github/v-firecrawl
github/v-firecrawl
0
Fork 0
Code Actions

Moved to utils/removeUnwantedElements, added unit tests

Browse Source
This commit is contained in:
rafaelsideguide 2024-06-18 09:46:42 -03:00
parent 8b3c3aae91
commit 6c726a02eb
3 changed files with 105 additions and 39 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
apps/api/src/scraper/WebScraper/single_url.ts
View File

@ -4,10 +4,10 @@ import { extractMetadata } from "./utils/metadata";
import dotenv from "dotenv"; import dotenv from "dotenv";
import { Document, PageOptions, FireEngineResponse } from "../../lib/entities"; import { Document, PageOptions, FireEngineResponse } from "../../lib/entities";
import { parseMarkdown } from "../../lib/html-to-markdown"; import { parseMarkdown } from "../../lib/html-to-markdown";
import { excludeNonMainTags } from "./utils/excludeTags";
import { urlSpecificParams } from "./utils/custom/website_params"; import { urlSpecificParams } from "./utils/custom/website_params";
import { fetchAndProcessPdf } from "./utils/pdfProcessor"; import { fetchAndProcessPdf } from "./utils/pdfProcessor";
import { handleCustomScraping } from "./custom/handleCustomScraping"; import { handleCustomScraping } from "./custom/handleCustomScraping";
import { removeUnwantedElements } from "./utils/removeUnwantedElements";
import axios from "axios"; import axios from "axios";
dotenv.config(); dotenv.config();
@ -313,44 +313,6 @@ export async function scrapSingleUrl(
): Promise<Document> { ): Promise<Document> {
urlToScrap = urlToScrap.trim(); urlToScrap = urlToScrap.trim();
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 = [pageOptions.removeTags];
}
if (Array.isArray(pageOptions.removeTags)) {
pageOptions.removeTags.forEach((tag) => {
let elementsToRemove;
if (tag.startsWith("*") && tag.endsWith("*")) {
const regexPattern = new RegExp(`\\b${tag.slice(1, -1)}\\b`);
elementsToRemove = soup('*').filter((index, element) => {
const classNames = soup(element).attr('class');
return classNames && classNames.split(/\s+/).some(className => regexPattern.test(className));
});
} else {
elementsToRemove = soup(tag);
}
elementsToRemove.remove();
});
}
}
if (pageOptions.onlyMainContent) {
// remove any other tags that are not in the main content
excludeNonMainTags.forEach((tag) => {
const elementsToRemove = soup(tag);
elementsToRemove.remove();
});
}
const cleanedHtml = soup.html();
return cleanedHtml;
};
const attemptScraping = async ( const attemptScraping = async (
url: string, url: string,
method: (typeof baseScrapers)[number] method: (typeof baseScrapers)[number]

63
apps/api/src/scraper/WebScraper/utils/__tests__/removeUnwantedElements.test.ts Normal file
View File

@ -0,0 +1,63 @@
import { removeUnwantedElements } from "../removeUnwantedElements";
import { PageOptions } from "../../../../lib/entities";
describe('removeUnwantedElements', () => {
it('should remove script, style, iframe, noscript, meta, and head tags', () => {
const html = `<html><head><title>Test</title></head><body><script>alert('test');</script><div>Content</div></body></html>`;
const options: PageOptions = {};
const result = removeUnwantedElements(html, options);
expect(result).not.toContain('<script>');
expect(result).not.toContain('<head>');
expect(result).toContain('Content');
});
it('should remove specified tags passed as string', () => {
const html = `<div><span>Remove</span><p>Keep</p></div>`;
const options: PageOptions = { removeTags: 'span' };
const result = removeUnwantedElements(html, options);
expect(result).not.toContain('<span>');
expect(result).toContain('<p>Keep</p>');
});
it('should remove specified tags passed as array', () => {
const html = `<div><span>Remove</span><p>Remove</p><a>Keep</a></div>`;
const options: PageOptions = { removeTags: ['span', 'p'] };
const result = removeUnwantedElements(html, options);
expect(result).not.toContain('<span>');
expect(result).not.toContain('<p>');
expect(result).toContain('<a>Keep</a>');
});
it('should handle class selectors', () => {
const html = `<div class="test">Remove</div><div class="keep">Keep</div>`;
const options: PageOptions = { removeTags: '.test' };
const result = removeUnwantedElements(html, options);
expect(result).not.toContain('class="test"');
expect(result).toContain('class="keep"');
});
it('should handle id selectors', () => {
const html = `<div id="test">Remove</div><div id="keep">Keep</div>`;
const options: PageOptions = { removeTags: '#test' };
const result = removeUnwantedElements(html, options);
expect(result).not.toContain('id="test"');
expect(result).toContain('id="keep"');
});
it('should handle regex patterns in class names', () => {
const html = `<div class="test-123">Remove</div><div class="test-abc">Remove</div><div class="keep">Keep</div>`;
const options: PageOptions = { removeTags: ['*test*'] };
const result = removeUnwantedElements(html, options);
expect(result).not.toContain('class="test-123"');
expect(result).not.toContain('class="test-abc"');
expect(result).toContain('class="keep"');
});
it('should remove non-main content if onlyMainContent is true', () => {
const html = `<div><main>Main Content</main><aside>Remove</aside></div>`;
const options: PageOptions = { onlyMainContent: true };
const result = removeUnwantedElements(html, options);
expect(result).toContain('Main Content');
expect(result).not.toContain('<aside>');
});
});

41
apps/api/src/scraper/WebScraper/utils/removeUnwantedElements.ts Normal file
View File

@ -0,0 +1,41 @@
import cheerio, { AnyNode, Cheerio } from "cheerio";
import { PageOptions } from "../../../lib/entities";
import { excludeNonMainTags } from "./excludeTags";
export 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 = [pageOptions.removeTags];
}
if (Array.isArray(pageOptions.removeTags)) {
pageOptions.removeTags.forEach((tag) => {
let elementsToRemove: Cheerio<AnyNode>;
if (tag.startsWith("*") && tag.endsWith("*")) {
const regexPattern = new RegExp(`\\b${tag.slice(1, -1)}\\b`);
elementsToRemove = soup('*').filter((index, element) => {
const classNames = soup(element).attr('class');
return classNames && classNames.split(/\s+/).some(className => regexPattern.test(className));
});
} else {
elementsToRemove = soup(tag);
}
elementsToRemove.remove();
});
}
}
if (pageOptions.onlyMainContent) {
// remove any other tags that are not in the main content
excludeNonMainTags.forEach((tag) => {
const elementsToRemove = soup(tag);
elementsToRemove.remove();
});
}
const cleanedHtml = soup.html();
return cleanedHtml;
};