parsing ads works for duckduckgo, google, bing. tested.

This commit is contained in:
Nikolai Tschacher
2019-07-07 19:38:28 +02:00
parent bbebe3ce60
commit a413cb54ef
12 changed files with 211 additions and 178 deletions

View File

@ -1,10 +1,7 @@
const se_scraper = require('./../index.js');
var assert = require('chai').assert;
/*
* Use chai and mocha for tests.
* https://mochajs.org/#installation
*/
const chai = require('chai');
chai.use(require('chai-string'));
const assert = chai.assert;
const normal_search_keywords = ['apple tree', 'weather tomorrow'];
@ -12,12 +9,9 @@ async function normal_search_test() {
let config = {
compress: false,
debug_level: 1,
keyword_file: '',
headless: true,
output_file: '',
block_assets: true,
user_agent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.110 Safari/537.36',
random_user_agent: false,
block_assets: false,
random_user_agent: true,
};
let scrape_config = {
@ -83,12 +77,9 @@ async function effective_query_test() {
let config = {
compress: false,
debug_level: 1,
keyword_file: '',
headless: true,
output_file: '',
block_assets: true,
user_agent: 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.110 Safari/537.36',
random_user_agent: false,
random_user_agent: true,
};
let scrape_config = {
@ -129,7 +120,72 @@ function test_case_effective_query(response) {
}
}
(async () => {
await normal_search_test();
await effective_query_test();
})();
const ads_keywords = ['cloud services', 'buy shoes'];
async function ads_test() {
let config = {
compress: false,
debug_level: 1,
headless: true,
block_assets: false,
random_user_agent: false,
};
let scrape_config = {
search_engine: 'duckduckgo',
keywords: ads_keywords,
num_pages: 1,
};
console.log('ads_test()');
test_case_ads_test( await se_scraper.scrape(config, scrape_config) );
}
function test_case_ads_test(response) {
assert.equal(response.metadata.num_requests, 2);
for (let query in response.results) {
assert.containsAllKeys(response.results, ads_keywords, 'not all keywords were scraped.');
for (let page_number in response.results[query]) {
assert.isNumber(parseInt(page_number), 'page_number must be numeric');
let obj = response.results[query][page_number];
assert.containsAllKeys(obj, ['results', 'time', 'effective_query', 'ads'], 'not all keys are in the object');
assert.isAtLeast(obj.results.length, 6, 'results must have at least 6 SERP objects');
assert.typeOf(Date.parse(obj.time), 'number', 'time should be a valid date');
assert.isAtLeast(obj.ads.length, 2, 'ads must have at least 2 SERP object');
for (let res of obj.ads) {
assert.isOk(res.tracking_link, 'link must be ok');
assert.typeOf(res.tracking_link, 'string', 'link must be string');
assert.isAtLeast(res.tracking_link.length, 5, 'link must have at least 5 chars');
assert.isOk(res.visible_link, 'visible_link must be ok');
assert.typeOf(res.visible_link, 'string', 'visible_link must be string');
assert.isAtLeast(res.visible_link.length, 5, 'visible_link must have at least 5 chars');
assert.isOk(res.title, 'title must be ok');
assert.typeOf(res.title, 'string', 'title must be string');
assert.isAtLeast(res.title.length, 10, 'title must have at least 10 chars');
assert.isOk(res.snippet, 'snippet must be ok');
assert.typeOf(res.snippet, 'string', 'snippet must be string');
assert.isAtLeast(res.snippet.length, 10, 'snippet must have at least 10 chars');
}
}
}
}
describe('Duckduckgo', function(){
this.timeout(30000);
it('normal search', normal_search_test);
it('effective query', effective_query_test);
it('finds ads', ads_test);
});