@opencraw/studio 0.0.39 → 0.0.41

package/dist/index.esm.js

@@ -3,7 +3,7 @@ import { createServer } from 'node:http';
 import { promisify, stripVTControlCharacters } from 'node:util';
 import { fileURLToPath, pathToFileURL } from 'node:url';
 import { z } from 'zod';
-import { BODY_KINDS, formatOfResponse, HttpClient, BrowserClient, NODE_ID_ATTRIBUTE, sessionStateOf, readRecipeSource, parseInputRecipe, parseOutputRecipe, calloutLabel, bindRecipeSet, memorySink, createCrawler, traceLine, UnknownHookError, rowsOfCells, parseXml, marginRows, findDeckTables, findGridTables, analyzeTables, parseRegion, regionText, deckRegionText, isDeckDocument, resolveMissingPolicy, isWorkbookDocument, RecipeBindingError, RecipeValidationError, countMatches, matchesOf, tryParseJson, isPdfDocument, BrowserProfiles } from '@opencraw/core';
+import { BODY_KINDS, videoIdOf, formatOfResponse, HttpClient, BrowserClient, NODE_ID_ATTRIBUTE, sessionStateOf, readRecipeSource, parseInputRecipe, parseOutputRecipe, calloutLabel, bindRecipeSet, memorySink, createCrawler, traceLine, UnknownHookError, rowsOfCells, parseXml, marginRows, findDeckTables, findGridTables, analyzeTables, parseRegion, regionText, deckRegionText, isDeckDocument, resolveMissingPolicy, isWorkbookDocument, RecipeBindingError, RecipeValidationError, countMatches, matchesOf, tryParseJson, isPdfDocument, BrowserProfiles } from '@opencraw/core';
 import { randomBytes, randomUUID, createHash } from 'node:crypto';
 import { execFile } from 'node:child_process';
 import { mkdir, writeFile, chmod, rm, readFile, rename, stat } from 'node:fs/promises';
@@ -679,7 +679,7 @@ const sampleRunViewSchema = z.object({
  * browser); `json`, `pdf`, `csv`, `xlsx`, `markdown`, `xml`... are documents (read with a request); `unknown` when it could not be fetched.
  */
 const detectUrlViewSchema = z.object({
-  kind: z.enum(['html', 'json', 'jsonl', 'text', 'pdf', 'csv', 'xlsx', 'pptx', 'yaml', 'markdown', 'xml', 'docx', 'unknown'])
+  kind: z.enum(['html', 'json', 'jsonl', 'text', 'pdf', 'csv', 'xlsx', 'pptx', 'yaml', 'markdown', 'xml', 'docx', 'youtube', 'unknown'])
 });
 
 /** What `fetch-start-page` answers with. */
@@ -1202,12 +1202,14 @@ const TIMEOUT_MS = 20_000;
  * What an address returns: the format the engine reads its response as (content type, then what the first bytes
  * show, then the file extension in the address), and, for an API that serves JSON as `text/plain` (common), JSON when
  * the text starts like an object or an array. Only the start of the response is read: a large file is not downloaded.
- * An address that cannot be fetched is `unknown`; the recipe started from it then reports why on its first snapshot.
+ * The address of a YouTube video is `youtube` without being fetched. An address that cannot be fetched is `unknown`; the recipe started from it then reports why on its first snapshot.
  *
  * @param url - The address.
  * @returns The kind: `html` is a page to open in a browser, anything else a document to read.
  */
 async function detectUrlKind(url) {
+  // A video's address is read as a video (its details and transcript), not fetched as the page it opens in a browser.
+  if (/^https?:/i.test(url.trim()) && videoIdOf(url) !== undefined) return 'youtube';
   const controller = new AbortController();
   const timer = setTimeout(() => {
     controller.abort();

package/dist/page-snapshot/detect-url.use-case.d.ts

@@ -1,10 +1,10 @@
 /** What an address returns, in the engine's own words (`html` is a page; anything else is a document it reads), or `unknown` when it cannot be fetched. */
-export type UrlKind = 'html' | 'json' | 'jsonl' | 'text' | 'pdf' | 'csv' | 'xlsx' | 'pptx' | 'yaml' | 'markdown' | 'xml' | 'docx' | 'unknown';
+export type UrlKind = 'html' | 'json' | 'jsonl' | 'text' | 'pdf' | 'csv' | 'xlsx' | 'pptx' | 'yaml' | 'markdown' | 'xml' | 'docx' | 'youtube' | 'unknown';
 /**
  * What an address returns: the format the engine reads its response as (content type, then what the first bytes
  * show, then the file extension in the address), and, for an API that serves JSON as `text/plain` (common), JSON when
  * the text starts like an object or an array. Only the start of the response is read: a large file is not downloaded.
- * An address that cannot be fetched is `unknown`; the recipe started from it then reports why on its first snapshot.
+ * The address of a YouTube video is `youtube` without being fetched. An address that cannot be fetched is `unknown`; the recipe started from it then reports why on its first snapshot.
  *
  * @param url - The address.
  * @returns The kind: `html` is a page to open in a browser, anything else a document to read.

package/dist/src/page-snapshot/detect-url.use-case.d.ts

@@ -1,10 +1,10 @@
 /** What an address returns, in the engine's own words (`html` is a page; anything else is a document it reads), or `unknown` when it cannot be fetched. */
-export type UrlKind = 'html' | 'json' | 'jsonl' | 'text' | 'pdf' | 'csv' | 'xlsx' | 'pptx' | 'yaml' | 'markdown' | 'xml' | 'docx' | 'unknown';
+export type UrlKind = 'html' | 'json' | 'jsonl' | 'text' | 'pdf' | 'csv' | 'xlsx' | 'pptx' | 'yaml' | 'markdown' | 'xml' | 'docx' | 'youtube' | 'unknown';
 /**
  * What an address returns: the format the engine reads its response as (content type, then what the first bytes
  * show, then the file extension in the address), and, for an API that serves JSON as `text/plain` (common), JSON when
  * the text starts like an object or an array. Only the start of the response is read: a large file is not downloaded.
- * An address that cannot be fetched is `unknown`; the recipe started from it then reports why on its first snapshot.
+ * The address of a YouTube video is `youtube` without being fetched. An address that cannot be fetched is `unknown`; the recipe started from it then reports why on its first snapshot.
  *
  * @param url - The address.
  * @returns The kind: `html` is a page to open in a browser, anything else a document to read.

package/dist/src/studio-api/detect-url-view.contract.d.ts

@@ -18,6 +18,7 @@ export declare const detectUrlViewSchema: z.ZodObject<{
         markdown: "markdown";
         xml: "xml";
         docx: "docx";
+        youtube: "youtube";
     }>;
 }, z.core.$strip>;
 export type DetectUrlView = z.infer<typeof detectUrlViewSchema>;

package/dist/src/studio-api/snapshot-view.contract.d.ts

@@ -24,6 +24,7 @@ export declare const snapshotViewSchema: z.ZodObject<{
         markdown: "markdown";
         xml: "xml";
         docx: "docx";
+        youtube: "youtube";
     }>>;
     notice: z.ZodOptional<z.ZodString>;
 }, z.core.$strip>;

package/dist/studio-api/detect-url-view.contract.d.ts

@@ -18,6 +18,7 @@ export declare const detectUrlViewSchema: z.ZodObject<{
         xlsx: "xlsx";
         xml: "xml";
         yaml: "yaml";
+        youtube: "youtube";
     }>;
 }, z.core.$strip>;
 export type DetectUrlView = z.infer<typeof detectUrlViewSchema>;

package/dist/studio-api/snapshot-view.contract.d.ts

@@ -24,6 +24,7 @@ export declare const snapshotViewSchema: z.ZodObject<{
         xlsx: "xlsx";
         xml: "xml";
         yaml: "yaml";
+        youtube: "youtube";
     }>>;
     notice: z.ZodOptional<z.ZodString>;
 }, z.core.$strip>;