{ "openapi": "3.0.1", "info": { "title": "Google News Scraper — Multi-Query Topic & Brand Monitoring", "description": "Google News RSS scraper with date-range filters, publisher filters, 30 language editions, optional enrichment (real article URL + og:image decoded from the Google News redirect) and FREE webhook + Telegram alerts.", "version": "0.5", "x-build-id": "qPhmM0kYzwnyUReM4" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/halobartku~google-news-scraper/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-halobartku-google-news-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/halobartku~google-news-scraper/runs": { "post": { "operationId": "runs-sync-halobartku-google-news-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/halobartku~google-news-scraper/run-sync": { "post": { "operationId": "run-sync-halobartku-google-news-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } } }, "components": { "schemas": { "inputSchema": { "type": "object", "properties": { "queries": { "title": "Topic queries", "maxItems": 50, "type": "array", "description": "Topics, brands, competitors or exact-phrase searches (use quotes). Each harvested query is one paid event. Leave empty if you only want section feeds from Topics.", "items": { "type": "string" } }, "topics": { "title": "Google News sections (optional)", "type": "array", "description": "Harvest Google News' own curated section feeds instead of (or alongside) your queries. Each harvested section is one paid event, same as a query. Sections return the last ~2 days of headlines and ignore Time range and the query filters.", "items": { "type": "string", "enum": [ "TOP", "WORLD", "NATION", "BUSINESS", "TECHNOLOGY", "ENTERTAINMENT", "SPORTS", "SCIENCE", "HEALTH" ], "enumTitles": [ "Top stories", "World", "Nation", "Business", "Technology", "Entertainment", "Sports", "Science", "Health" ] }, "default": [] }, "timeRange": { "title": "Time range", "enum": [ "", "1h", "6h", "12h", "1d", "3d", "7d", "14d", "30d", "1y" ], "type": "string", "description": "Only return articles published inside this window (Google News 'when:' operator). Ignored when Date from / Date to are set. Note: Google returns zero results for month-style windows, so only the windows that actually work are offered here.", "default": "" }, "dateFrom": { "title": "Date from (YYYY-MM-DD)", "type": "string", "description": "Only articles published on or after this date. Overrides Time range. Format YYYY-MM-DD." }, "dateTo": { "title": "Date to (YYYY-MM-DD)", "type": "string", "description": "Only articles published on or before this date. Use together with Date from. Format YYYY-MM-DD." }, "siteFilter": { "title": "Only these domains", "type": "array", "description": "Restrict results to these publisher domains, e.g. reuters.com, bbc.com. Several domains are OR-ed together.", "default": [], "items": { "type": "string" } }, "excludeWords": { "title": "Exclude words", "type": "array", "description": "Drop articles containing these words or phrases. Applied by Google's own '-word' operator, so excluded articles are never fetched.", "default": [], "items": { "type": "string" } }, "language": { "title": "Language (hl)", "enum": [ "en", "de", "fr", "es", "pl", "it", "pt", "nl", "ja", "ko", "zh", "sv", "da", "nb", "fi", "cs", "sk", "hu", "ro", "el", "tr", "ru", "uk", "ar", "he", "hi", "id", "th", "vi", "bg" ], "type": "string", "description": "News edition language. Country and edition default to the matching country unless you set them below.", "default": "en" }, "geo": { "title": "Country (gl)", "enum": [ "", "us", "gb", "ca", "au", "ie", "in", "sg", "za", "nz", "ng", "ph", "de", "at", "ch", "fr", "be", "es", "mx", "ar", "co", "cl", "it", "pt", "br", "nl", "pl", "se", "dk", "no", "fi", "cz", "sk", "hu", "ro", "gr", "tr", "ru", "ua", "il", "eg", "sa", "ae", "jp", "kr", "cn", "tw", "hk", "id", "th", "vn", "my" ], "type": "string", "description": "Country edition. Leave default to use the language's home country. English + gb/in/au/ca/ie/sg/za gives you the local English edition.", "default": "" }, "regionLanguage": { "title": "Raw edition override (ceid)", "type": "string", "description": "Power users only. Google News edition token in COUNTRY:lang form, e.g. GB:en, BR:pt-419, TW:zh-Hant. Overrides Language and Country when set.", "default": "" }, "sortBy": { "title": "Sort by", "enum": [ "relevance", "newest", "oldest" ], "type": "string", "description": "Google's feed order is relevance, not time. Pick Newest first to get a chronological feed — and to make 'Max items per query' keep the newest items rather than the top-ranked ones.", "default": "relevance" }, "maxPerQuery": { "title": "Max items per query", "minimum": 1, "maximum": 1000, "type": "integer", "description": "Cap rows per query or section. A typical topic returns 20-100 items; sections return 38-70.", "default": 100 }, "maxTotal": { "title": "Max items for the whole run", "minimum": 0, "maximum": 50000, "type": "integer", "description": "Hard cap across all queries and sections. 0 = no cap. Once the cap is full the remaining entries are skipped and NOT charged.", "default": 0 }, "webhookUrl": { "title": "Webhook URL (free notifications)", "type": "string", "description": "Optional. POSTs a JSON run summary to this URL on EVERY finished run (including empty ones — heartbeat for monitoring pipelines). Free, fail-soft: a broken webhook never fails your run.", "default": "" }, "telegramBotToken": { "title": "Telegram bot token", "type": "string", "description": "Optional. Bot token from @BotFather. A human-readable digest (top headlines) is sent only when there is something to read. Pair with telegramChatId. Free, fail-soft.", "default": "" }, "telegramChatId": { "title": "Telegram chat ID", "type": "string", "description": "Optional. Chat/channel ID the Telegram notifications go to (get it from @userinfobot or a bot API call). Ignored when telegramBotToken is empty.", "default": "" }, "enrich": { "title": "Enrich: resolve real article URLs + images", "type": "boolean", "description": "Optional. Decode Google News redirect URLs to the real publisher article URL and pull og:image + og:description from the article page (article_url, image_url, article_description fields). Fail-soft: rows that cannot be decoded or whose publisher blocks us are delivered with empty enrichment fields — never dropped, never charged extra. Adds ~2 requests per enriched row.", "default": false }, "enrichLimit": { "title": "Enrich row limit", "minimum": 0, "maximum": 1000, "type": "integer", "description": "How many rows (per entry, in delivery order) to enrich when enrich=true. Rows beyond the limit keep empty article_url/image_url/article_description. Default 10.", "default": 10 } } }, "runsResponseSchema": { "type": "object", "properties": { "data": { "type": "object", "properties": { "id": { "type": "string" }, "actId": { "type": "string" }, "userId": { "type": "string" }, "startedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "finishedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "status": { "type": "string", "example": "READY" }, "meta": { "type": "object", "properties": { "origin": { "type": "string", "example": "API" }, "userAgent": { "type": "string" } } }, "stats": { "type": "object", "properties": { "inputBodyLen": { "type": "integer", "example": 2000 }, "rebootCount": { "type": "integer", "example": 0 }, "restartCount": { "type": "integer", "example": 0 }, "resurrectCount": { "type": "integer", "example": 0 }, "computeUnits": { "type": "integer", "example": 0 } } }, "options": { "type": "object", "properties": { "build": { "type": "string", "example": "latest" }, "timeoutSecs": { "type": "integer", "example": 300 }, "memoryMbytes": { "type": "integer", "example": 1024 }, "diskMbytes": { "type": "integer", "example": 2048 } } }, "buildId": { "type": "string" }, "defaultKeyValueStoreId": { "type": "string" }, "defaultDatasetId": { "type": "string" }, "defaultRequestQueueId": { "type": "string" }, "buildNumber": { "type": "string", "example": "1.0.0" }, "containerUrl": { "type": "string" }, "usage": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "integer", "example": 1 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 } } }, "usageTotalUsd": { "type": "number", "example": 0.00005 }, "usageUsd": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "number", "example": 0.00005 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 } } } } } } } } } }