andreferraro
6e3d57619b
- Add standalone CLI script scripts/extract_google_news.py for Google News RSS scraping - Integrate foxcape in headless mode as primary stealth anti-bot engine - Implement parallel article URL resolution using googlenewsdecoder and ThreadPoolExecutor - Support language and regional locale mapping (-l, --lang, --locale) - Implement real-time progress logging in stderr and --silent flag - Add unit, integration, and live E2E tests in tests/test_extract_google_news.py - Add full SpecKit documentation (specs/002-google-news-extractor/) - Create comprehensive README.md covering both NLP Classifier and Google News Extractor
1 line
4.2 KiB
JSON
1 line
4.2 KiB
JSON
{"nodes": [{"id": "$graphify-root$_specs_002_google_news_extractor_data_model_md", "label": "data-model.md", "file_type": "document", "node_kind": "page", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L1"}, {"id": "$graphify-root$_specs_002_google_news_extractor_data_model_data_model_google_news_headlines_extractor", "label": "Data Model: Google News Headlines Extractor", "file_type": "document", "node_kind": "heading", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L1"}, {"id": "$graphify-root$_specs_002_google_news_extractor_data_model_1_entidades_de_dom\u00ednio_dtos", "label": "1. Entidades de Dom\u00ednio & DTOs", "file_type": "document", "node_kind": "heading", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L3"}, {"id": "$graphify-root$_specs_002_google_news_extractor_data_model_1_1_searchquery_par\u00e2metros_da_busca", "label": "1.1 SearchQuery (Par\u00e2metros da Busca)", "file_type": "document", "node_kind": "heading", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L5"}, {"id": "$graphify-root$_specs_002_google_news_extractor_data_model_1_2_newsarticle_item_de_not\u00edcia", "label": "1.2 NewsArticle (Item de Not\u00edcia)", "file_type": "document", "node_kind": "heading", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L15"}, {"id": "$graphify-root$_specs_002_google_news_extractor_data_model_1_3_extractionresult_sa\u00edda_estruturada_consolidada", "label": "1.3 ExtractionResult (Sa\u00edda Estruturada Consolidada)", "file_type": "document", "node_kind": "heading", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L26"}, {"id": "$graphify-root$_specs_002_google_news_extractor_data_model_2_esquema_json_de_sa\u00edda", "label": "2. Esquema JSON de Sa\u00edda", "file_type": "document", "node_kind": "heading", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L41"}], "edges": [{"source": "$graphify-root$_specs_002_google_news_extractor_data_model_md", "target": "$graphify-root$_specs_002_google_news_extractor_data_model_data_model_google_news_headlines_extractor", "relation": "contains", "confidence": "EXTRACTED", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L1", "weight": 1.0}, {"source": "$graphify-root$_specs_002_google_news_extractor_data_model_data_model_google_news_headlines_extractor", "target": "$graphify-root$_specs_002_google_news_extractor_data_model_1_entidades_de_dom\u00ednio_dtos", "relation": "contains", "confidence": "EXTRACTED", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L3", "weight": 1.0}, {"source": "$graphify-root$_specs_002_google_news_extractor_data_model_1_entidades_de_dom\u00ednio_dtos", "target": "$graphify-root$_specs_002_google_news_extractor_data_model_1_1_searchquery_par\u00e2metros_da_busca", "relation": "contains", "confidence": "EXTRACTED", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L5", "weight": 1.0}, {"source": "$graphify-root$_specs_002_google_news_extractor_data_model_1_entidades_de_dom\u00ednio_dtos", "target": "$graphify-root$_specs_002_google_news_extractor_data_model_1_2_newsarticle_item_de_not\u00edcia", "relation": "contains", "confidence": "EXTRACTED", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L15", "weight": 1.0}, {"source": "$graphify-root$_specs_002_google_news_extractor_data_model_1_entidades_de_dom\u00ednio_dtos", "target": "$graphify-root$_specs_002_google_news_extractor_data_model_1_3_extractionresult_sa\u00edda_estruturada_consolidada", "relation": "contains", "confidence": "EXTRACTED", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L26", "weight": 1.0}, {"source": "$graphify-root$_specs_002_google_news_extractor_data_model_data_model_google_news_headlines_extractor", "target": "$graphify-root$_specs_002_google_news_extractor_data_model_2_esquema_json_de_sa\u00edda", "relation": "contains", "confidence": "EXTRACTED", "source_file": "specs/002-google-news-extractor/data-model.md", "source_location": "L41", "weight": 1.0}], "input_tokens": 0, "output_tokens": 0} |