From 676d6e8ab5f7a1fd14ff5b76f8289db7543082c4 Mon Sep 17 00:00:00 2001 From: rafaelsideguide <150964962+rafaelsideguide@users.noreply.github.com> Date: Thu, 13 Jun 2024 10:51:05 -0300 Subject: [PATCH] Added pageOptions.removeTags --- apps/api/openapi.json | 19 +++++++++++ .../src/__tests__/e2e_withAuth/index.test.ts | 34 +++++++++++++++++++ apps/api/src/controllers/crawl.ts | 10 ++++-- apps/api/src/controllers/crawlPreview.ts | 2 +- apps/api/src/controllers/search.ts | 2 ++ apps/api/src/lib/entities.ts | 1 + apps/api/src/scraper/WebScraper/index.ts | 7 +++- apps/api/src/scraper/WebScraper/single_url.ts | 13 +++++++ 8 files changed, 84 insertions(+), 4 deletions(-) diff --git a/apps/api/openapi.json b/apps/api/openapi.json index a755e37..b07e43f 100644 --- a/apps/api/openapi.json +++ b/apps/api/openapi.json @@ -61,6 +61,13 @@ "description": "Wait x amount of milliseconds for the page to load to fetch content", "default": 0 }, + "removeTags": { + "type": "array", + "items": { + "type": "string" + }, + "description": "Tags, classes and ids to remove from the page. Use comma separated values. Example: 'script, .ad, #footer'" + }, "headers": { "type": "object", "description": "Headers to send with the request. Can be used to send cookies, user-agent, etc." @@ -194,6 +201,11 @@ "type": "integer", "description": "Maximum number of pages to crawl", "default": 10000 + }, + "allowBackwardCrawling": { + "type": "boolean", + "description": "Allow backward crawling (crawl from the base URL to the previous URLs)", + "default": false } } }, @@ -219,6 +231,13 @@ "type": "object", "description": "Headers to send with the request when scraping. Can be used to send cookies, user-agent, etc." }, + "removeTags": { + "type": "array", + "items": { + "type": "string" + }, + "description": "Tags, classes and ids to remove from the page. Use comma separated values. Example: 'script, .ad, #footer'" + }, "replaceAllPathsWithAbsolutePaths": { "type": "boolean", "description": "Replace all relative paths with absolute paths for images and links", diff --git a/apps/api/src/__tests__/e2e_withAuth/index.test.ts b/apps/api/src/__tests__/e2e_withAuth/index.test.ts index 02e4a47..3423b3a 100644 --- a/apps/api/src/__tests__/e2e_withAuth/index.test.ts +++ b/apps/api/src/__tests__/e2e_withAuth/index.test.ts @@ -136,6 +136,40 @@ describe("E2E Tests for API Routes", () => { expect(response.body.data.content).toContain('We present spectrophotometric observations of the Broad Line Radio Galaxy'); }, 60000); // 60 seconds + it.concurrent("should return a successful response with a valid API key with removeTags option", async () => { + const responseWithoutRemoveTags = await request(TEST_URL) + .post("/v0/scrape") + .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`) + .set("Content-Type", "application/json") + .send({ url: "https://www.scrapethissite.com/" }); + expect(responseWithoutRemoveTags.statusCode).toBe(200); + expect(responseWithoutRemoveTags.body).toHaveProperty("data"); + expect(responseWithoutRemoveTags.body.data).toHaveProperty("content"); + expect(responseWithoutRemoveTags.body.data).toHaveProperty("markdown"); + expect(responseWithoutRemoveTags.body.data).toHaveProperty("metadata"); + expect(responseWithoutRemoveTags.body.data).not.toHaveProperty("html"); + expect(responseWithoutRemoveTags.body.data.content).toContain("Scrape This Site"); + expect(responseWithoutRemoveTags.body.data.content).toContain("Lessons and Videos"); // #footer + expect(responseWithoutRemoveTags.body.data.content).toContain("[Sandbox]("); // .nav + expect(responseWithoutRemoveTags.body.data.content).toContain("web scraping"); // strong + + const response = await request(TEST_URL) + .post("/v0/scrape") + .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`) + .set("Content-Type", "application/json") + .send({ url: "https://www.scrapethissite.com/", pageOptions: { removeTags: ['.nav', '#footer', 'strong'] } }); + expect(response.statusCode).toBe(200); + expect(response.body).toHaveProperty("data"); + expect(response.body.data).toHaveProperty("content"); + expect(response.body.data).toHaveProperty("markdown"); + expect(response.body.data).toHaveProperty("metadata"); + expect(response.body.data).not.toHaveProperty("html"); + expect(response.body.data.content).toContain("Scrape This Site"); + expect(response.body.data.content).not.toContain("Lessons and Videos"); // #footer + expect(response.body.data.content).not.toContain("[Sandbox]("); // .nav + expect(response.body.data.content).not.toContain("web scraping"); // strong + }, 30000); // 30 seconds timeout + // TODO: add this test back once we nail the waitFor option to be more deterministic // it.concurrent("should return a successful response with a valid API key and waitFor option", async () => { // const startTime = Date.now(); diff --git a/apps/api/src/controllers/crawl.ts b/apps/api/src/controllers/crawl.ts index 58d01e2..7eab78f 100644 --- a/apps/api/src/controllers/crawl.ts +++ b/apps/api/src/controllers/crawl.ts @@ -55,8 +55,14 @@ export async function crawlController(req: Request, res: Response) { } const mode = req.body.mode ?? "crawl"; - const crawlerOptions = req.body.crawlerOptions ?? { allowBackwardCrawling: false }; - const pageOptions = req.body.pageOptions ?? { onlyMainContent: false, includeHtml: false }; + const crawlerOptions = req.body.crawlerOptions ?? { + allowBackwardCrawling: false + }; + const pageOptions = req.body.pageOptions ?? { + onlyMainContent: false, + includeHtml: false, + removeTags: [] + }; if (mode === "single_urls" && !url.includes(",")) { try { diff --git a/apps/api/src/controllers/crawlPreview.ts b/apps/api/src/controllers/crawlPreview.ts index d3e9afe..2c3dc4e 100644 --- a/apps/api/src/controllers/crawlPreview.ts +++ b/apps/api/src/controllers/crawlPreview.ts @@ -26,7 +26,7 @@ export async function crawlPreviewController(req: Request, res: Response) { const mode = req.body.mode ?? "crawl"; const crawlerOptions = req.body.crawlerOptions ?? {}; - const pageOptions = req.body.pageOptions ?? { onlyMainContent: false, includeHtml: false }; + const pageOptions = req.body.pageOptions ?? { onlyMainContent: false, includeHtml: false, removeTags: [] }; const job = await addWebScraperJob({ url: url, diff --git a/apps/api/src/controllers/search.ts b/apps/api/src/controllers/search.ts index 7474aae..abbc357 100644 --- a/apps/api/src/controllers/search.ts +++ b/apps/api/src/controllers/search.ts @@ -85,6 +85,7 @@ export async function searchHelper( onlyMainContent: pageOptions?.onlyMainContent ?? true, fetchPageContent: pageOptions?.fetchPageContent ?? true, includeHtml: pageOptions?.includeHtml ?? false, + removeTags: pageOptions?.removeTags ?? [], fallback: false, }, }); @@ -139,6 +140,7 @@ export async function searchController(req: Request, res: Response) { includeHtml: false, onlyMainContent: true, fetchPageContent: true, + removeTags: [], fallback: false, }; const origin = req.body.origin ?? "api"; diff --git a/apps/api/src/lib/entities.ts b/apps/api/src/lib/entities.ts index 81bf12c..92170c1 100644 --- a/apps/api/src/lib/entities.ts +++ b/apps/api/src/lib/entities.ts @@ -19,6 +19,7 @@ export type PageOptions = { screenshot?: boolean; headers?: Record; replaceAllPathsWithAbsolutePaths?: boolean; + removeTags?: string | string[]; }; export type ExtractorOptions = { diff --git a/apps/api/src/scraper/WebScraper/index.ts b/apps/api/src/scraper/WebScraper/index.ts index f432f43..1a6ffd0 100644 --- a/apps/api/src/scraper/WebScraper/index.ts +++ b/apps/api/src/scraper/WebScraper/index.ts @@ -475,7 +475,12 @@ export class WebScraperDataProvider { this.limit = options.crawlerOptions?.limit ?? 10000; this.generateImgAltText = options.crawlerOptions?.generateImgAltText ?? false; - this.pageOptions = options.pageOptions ?? { onlyMainContent: false, includeHtml: false, replaceAllPathsWithAbsolutePaths: false }; + this.pageOptions = options.pageOptions ?? { + onlyMainContent: false, + includeHtml: false, + replaceAllPathsWithAbsolutePaths: false, + removeTags: [] + }; this.extractorOptions = options.extractorOptions ?? {mode: "markdown"} this.replaceAllPathsWithAbsolutePaths = options.crawlerOptions?.replaceAllPathsWithAbsolutePaths ?? options.pageOptions?.replaceAllPathsWithAbsolutePaths ?? false; //! @nicolas, for some reason this was being injected and breaking everything. Don't have time to find source of the issue so adding this check diff --git a/apps/api/src/scraper/WebScraper/single_url.ts b/apps/api/src/scraper/WebScraper/single_url.ts index c2dcea1..a16f6f0 100644 --- a/apps/api/src/scraper/WebScraper/single_url.ts +++ b/apps/api/src/scraper/WebScraper/single_url.ts @@ -304,6 +304,19 @@ export async function scrapSingleUrl( const removeUnwantedElements = (html: string, pageOptions: PageOptions) => { const soup = cheerio.load(html); soup("script, style, iframe, noscript, meta, head").remove(); + + if (pageOptions.removeTags) { + if (typeof pageOptions.removeTags === 'string') { + pageOptions.removeTags.split(',').forEach((tag) => { + soup(tag.trim()).remove(); + }); + } else if (Array.isArray(pageOptions.removeTags)) { + pageOptions.removeTags.forEach((tag) => { + soup(tag).remove(); + }); + } + } + if (pageOptions.onlyMainContent) { // remove any other tags that are not in the main content excludeNonMainTags.forEach((tag) => {