From f48ff36b32978fb4d0c54dd7efcbaad5f22e9a5d Mon Sep 17 00:00:00 2001 From: rafaelsideguide <150964962+rafaelsideguide@users.noreply.github.com> Date: Wed, 31 Jul 2024 09:28:43 -0300 Subject: [PATCH] added .inc files and forced lower case comparison --- apps/api/src/scraper/WebScraper/crawler.ts | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/apps/api/src/scraper/WebScraper/crawler.ts b/apps/api/src/scraper/WebScraper/crawler.ts index 5ee8cda..60dc301 100644 --- a/apps/api/src/scraper/WebScraper/crawler.ts +++ b/apps/api/src/scraper/WebScraper/crawler.ts @@ -164,7 +164,7 @@ export class WebCrawler { concurrencyLimit, inProgress ); - + if ( urls.length === 0 && this.filterLinks([this.initialUrl], limit, this.maxCrawledDepth).length > 0 @@ -420,9 +420,10 @@ export class WebCrawler { ".woff", ".ttf", ".woff2", - ".webp" + ".webp", + ".inc" ]; - return fileExtensions.some((ext) => url.endsWith(ext)); + return fileExtensions.some((ext) => url.toLowerCase().endsWith(ext)); } private isSocialMediaOrEmail(url: string): boolean {