browsertrix-crawler/tests/custom_driver.test.js
Ilya Kreymer 0e0b85d7c3
Customizable extract selectors + typo fix (0.4.2) (#72)
* fix typo in setting crawler.capturePrefix which caused directFetchCapture() to fail, causing non-HTML urls to fail.
- wrap directFetchCapture() to retry browser loading in case of failure

* custom link extraction improvements (improvements for #25) 
- extractLinks() returns a list of link URLs to allow for more flexibility in custom driver
- rename queueUrls() to queueInScopeUrls() to indicate the filtering is performed
- loadPage accepts a list of select opts {selector, extract, isAttribute} and defaults to {"a[href]", "href", false}
- tests: add test for custom driver which uses custom selector

* tests
- tests: all tests uses 'test-crawls' instead of crawls
- consolidation: combine initial crawl + rollover, combine warc, text tests into basic_crawl.test.js
- add custom driver test and fixture to test custom link extraction

* add to CHANGES, bump to 0.4.2
2021-07-23 18:31:43 -07:00

34 lines
997 B
JavaScript

const child_process = require("child_process");
const fs = require("fs");
test("ensure custom driver with custom selector crawls JS files as pages", async () => {
jest.setTimeout(30000);
try {
child_process.execSync("docker run -v $PWD/tests/fixtures:/tests/fixtures -v $PWD/test-crawls:/crawls webrecorder/browsertrix-crawler crawl --url https://www.iana.org/ --collection custom-driver-1 --driver /tests/fixtures/driver-1.js");
}
catch (error) {
console.log(error);
}
const crawledPages = fs.readFileSync("test-crawls/collections/custom-driver-1/pages/pages.jsonl", "utf8");
const pages = new Set();
for (const line of crawledPages.trim().split("\n")) {
const url = JSON.parse(line).url;
if (!url) {
continue;
}
pages.add(url);
}
const expectedPages = new Set([
"https://www.iana.org/",
"https://www.iana.org/_js/2013.1/jquery.js",
"https://www.iana.org/_js/2013.1/iana.js"
]);
expect(pages).toEqual(expectedPages);
});