AI/firecrawl
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/mendableai/firecrawl synced 2025-08-14 02:35:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix: treat existing screenshot as a scraper success condition

Browse Source
This commit is contained in:
Gergő Móricz 2024-08-20 22:24:18 +02:00
parent 0c48c8a436
commit 1368f9a87f
2 changed files with 12 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
apps/api/src/controllers/v1/types.ts
View File

@ -279,6 +279,16 @@ export function legacyScrapeOptions(x: ScrapeOptions): PageOptions {
} }
export function legacyDocumentConverter(doc: any): Document { export function legacyDocumentConverter(doc: any): Document {
if (doc.metadata.screenshot) {
doc.screenshot = doc.metadata.screenshot;
delete doc.metadata.screenshot;
}
if (doc.metadata.fullPageScreenshot) {
doc.fullPageScreenshot = doc.metadata.fullPageScreenshot;
delete doc.metadata.fullPageScreenshot;
}
return { return {
markdown: doc.markdown, markdown: doc.markdown,
links: doc.linksOnPage, links: doc.linksOnPage,

4
apps/api/src/scraper/WebScraper/single_url.ts
View File

@ -340,8 +340,8 @@ export async function scrapSingleUrl(
pageError = undefined; pageError = undefined;
} }
if (text && text.trim().length >= 100) { if ((text && text.trim().length >= 100) || (typeof screenshot === "string" && screenshot.length > 0)) {
Logger.debug(`⛏️ ${scraper}: Successfully scraped ${urlToScrap} with text length >= 100, breaking`); Logger.debug(`⛏️ ${scraper}: Successfully scraped ${urlToScrap} with text length >= 100 or screenshot, breaking`);
break; break;
} }
if (pageStatusCode && (pageStatusCode == 404 || pageStatusCode == 500)) { if (pageStatusCode && (pageStatusCode == 404 || pageStatusCode == 500)) {