2017-12-11 05:51:04 +01:00
|
|
|
// Copyright 2017 Frédéric Guillot. All rights reserved.
|
|
|
|
// Use of this source code is governed by the Apache 2.0
|
|
|
|
// license that can be found in the LICENSE file.
|
|
|
|
|
|
|
|
package scraper
|
|
|
|
|
|
|
|
// List of predefined scraper rules (alphabetically sorted)
|
|
|
|
// domain => CSS selectors
|
|
|
|
var predefinedRules = map[string]string{
|
2017-12-14 06:30:40 +01:00
|
|
|
"github.com": "article.entry-content",
|
|
|
|
"igen.fr": "section.corps",
|
2017-12-11 05:51:04 +01:00
|
|
|
"lemonde.fr": "div#articleBody",
|
|
|
|
"lesjoiesducode.fr": ".blog-post-content img",
|
|
|
|
"linux.com": "div.content, div[property]",
|
2017-12-14 06:30:40 +01:00
|
|
|
"medium.com": ".section-content",
|
2017-12-11 05:51:04 +01:00
|
|
|
"opensource.com": "div[property]",
|
2017-12-14 06:30:40 +01:00
|
|
|
"osnews.com": "div.newscontent1",
|
2017-12-11 05:51:04 +01:00
|
|
|
"phoronix.com": "div.content",
|
|
|
|
"techcrunch.com": "div.article-entry",
|
2017-12-14 06:30:40 +01:00
|
|
|
"theregister.co.uk": "#body",
|
|
|
|
"wired.com": "main figure, article",
|
2017-12-11 05:51:04 +01:00
|
|
|
}
|