[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_156":157,"_158":159},"routes/_app.x.$handle.$name",{"_11":12,"_13":14,"_93":-7,"_94":95,"_29":-7,"_96":-5,"_15":97,"_98":99,"_100":101,"_102":103,"_104":105,"_106":-5,"_107":-5,"_108":-5,"_109":38,"_110":111,"_112":50,"_113":-5,"_114":-5,"_115":116},"routes/_app.x.$handle.$name.branches",{"_11":12,"_13":14,"_15":16,"_17":18},"handle","raddman","name","video","project",{"_20":28,"_13":14,"_22":36,"_35":-5,"_37":38,"_39":40,"_41":40,"_42":43,"_44":-5,"_45":46,"_47":48,"_49":50,"_51":-5,"_52":-5,"_53":54,"_55":56,"_57":58,"_59":60,"_61":62,"_63":-5,"_64":65,"_66":65,"_67":68,"_69":68,"_70":68,"_71":65,"_72":65,"_73":74,"_75":76,"_77":78,"_79":38},"branches",[19],{"_20":21,"_22":23,"_24":25,"_13":26,"_27":28,"_29":30,"_31":-5,"_32":-5,"_33":34,"_35":-5},"id","2e9d81ea-5c1e-11f1-924f-ee650bb23af1","createdAt",["D",1780142012604],"updatedAt",["D",1781303607735],"main","projectId","2e9ca14e-5c1e-11f1-924f-ee650bb23af1","version",26,"forkedBranchId","forkedBranchVersion","lockfileKey","1781303608223-esm.town_v_raddman_video@26-main","deletedAt",["D",1780142012597],"legacyEnvVars",false,"privacy","public","httpPrivacy","description","Host","imageUrl","userId","e1d9874c-49a0-4fd0-a196-e29af0fc3985","pinnedFileId","3b41e0c6-5c1e-11f1-924f-ee650bb23af1","pinPreview",true,"pinMigrated","screenshotVersion","likeCount",0,"orgId","019e78ba-264d-73c7-b4fa-c98be63ae754","tursoName","2e9ca14e5c1e11f1924fee650bb23af1","tursoHostname","2e9ca14e5c1e11f1924fee650bb23af1-val-town.aws-us-east-2.turso.io","tursoMeasuredAt",["D",1789471840176],"pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_11":12,"_92":88},"org",{"_11":12,"_85":86,"_87":88,"_45":46,"_89":90,"_91":38},"branch",{"_29":30,"_13":26,"_20":21,"_27":28,"_31":-5,"_32":-5,"_80":-5,"_81":-5,"_82":-5,"_83":14,"_84":12},"isOwned","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","displayName","Rain","avatar","https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/9e1a4958-e7ce-4b51-c18a-bd85c98bf600/square512?exp=1789527599&sig=4fc3a62a20555216db2a6e42a409235d3f742de4c96a97a16d42ca053fd1b151","tier","free","canRestrictVals","profileImageUrl","branchName","path","","adoptOrgId",{"_20":28,"_13":14,"_22":151,"_35":-5,"_37":38,"_39":40,"_41":40,"_42":43,"_44":-5,"_45":46,"_47":48,"_49":50,"_51":-5,"_52":-5,"_53":54,"_55":56,"_57":58,"_59":60,"_61":152,"_63":-5,"_64":65,"_66":65,"_67":68,"_69":68,"_70":68,"_71":65,"_72":65,"_73":153,"_75":154,"_77":155,"_79":38},"tree",[146],"files",[146],"counts",{"_135":54,"_17":136,"_137":54,"_138":139,"_140":141},"contents",[117],"file","upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},{"_20":118,"_13":119,"_120":48,"_121":-5,"_122":21,"_123":30,"_124":50,"_22":125,"_29":126,"_127":128,"_129":130,"_131":38,"_132":133,"_94":134},"44f51da0-38ba-4655-9ec1-387179bc98a1","main.ts","fileId","parentId","branchId","branchVersion","latest",["D",1781303607735],25,"fileType","http","content","// Video mirror: requests to /v3/ are proxied from an upstream host and\n// streamed back to the client. The upstream domain is NEVER exposed to the\n// client — not in URLs, headers, redirects, or error messages.\n\nconst UPSTREAM_HOST = Deno.env.get(\"UPSTREAM_HOST\") ?? \"\";\nconst UPSTREAM_PREFIX = Deno.env.get(\"UPSTREAM_PREFIX\") ?? \"/\";\nconst MIRROR_PREFIX = \"/v3/\";\n\n// Second mirror: /v4/ proxies from the Vercel blob store.\nconst VERCEL_PREFIX = \"/v4/\";\nconst VERCEL_HOST = Deno.env.get(\"VERCELURL\") ?? \"\";\nconst VERCEL_UPSTREAM_PREFIX = Deno.env.get(\"VERCEL_UPSTREAM_PREFIX\") ?? \"/\";\n\nconst FORWARD_REQUEST_HEADERS = [\n \"range\",\n \"if-range\",\n \"accept\",\n];\n\nconst SAFE_RESPONSE_HEADERS = [\n \"content-type\",\n \"content-length\",\n \"content-range\",\n \"accept-ranges\",\n \"cache-control\",\n \"expires\",\n];\n\n// Extensions the upstream stores with their REAL name (no .png disguise).\nconst KEEP_ORIGINAL_EXTS = new Set([\n \"png\",\n \"jpg\",\n \"jpeg\",\n \"gif\",\n \"mp4\",\n \"mp3\",\n \"pdf\",\n \"txt\",\n]);\n\n// Extensions that should play/show INLINE in the browser.\nconst INLINE_EXTS = new Set([\n // images\n \"png\",\n \"jpg\",\n \"jpeg\",\n \"gif\",\n \"webp\",\n \"avif\",\n \"bmp\",\n \"svg\",\n \"ico\",\n // streamable video\n \"mp4\",\n \"webm\",\n]);\n\nconst STALL_MS = 8000; // Replicate's CDN drops fast -> shorter stall window\nconst MAX_RESUMES = 25; // consecutive failed resumes (without progress) before giving up\nconst CONNECT_TIMEOUT_MS = 15000; // max wait for upstream to send response headers\n\nfunction sleep(ms: number): Promise {\n return new Promise((r) => setTimeout(r, ms));\n}\n\nfunction jsonError(status: number): Response {\n return Response.json({ message: \"dont work\" }, { status });\n}\n\n// fetch that aborts if the upstream doesn't send response HEADERS in time.\n// The body keeps streaming freely once headers arrive (timer is cleared the\n// moment fetch() resolves, which is when headers are received).\nasync function fetchWithTimeout(\n url: string,\n opts: RequestInit,\n timeoutMs: number,\n): Promise {\n const ac = new AbortController();\n const timer = setTimeout(() => ac.abort(), timeoutMs);\n try {\n return await fetch(url, { ...opts, signal: ac.signal });\n } finally {\n clearTimeout(timer);\n }\n}\n\n// Read a single chunk but give up if the upstream goes quiet for `ms`.\nasync function readWithTimeout(\n reader: ReadableStreamDefaultReader,\n ms: number,\n): Promise | \"timeout\"> {\n let timer: number | undefined;\n const timeout = new Promise<\"timeout\">((r) => {\n timer = setTimeout(() => r(\"timeout\"), ms);\n });\n try {\n return await Promise.race([reader.read(), timeout]);\n } finally {\n clearTimeout(timer);\n }\n}\n\n// Get the real filename + ext from the path, stripping a decoy \".png\".\nfunction realFile(rest: string): { name: string; ext: string } {\n const seg = rest.split(\"/\").pop() || rest;\n let name = seg;\n if (name.toLowerCase().endsWith(\".png\")) {\n const stripped = name.slice(0, -4);\n if (stripped.includes(\".\")) name = stripped; // .glb.png -> .glb\n // else: it's a real \".png\" image, leave it\n }\n const dot = name.lastIndexOf(\".\");\n const ext = dot >= 0 ? name.slice(dot + 1).toLowerCase() : \"\";\n return { name, ext };\n}\n\n// Stream upstream to the client, transparently resuming via Range requests if\n// the upstream connection stalls or drops mid-transfer. The client sees one\n// smooth, continuous download. Respects backpressure (returns after each chunk)\n// so large videos aren't buffered entirely in memory.\nfunction resilientBody(\n upstreamURL: string,\n baseHeaders: Headers,\n startResp: Response,\n startOffset: number,\n totalSize: number, // expected final byte offset (-1 if unknown)\n): ReadableStream {\n let resp: Response | null = startResp;\n let reader: ReadableStreamDefaultReader | null = null;\n let offset = startOffset;\n let skip = 0;\n let failedResumes = 0; // consecutive resumes that made NO progress\n\n // Make sure we have a live reader pointing at `offset`. Returns false only\n // if it has already closed/errored the controller.\n async function ensureReader(\n controller: ReadableStreamDefaultController,\n ): Promise {\n while (true) {\n if (reader) return true;\n\n if (!resp) {\n if (failedResumes >= MAX_RESUMES) {\n controller.error(new Error(\"too many upstream resumes\"));\n return false;\n }\n failedResumes++;\n const backoff = Math.min(250 * failedResumes, 2000);\n\n const h = new Headers(baseHeaders);\n h.set(\"range\", `bytes=${offset}-`);\n console.log(\n `[resume] requesting bytes=${offset}- (attempt ${failedResumes})`,\n );\n try {\n resp = await fetchWithTimeout(\n upstreamURL,\n { method: \"GET\", headers: h, redirect: \"manual\" },\n CONNECT_TIMEOUT_MS,\n );\n } catch {\n resp = null;\n await sleep(backoff);\n continue;\n }\n console.log(\n `[resume] upstream replied ${resp.status}, content-range=${\n resp.headers.get(\"content-range\")\n }`,\n );\n if (resp.status >= 300 || !resp.body) {\n console.log(`[resume] unusable status ${resp.status}, retrying`);\n try {\n await resp.body?.cancel();\n } catch { /* ignore */ }\n resp = null;\n await sleep(backoff);\n continue;\n }\n }\n\n // If a resumed upstream ignored our Range and sent a full 200, it\n // restarts at byte 0 — so we must skip the bytes already delivered.\n skip = (resp.status === 200 && offset > startOffset)\n ? offset - startOffset\n : 0;\n reader = resp.body!.getReader();\n return true;\n }\n }\n\n return new ReadableStream({\n async pull(controller) {\n while (true) {\n const ok = await ensureReader(controller);\n if (!ok) return;\n\n let result: ReadableStreamReadResult | \"timeout\";\n try {\n result = await readWithTimeout(reader!, STALL_MS);\n } catch {\n // Connection error mid-stream -> resume from `offset`.\n try {\n reader!.releaseLock();\n } catch { /* ignore */ }\n reader = null;\n resp = null;\n continue;\n }\n\n if (result === \"timeout\") {\n console.log(`[stall] at offset=${offset}, fails=${failedResumes}`);\n try {\n await reader!.cancel();\n } catch { /* ignore */ }\n reader = null;\n resp = null;\n continue;\n }\n\n const { done, value } = result;\n\n if (done) {\n try {\n reader!.releaseLock();\n } catch { /* ignore */ }\n reader = null;\n\n // Upstream ended. If we KNOW the total and haven't reached it, the\n // CDN cut us off early -> resume instead of truncating.\n if (totalSize >= 0 && offset < totalSize) {\n console.log(`[short-eof] got ${offset}/${totalSize}, resuming`);\n resp = null;\n continue;\n }\n console.log(`[done] delivered ${offset} bytes total`);\n controller.close();\n return;\n }\n\n if (value && value.length) {\n let chunk = value;\n if (skip > 0) {\n if (skip >= chunk.length) {\n skip -= chunk.length;\n continue; // whole chunk already delivered, drop it\n }\n chunk = chunk.subarray(skip);\n skip = 0;\n }\n offset += chunk.length;\n failedResumes = 0; // real progress -> allow unlimited future resumes\n controller.enqueue(chunk);\n return; // hand control back; respect backpressure\n }\n // empty chunk -> loop and read again\n }\n },\n cancel() {\n try {\n reader?.cancel();\n } catch { /* ignore */ }\n try {\n resp?.body?.cancel();\n } catch { /* ignore */ }\n },\n });\n}\n\nasync function tryFetch(\n upstreamURL: string,\n method: string,\n headers: Headers,\n): Promise {\n let resp: Response;\n try {\n resp = await fetchWithTimeout(\n upstreamURL,\n {\n method: method === \"HEAD\" ? \"HEAD\" : \"GET\",\n headers,\n redirect: \"follow\",\n },\n CONNECT_TIMEOUT_MS,\n );\n } catch {\n return null;\n }\n // Redirects or errors -> not usable, drain and bail.\n // 206 (Partial Content) is a SUCCESS for range requests — keep it.\n if (resp.status >= 300 && resp.status !== 304) {\n try {\n await resp.body?.cancel();\n } catch { /* ignore */ }\n return null;\n }\n return resp;\n}\n\nexport default async function (req: Request): Promise {\n const url = new URL(req.url);\n\n if (req.method === \"OPTIONS\") {\n return new Response(null, {\n status: 204,\n headers: {\n \"access-control-allow-origin\": \"*\",\n \"access-control-allow-methods\": \"GET, HEAD, OPTIONS\",\n \"access-control-allow-headers\": \"range, if-range, accept, content-type\",\n \"access-control-max-age\": \"86400\",\n },\n });\n }\n\n if (req.method !== \"GET\" && req.method !== \"HEAD\") {\n return jsonError(405);\n }\n\n // Pick which upstream to mirror from based on the path prefix.\n let upstreamHostBase: string;\n let rest: string;\n if (url.pathname.startsWith(VERCEL_PREFIX)) {\n upstreamHostBase = VERCEL_HOST + VERCEL_UPSTREAM_PREFIX;\n rest = url.pathname.slice(VERCEL_PREFIX.length);\n } else if (url.pathname.startsWith(MIRROR_PREFIX)) {\n upstreamHostBase = UPSTREAM_HOST + UPSTREAM_PREFIX;\n rest = url.pathname.slice(MIRROR_PREFIX.length);\n } else {\n return jsonError(404);\n }\n\n const { name, ext } = realFile(rest);\n\n // Build candidate upstream paths.\n const base = upstreamHostBase + rest;\n const alreadyPng = rest.toLowerCase().endsWith(\".png\");\n const candidates: string[] = (alreadyPng || KEEP_ORIGINAL_EXTS.has(ext))\n ? [base + url.search]\n : [base + \".png\" + url.search, base + url.search];\n\n const upstreamHeaders = new Headers();\n for (const h of FORWARD_REQUEST_HEADERS) {\n const v = req.headers.get(h);\n if (v) upstreamHeaders.set(h, v);\n }\n\n let upstreamResp: Response | null = null;\n let winningURL = \"\";\n for (const candidate of candidates) {\n upstreamResp = await tryFetch(candidate, req.method, upstreamHeaders);\n if (upstreamResp) {\n // Use the FINAL url after redirects (e.g. the signed GCS/S3 url that\n // replicate.delivery 302s to). That target honors Range requests;\n // the front door re-mints a fresh byte-0 url on every hit, which breaks\n // resume. Resume against the resolved url.\n winningURL = upstreamResp.url || candidate;\n break;\n }\n }\n\n if (!upstreamResp) {\n return jsonError(404);\n }\n\n const respHeaders = new Headers();\n for (const h of SAFE_RESPONSE_HEADERS) {\n const v = upstreamResp.headers.get(h);\n if (v) respHeaders.set(h, v);\n }\n\n // When we resilient-stream a full (non-range) GET, we may stitch the body\n // from multiple upstream connections. Forwarding the upstream content-length\n // is dangerous: a single short hop makes the browser count bytes against the\n // wrong total and stall. Drop it for full GETs and stream to natural EOF.\n if (\n req.method === \"GET\" && !req.headers.has(\"range\") &&\n upstreamResp.status === 200\n ) {\n respHeaders.delete(\"content-length\");\n }\n\n // If the client asked for a range but upstream answered with a full 200,\n // strip range-y headers so the browser treats it as a complete download.\n const clientWantedRange = req.headers.has(\"range\");\n if (clientWantedRange && upstreamResp.status === 200) {\n respHeaders.delete(\"content-range\");\n }\n if (upstreamResp.status === 206 && !respHeaders.has(\"content-range\")) {\n respHeaders.delete(\"content-range\");\n }\n respHeaders.set(\"access-control-allow-origin\", \"*\");\n respHeaders.set(\"referrer-policy\", \"no-referrer\");\n\n // Only advertise range support if upstream actually honored/announced it.\n const upstreamAcceptRanges = upstreamResp.headers.get(\"accept-ranges\");\n const isPartial = upstreamResp.status === 206;\n if (isPartial || upstreamAcceptRanges === \"bytes\") {\n respHeaders.set(\"accept-ranges\", \"bytes\");\n } else {\n respHeaders.delete(\"accept-ranges\");\n respHeaders.delete(\"content-range\");\n }\n\n if (req.headers.get(\"range\") && upstreamResp.status === 200) {\n respHeaders.delete(\"content-range\");\n }\n\n // Decide inline vs download by the REAL extension.\n if (!INLINE_EXTS.has(ext)) {\n respHeaders.set(\"content-type\", \"application/octet-stream\");\n const safeName = (name || \"download\").replace(/\"/g, \"\");\n respHeaders.set(\n \"content-disposition\",\n `attachment; filename=\"${safeName}\"`,\n );\n }\n // else: leave upstream content-type so images/mp4/webm render inline.\n\n // Figure out the byte offset this response starts at.\n let startOffset = 0;\n const cr = upstreamResp.headers.get(\"content-range\"); // \"bytes 500-999/1234\"\n if (cr) {\n const m = cr.match(/bytes\\s+(\\d+)-/i);\n if (m) startOffset = parseInt(m[1], 10);\n } else {\n const rangeReq = req.headers.get(\"range\"); // \"bytes=500-\"\n if (rangeReq) {\n const m = rangeReq.match(/bytes=(\\d+)-/i);\n if (m) startOffset = parseInt(m[1], 10);\n }\n }\n\n // Determine the TOTAL expected size so we can detect a premature EOF.\n let totalSize = -1;\n const crTotal = upstreamResp.headers.get(\"content-range\"); // \"bytes 0-/12345\"\n if (crTotal) {\n const m = crTotal.match(/\\/(\\d+)\\s*$/);\n if (m) totalSize = parseInt(m[1], 10);\n }\n if (totalSize < 0) {\n const cl = upstreamResp.headers.get(\"content-length\");\n if (cl) {\n const n = parseInt(cl, 10);\n if (!Number.isNaN(n)) {\n totalSize = upstreamResp.status === 206 ? startOffset + n : n;\n }\n }\n }\n console.log(\n `[start] offset=${startOffset} totalSize=${totalSize} status=${upstreamResp.status}`,\n );\n\n // For HEAD, no body. For GET, wrap so we transparently resume from the\n // resolved signed url if the CDN cuts the connection halfway.\n const body = req.method === \"HEAD\" ? null : resilientBody(\n winningURL,\n upstreamHeaders,\n upstreamResp,\n startOffset,\n totalSize,\n );\n\n return new Response(body, {\n status: upstreamResp.status,\n headers: respHeaders,\n });\n}","deleted","label","videos",[119],"pulls",1,"forks","envVars",{"_142":144,"_145":54},"connections",{"_142":54,"_143":54},"count","unfilledCount",3,"emptyCount",{"_20":118,"_13":119,"_120":48,"_121":-5,"_122":21,"_123":30,"_124":50,"_22":147,"_29":126,"_127":128,"_129":130,"_131":38,"_132":133,"_148":149,"_94":150},["D",1781303607735],"children",[],[119],["D",1780142012597],["D",1789471840176],{"_11":12,"_92":88},{"_11":12,"_85":86,"_87":88,"_45":46,"_89":90,"_91":38},{"_29":30,"_13":26,"_20":21,"_27":28,"_31":-5,"_32":-5,"_80":-5,"_81":-5,"_82":-5,"_83":14,"_84":12},"ctx",{"_73":-5,"_114":189,"_190":191,"_192":193,"_194":38,"_195":38,"_196":38,"_197":50,"_198":-5,"_199":-7,"_200":-7,"_201":54,"_202":38,"_203":38,"_204":-5,"_205":38,"_206":38,"_207":38,"_208":-5},"ENV",{"_160":161,"_162":163,"_164":26,"_165":166,"_167":168,"_169":170,"_171":172,"_173":174,"_175":176,"_177":178,"_179":180,"_181":182,"_183":184,"_185":186,"_187":188},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","e87eb04e707cf64e7e532c8e8ec395f209e17f15","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]

Branches1

Branches
1