFromJson();\n } catch (_) {}\n },\n\n rebuildAnsweredMap() {\n const map = {};\n for (const [qid, a] of Object.entries(this.answers || {})) {\n if (a && String(a.answer_text || '').trim()) map[qid] = true;\n }\n this.answeredMap = map;\n },\n\n // Microphone + Whisper\n async startRecording() {\n try {\n this.isTranscribing = true;\n this.mediaStream = await navigator.mediaDevices.getUserMedia({ audio: true });\n this.chunks = [];\n this.recorder = new MediaRecorder(this.mediaStream);\n this.recorder.ondataavailable = (e) => { if (e.data && e.data.size > 0) this.chunks.push(e.data); };\n this.recorder.onstop = async () => {\n try {\n const blob = new Blob(this.chunks, { type: 'audio/webm' });\n await this.transcribeBlob(blob);\n } catch (e) {\n this.errorMessage = 'Recording error: ' + (e?.message || String(e));\n } finally {\n this.isTranscribing = false;\n this.cleanupMedia();\n }\n };\n this.recorder.start();\n } catch (e) {\n this.isTranscribing = false;\n this.errorMessage = 'Microphone error: ' + (e?.message || String(e));\n this.cleanupMedia();\n }\n },\n stopRecording() { try { this.recorder && this.recorder.stop(); } catch (_) {} },\n cleanupMedia() {\n try { this.mediaStream?.getTracks()?.forEach(t => t.stop()); } catch (_) {}\n this.mediaStream = null; this.recorder = null; this.chunks = [];\n },\n async transcribeBlob(blob) {\n // Require key if no server key\n if (!this.hasServerKey && !this.userApiKey) {\n this.errorMessage = 'please add api key';\n return;\n }\n const form = new FormData();\n form.append('file', blob, 'audio.webm');\n form.append('userApiKey', this.userApiKey || '');\n try {\n const r = await fetch('/api/transcribe', { method: 'POST', body: form });\n if (!r.ok) throw new Error('transcription failed');\n const j = await r.json();\n const text = (j && j.text) ? String(j.text) : '';\n if (text) this.insertTranscriptIntoAnswer(text);\n } catch (e) {\n this.errorMessage = 'Transcription error: ' + (e?.message || String(e));\n }\n },\n\n // Minimal legacy trace helpers (no-ops unless trace is visible)\n computeTraceMaxHeight() { return this.traceExpanded ? '' : 'max-height: 100px'; },\n toggleTool(idx) { this.expandedToolMap[idx] = !this.expandedToolMap[idx]; this.queueScrollToBottom(); },\n prettify(obj) { try { return JSON.stringify(obj, null, 2); } catch (_) { return String(obj); } },\n async copyEventsJson() {\n try {\n const json = JSON.stringify(this.traceEvents || [], null, 2);\n await navigator.clipboard.writeText(json);\n this.copyEventsText = 'copied';\n setTimeout(() => { this.copyEventsText = 'copy json'; }, 2000);\n } catch (_) {}\n },\n queueScrollToBottom() {\n try {\n const el = this.$refs && this.$refs.traceContainer;\n if (!el) return;\n const doScroll = () => { el.scrollTop = el.scrollHeight; };\n if (typeof this.$nextTick === 'function') this.$nextTick(doScroll); else requestAnimationFrame(doScroll);\n } catch (_) {}\n },\n\n // Sidebar helpers (keep minimal compatibility with layout)\n toggleSidebar() { this.sidebarOpen = !this.sidebarOpen; },\n collapseSidebar() { this.sidebarTransitioning = true; this.sidebarCollapsed = true; setTimeout(() => { this.sidebarTransitioning = false; }, this.sidebarAnimMs); },\n expandSidebar() { this.sidebarTransitioning = true; this.sidebarCollapsed = false; setTimeout(() => { this.sidebarTransitioning = false; }, this.sidebarAnimMs); },\n\n // Loading message placeholder\n get loadingMessage() { return 'Loading...'; },\n\n // i18n minimal for Lorraine\n get i18n() {\n const t = {\n english: {\n title: 'Lorraine',\n subtitle: 'Answer prompts in your voice. Build a dataset for your personal style.',\n apiKey: 'API Key:', change: 'Change', clear: 'Clear', getKeyHere: 'Get your free key here',\n yourVersion: 'Your version',\n prev: 'Previous', next: 'Next',\n record: 'Record', stop: 'Stop', insertTranscript: 'Insert transcript',\n export: 'Download answers JSON', language: 'language'\n }\n };\n return t[this.selectedLanguage] || t.english;\n }\n };\n}\n\n\n",[],[186,283],"76b7b9ba-7618-11f0-a463-0224a6c84d84","init.js","76b7ba6e-7618-11f0-a463-0224a6c84d84",14,["D",1754850789077],"import researchApp from './researchApp.js';\nimport lorraineApp from './lorraineApp.js';\n\n// Expose globally so existing x-data=\"researchApp()\" continues to work\nwindow.researchApp = researchApp;\nwindow.lorraineApp = lorraineApp;\n\n// Register with Alpine when ready\nif (window.Alpine && typeof window.Alpine.data === 'function') {\n window.Alpine.data('researchApp', researchApp);\n} else {\n document.addEventListener('alpine:init', () => {\n try {\n if (window.Alpine && typeof window.Alpine.data === 'function') {\n window.Alpine.data('researchApp', researchApp);\n window.Alpine.data('lorraineApp', lorraineApp);\n }\n } catch (_) { /* no-op */ }\n });\n}\n\n// If Alpine was deferred via window.deferLoadingAlpine, start it now\ntry { if (typeof window._startAlpine === 'function') window._startAlpine(); } catch (_) {}\n\n\n",[],[186,291],["D",1754850788185],[219,220,221],[186],"769b4ce4-7618-11f0-a93a-0224a6c84d84","questions_design.md","769b4d98-7618-11f0-a93a-0224a6c84d84",9,["D",1754850788959],"### Question design for collecting personalized multi‑context responses\n\nThis document describes the current, simple question/response format used in the app, plus an optional appendix of advanced facets for future phases. The immediate goal is to collect high‑quality, easy‑to‑author pairs for LoRA fine‑tuning of “style transfer” with minimal metadata.\n\n### Objectives\n- **Coverage**: Capture the breadth of situations people respond to (work, personal, public, service, etc.).\n- **Control**: Parameterize questions with facets (tone, formality, emotion, power dynamics) for targeted sampling.\n- **Parsimony**: Achieve high coverage without exploding the cross‑product (via stratified/structured sampling).\n- **Privacy**: Support redaction and sensitivity metadata.\n- **Trainability**: Produce consistent, machine‑readable input/output pairs with rich metadata for conditioning.\n\n### Minimal question schema (current)\n\nKeep authoring dead‑simple so anyone (including kids/teens/non‑tech users) can contribute consistent data.\n\n```json\n{\n \"id\": \"string\",\n \"label\": \"short descriptive name\",\n \"instruction\": \"single instruction text (includes the scenario)\",\n \"tags\": [\"grouping\", \"filters\"],\n \"example\": \"a neutral, normal-human example reply (optional but recommended)\"\n}\n```\n\nNotes\n- instruction merges the old instruction + source_text into one block. Write it in plain language.\n- example replaces the old neutral_sample. It’s what a normal human might say; concise; no sensitive data.\n- tags are lightweight for grouping/filters (e.g., friends, email, invite). No other fields are required.\n\nFile layout\n- Files live under `data/questions/` and may be sharded (e.g., `work.json`, `friends.json`, `youth/school.json`).\n- Aggregation endpoints combine files as needed:\n - `/data/questions/simple.json` (all simple questions, recursive)\n - `/data/questions/simple/by-groups.json?groups=work,friends` (selected groups)\n\nAuthoring rules\n- Keep instruction and example at a grade‑6 to grade‑9 reading level when possible.\n- Use whitespace/newlines in instruction to separate “context” and “what to do.”\n- Avoid extra formatting, bullet lists, or markup; keep it plain.\n- Never include real PII in example; paraphrase.\n\nUI presentation\n- Title/label is not emphasized.\n- The app shows a simple block:\n - “Instructions: ”\n - “Example: ” (always shown if present)\n\n### Minimal response schema (current)\n\n```json\n{\n \"question_id\": \"string\",\n \"raw_response\": \"user’s answer in their voice\",\n \"created_at\": \"ISO 8601\",\n \"neutral_distance_note\": \"optional freeform note\",\n \"response_latency_ms\": 1234\n}\n```\n\nNotes\n- raw_response captures exactly how the user would say it. No polishing required.\n- neutral_distance_note is optional (e.g., “I’d be more direct here than the example”).\n\n### Why this minimal spec works for LoRA\n- **Low friction**: fewer fields → faster authoring and better completion rates.\n- **Style focus**: instruction + example give enough context to generate user‑style responses.\n- **Simple conditioning**: tags enable basic slicing without heavy metadata.\n\n### Optional advanced facets (appendix)\n\nThe following facets are NOT required for the current collection, but remain useful if we later need more control or analysis. Treat them as an optional overlay stored separately or added in a later phase.\n\n### Core facets (dimensions) to parameterize a question (optional)\n- **Domain/context**: work, school/academia, family, partner, friends, social, public, customer support, sales, legal/admin, healthcare, travel/hospitality, community/volunteer, gaming, developer/collab (GitHub/PRs).\n- **Channel/medium**: in‑person, phone/voice, video, SMS/iMessage, WhatsApp, Slack/Teams, email, calendar invite, social post (X/IG/LinkedIn/TikTok caption), comment/reply, forum/Discord, GitHub issue/PR/review, blog, support ticket, survey, form.\n- **Audience role**: manager, direct report, peer, cross‑functional partner, vendor, customer, friend (close), friend (acquaintance), partner/spouse, child/parent/sibling, professor/TA/student, doctor/therapist, government/agency.\n- **Relationship closeness**: stranger, acquaintance, familiar, close, intimate.\n- **Power dynamics**: subordinate → equal → senior; high/low leverage; high/low stakes.\n- **Formality**: very informal, informal, neutral, formal, very formal.\n- **Tone/voice**: friendly, warm, enthusiastic, calm, concise, direct, diplomatic, firm, empathetic, humorous, witty, serious, inspirational, authoritative, deferential.\n- **Emotion state (author of response)**: use circumplex (valence: negative↔positive, arousal: low↔high) and/or Plutchik categories (anger, fear, sadness, disgust, surprise, anticipation, trust, joy) with intensities.\n- **Emotion target**: self‑oriented vs other‑oriented; directed at person vs topic vs situation.\n- **Speech act / intent**: ask/request, propose, inform, confirm, deny, clarify, instruct, announce, invite, accept, decline, schedule, reschedule, apologize, thank, congratulate, praise, critique, give feedback, complain, escalate, negotiate, set boundary, follow up, remind, nudge, check in, summarize, debrief, reflect.\n- **Urgency/time pressure**: none, low, medium, high, crisis.\n- **Constraints**: length (very short, short, medium, long), bullets vs paragraph, with/without emoji, with/without links, subject line required, call‑to‑action required, deadline date/time, reading level.\n- **Sensitivity/compliance**: PII present, confidential, legal exposure, HR policy, medical/financial, regulated industry, brand guideline needed.\n- **Cultural/region/language**: locale (en‑US, en‑UK, en‑IN, etc.), politeness norms, honorifics, code‑switching preferences.\n- **Time/contextual anchors**: time of day, weekday/weekend, before/after an event (launch, outage, interview, exam), proximity to deadline.\n- **Topic domain**: engineering, product, design, marketing, sales, ops, finance, HR, education, research, creative, household/admin, travel, events.\n- **Formatting preferences**: greetings/closings style, signature block, bullets, numbering, sections, TL;DR, action list.\n- **Style modifiers (pragmatics)**: directness, hedging, intensifiers, face‑saving, humor level, exclamation usage, emoji usage, colloquialisms, slang.\n\n### Personality and preference frameworks (optional)\n- **MBTI (16 types)**: e.g., INTJ, ENFP. Store declared type or unknown.\n- **Big Five (OCEAN)**: scores 0–100 per trait.\n- **Enneagram**: types 1–9, optional wing.\n- **DiSC**: D/I/S/C primary and secondary.\n- **Communication preferences**: directness (1–5), candor (1–5), warmth (1–5), humor (1–5), emoji usage (never→often), exclamation tolerance, sentence length preference, bullet vs paragraph preference.\n- **Professional voice toggles**: brand voice adjectives, do/don’t words, signature/greeting conventions.\n\n### Extended question item schema (optional overlay)\nIf/when needed, we can layer additional fields (domain, channel, tone, etc.) as a separate metadata file or by expanding items. See the earlier version of this document for the full list of optional fields.\n\n### Extended response schema (optional overlay)\nAdvanced fields like polished_response, self_ratings, or derived metrics can be added later if we need deeper analysis. For now we only capture the minimal fields listed above.\n\n### Enumerations and examples per facet\n- **Domain/context**\n - Work: 1:1, standup, status update, project kickoff, code review, PR description, incident postmortem, performance review, promotion/raise ask, interview scheduling, offer negotiation, vendor management.\n - Personal: partner check‑in, apology, boundary setting, planning a trip, money talk, household admin, parenting.\n - Friends/social: invite, RSVP, plan details, jokes/banter, group coordination, conflict resolution.\n - Public/social: announcement, thread reply, correction, hot take, event recap, launch post, mea culpa.\n - Service/support: report an issue, ask for refund, give praise, complaint escalation.\n - Admin/legal: lease inquiry, government forms, policy appeal.\n - Healthcare: appointment request, symptoms note, follow‑up question, insurance paperwork.\n - Travel/hospitality: booking request, change/reschedule, special request.\n - Education: email professor, TA feedback, peer review, assignment clarification.\n - Developer collaboration: GitHub issue, PR summary, code review comment, RFC response, design doc feedback.\n\n- **Speech act / intent (canonical set)**\n - Ask/request, propose/suggest, inform/update, confirm/deny, clarify, instruct, announce, invite, accept/decline, schedule/reschedule, apologize, thank, congratulate, praise, critique, feedback, complain, escalate, negotiate, set boundary, follow up, remind, nudge, check in, summarize, debrief, reflect.\n\n- **Tone/voice (mix‑and‑match)**\n - Friendly, warm, calm, concise, direct, diplomatic, firm, empathetic, humorous, witty, serious, inspirational, authoritative, deferential.\n\n### Sampling strategy (to avoid combinatorial explosion)\n- **Stratified grid**: Choose anchor levels for key axes: domain × channel × speech_act × (formality: low/med/high) × (valence: −/0/+) × (power: sub/equal/senior).\n- **Latin hypercube sampling** across remaining facets (tone modifiers, constraints, topic domains) to ensure even coverage.\n- **Phase coverage quotas** (suggested minima):\n - Work (email, Slack, PR/GitHub): 40–60 items\n - Personal/partner/family: 20–30\n - Friends/social: 20–30\n - Public/social media: 15–25\n - Service/support/admin: 15–25\n - Education/healthcare/travel: 10–20\n - Emotions sweep (same scenario across −/0/+ valence and low/high arousal): 15–25\n\n### Template library (examples; fill with scenario content)\n- **Work – status update (email, neutral)**: “Share a brief project status with next steps and date‑based milestones.”\n- **Work – request (Slack, equal)**: “Ask a peer to review your PR today.”\n- **Work – escalate (email, senior)**: “Escalate a blocker with clear impact and request for decision.”\n- **Work – feedback (1:1 note)**: “Give constructive feedback on missed deadline; keep rapport.”\n- **Work – negotiation (email)**: “Negotiate timeline extension by one week; offer trade‑offs.”\n- **Developer – PR summary**: “Write a PR description summarizing changes, risks, and testing notes.”\n- **Developer – code review**: “Request changes on a PR; be specific and respectful.”\n- **Personal – apology**: “Apologize for being late to dinner; own it and propose a make‑good.”\n- **Personal – boundary**: “Decline an invitation without hurting feelings.”\n- **Friends – planning**: “Coordinate a group dinner; propose 2 dates and a budget.”\n- **Friends – banter**: “Respond playfully to a joke; keep it light.”\n- **Partner – check‑in**: “Express appreciation after a tough week.”\n- **Public – announcement**: “Post about launching a small tool; concise and authentic.”\n- **Public – correction**: “Correct misinformation politely in a thread.”\n- **Support – refund**: “Request a refund for a faulty product; include order details.”\n- **Support – praise**: “Thank a customer support rep for great service.”\n- **Admin – policy appeal**: “Appeal a parking ticket; state facts and request leniency.”\n- **Healthcare – appointment**: “Ask to move an appointment earlier; mention symptom severity.”\n- **Travel – change flight**: “Request flight change due to conflict; provide preferences.”\n- **Education – email professor**: “Ask for clarification on assignment requirements.”\n\nEach template becomes a question by binding facets (channel, tone, formality, etc.) and a concrete `source_text` stimulus.\n\n### Prompt structure recommendations (for consistent collection)\n- **Instruction template**: “Reply in your own voice, matching the specified tone and constraints.”\n- **Context block**: scenario summary + `source_text` (if provided) + facets.\n- **Constraints**: explicit bullets (length, emoji, bullets vs paragraph, deadline).\n\nExample assembled question (human‑readable):\n- **Label**: Work email follow‑up (neutral, concise)\n- **Instruction**: Reply in your own voice. Keep it concise and warm. Include a clear ask.\n- **Source text**: “Circling back on the Q3 deliverables: can you share an updated timeline by Friday?”\n- **Facets**: domain: work; channel: email; audience: peer; power: equal; formality: neutral; tone: [concise, warm]; speech_act: [follow_up, request]; urgency: medium.\n\n### File structure and schemas\n- **Files**\n - `data/questions/*.json`: question banks by domain/group (simple schema above).\n - `data/responses/*.jsonl`: user responses (one JSON per line) using the minimal response schema.\n - `schemas/question.schema.json`, `schemas/response.schema.json`: JSON Schemas for validation.\n\n### Compatibility\n- For legacy items with `neutral_sample`, the app treats it as `example`.\n- Old `source_text` is now merged into `instruction`.\n\n### Phase plan (simplified)\n- Phase 1: ship with minimal schema (current). Focus on breadth and ease of authoring.\n- Phase 2 (optional): add overlays for emotions/power/tone on a subset if needed.\n- Phase 3 (optional): domain deep dives with richer metadata if training analysis demands it.\n\n### Privacy and ethics\n- Built‑in redaction UI; sensitivity flags; local‑first storage options.\n- Guidance to paraphrase sensitive specifics while preserving style markers.\n\n### Next steps\n- Implement JSON Schemas and a generator that samples templates × facets with quotas.\n- Build a UI that presents scenario + facets clearly, collects responses, self‑ratings, and optional rationale.\n- Add analytics to track coverage across facets and identify gaps for targeted sampling.\n\n\n",[],[192,302],["D",1754850788090],[217],[192],"767abf6a-7618-11f0-836d-0224a6c84d84","youth","767ac050-7618-11f0-836d-0224a6c84d84","7664478a-7618-11f0-bde4-0224a6c84d84",7,["D",1754850788746],[],[198,320,313],"questions","766446e0-7618-11f0-bde4-0224a6c84d84",6,["D",1754850788582],[215],[198,320],["D",1754850788481],[214],[198],["D",1754850754170],{"_95":94,"_102":98},{"_95":94,"_96":-5,"_97":98,"_18":53,"_99":100,"_101":24},{"_13":84,"_31":85,"_41":40,"_86":42,"_87":-5,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":43,"_93":94},"ctx",{"_77":-5,"_121":366,"_367":368,"_369":370,"_371":24,"_372":24,"_373":24,"_374":57,"_375":-5,"_376":-7,"_377":-7,"_378":61,"_379":24,"_380":24,"_381":-5,"_382":24,"_383":24,"_384":24,"_385":-5},"ENV",{"_337":338,"_339":340,"_341":85,"_342":343,"_344":345,"_346":347,"_348":349,"_350":351,"_352":353,"_354":355,"_356":357,"_358":359,"_360":361,"_362":363,"_364":365},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","5015b8df46402d8bba4f485ecc37aec202b0e06b","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"]

Diff · v19