cryptorugmunch
8d25702eca
Squashed from chore/license-relicense. Full message preserved in the original branch commitbb77eb5. See ADR-0002 for the decision rationale. Refs: ADR-0002, commitbb77eb5
53 lines
No EOL
1.2 KiB
JSON
53 lines
No EOL
1.2 KiB
JSON
{
|
|
"name": "The Guardian Article",
|
|
"description": "Extract article content from The Guardian",
|
|
"site": "theguardian.com",
|
|
"category": "news",
|
|
"icon": "\ud83d\udcf0",
|
|
"schema": {
|
|
"name": "guardian_article",
|
|
"base_selector": "article",
|
|
"fields": [
|
|
{
|
|
"name": "title",
|
|
"selector": "h1",
|
|
"type": "text"
|
|
},
|
|
{
|
|
"name": "author",
|
|
"selector": "[class*='byline']",
|
|
"type": "text"
|
|
},
|
|
{
|
|
"name": "published_date",
|
|
"selector": "time, [class*='date']",
|
|
"type": "attribute",
|
|
"attribute": "datetime"
|
|
},
|
|
{
|
|
"name": "content",
|
|
"selector": "[class*='article-body']",
|
|
"type": "text"
|
|
},
|
|
{
|
|
"name": "summary",
|
|
"selector": "[class*='standfirst']",
|
|
"type": "text"
|
|
},
|
|
{
|
|
"name": "image_url",
|
|
"selector": "figure img",
|
|
"type": "attribute",
|
|
"attribute": "src"
|
|
},
|
|
{
|
|
"name": "section",
|
|
"selector": "[class*='section'] a",
|
|
"type": "text"
|
|
}
|
|
]
|
|
},
|
|
"url_pattern": "*://www.theguardian.com/*",
|
|
"_license": "MIT",
|
|
"_copyright": "Copyright (c) 2026 Rug Munch Media LLC"
|
|
} |