{ "openapi": "3.0.1", "info": { "title": "Google News Scraper · Full Article Bodies + Entities", "description": "Scrape Google News with full publisher article bodies, not just snippets. Decodes the CBM redirect to the publisher page, extracts Article JSON-LD (title, body, author, image, keywords), and runs entity extraction (orgs/tickers/locations). 4 input kinds. Flat per-article billing. Pure HTTP", "version": "0.1", "x-build-id": "7CVlrDTKPcQUiJEas" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/memo23~google-news-scraper/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-memo23-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": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/memo23~google-news-scraper/runs": { "post": { "operationId": "runs-sync-memo23-google-news-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/memo23~google-news-scraper/run-sync": { "post": { "operationId": "run-sync-memo23-google-news-scraper", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$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", "required": [ "startInputs" ], "properties": { "startInputs": { "title": "Inputs (any mix)", "type": "array", "description": "Each entry is auto-classified:\n\n• **Search query** — bare string, e.g. `OpenAI earnings`\n• **Predefined topic name** — `BUSINESS`, `TECHNOLOGY`, `WORLD`, `NATION`, `ENTERTAINMENT`, `SPORTS`, `SCIENCE`, `HEALTH` (case-insensitive)\n• **Topic URL** — `https://news.google.com/topics/CAAqJ…`\n• **Publisher domain** — bare hostname, e.g. `bloomberg.com`\n• **Full-coverage cluster URL** — `https://news.google.com/stories/CA…`\n\nGoogle search operators work too: `-trump`, `\"exact phrase\"`, `site:wsj.com`, `OpenAI OR Anthropic`.", "items": { "type": "string" } }, "maxItems": { "title": "Max total rows", "minimum": 1, "type": "integer", "description": "Safety cap on the entire run. Each article = one row. Default 500.", "default": 500 }, "maxArticlesPerInput": { "title": "Max articles per input", "minimum": 0, "type": "integer", "description": "Per-input cap. Google News RSS feeds return up to ~100 articles per query. Default 100; set 0 for no per-input cap (overall `maxItems` still applies).", "default": 100 }, "resolveUrls": { "title": "Resolve Google News redirect → publisher URL", "type": "boolean", "description": "Decode each Google News CBM redirect to its real publisher URL using the same batchexecute endpoint the Google News client uses. One HTTP hop per article. Cheaper than `enrichBody`. Implied by `enrichBody:true`. Charges a `url-resolved` event.", "default": false }, "enrichBody": { "title": "Enrich with publisher article body", "type": "boolean", "description": "ALSO fetch the publisher page and extract the full body, headline, image, author, section, keywords, wordCount from Article/NewsArticle JSON-LD (with Mozilla Readability as the fallback algorithm — the same one Firefox Reader View uses). Adds a second HTTP hop per article on top of url-resolved. Charges a `body-enriched` event. Recommended for LLM/RAG buyers.", "default": false }, "enableCfBypass": { "title": "Cloudflare-bypass for hard-anti-bot publishers", "type": "boolean", "description": "When impit Firefox + Chrome both fail on a publisher (Bloomberg, NYT, WSJ, FT, sometimes Yahoo Finance), fall back to a CF-bypass service for the final attempt. No setup on your side — the bypass token is built into the actor, and there is **no extra charge** (the cost is absorbed into the base per-article price).\n\n**Defaults to ON whenever `enrichBody` is enabled** (the publishers worth reading sit behind Cloudflare, so bodies should just arrive). Set to `false` to skip walled publishers.", "default": true }, "extractEntities": { "title": "Extract named entities (people / orgs / tickers / locations)", "type": "boolean", "description": "Rule-based entity extraction on the article body or title. Cheap, no extra HTTP cost. Defaults to ON when `enrichBody` is enabled (body gives much better recall than title alone).", "default": false }, "regionLanguage": { "title": "Region + language (combined)", "type": "string", "description": "Convenience input — combined Google News locale token. Examples: `US:en`, `GB:en`, `DE:de`, `JP:ja`, `FR:fr`, `ES:es`, `IN:en`, `AU:en`, `BR:pt`. When set, overrides separate `country` + `language` below.", "default": "US:en" }, "timeframe": { "title": "Quick-select timeframe", "enum": [ "all", "last_hour", "last_day", "last_week", "last_month", "last_year" ], "type": "string", "description": "Restrict to articles from this window via Google News `when:` operator. Applies to search queries + publisher inputs; topic / cluster feeds ignore this (their windows are fixed by Google). Use `since` / `until` for exact date ranges.", "default": "all" }, "country": { "title": "Country (Google News gl=) — advanced override", "type": "string", "description": "ISO country code. Used only when `regionLanguage` isn't set. Examples: `US`, `GB`, `DE`, `IN`, `AU`. Default `US`.", "default": "US" }, "language": { "title": "Language (Google News hl=) — advanced override", "type": "string", "description": "ISO language code. Used only when `regionLanguage` isn't set. Examples: `en`, `de`, `ja`, `es`, `fr`. Default `en`.", "default": "en" }, "since": { "title": "Drop articles older than (ISO date)", "type": "string", "description": "Optional. Skip articles with `publishedAt < since`. Format: `YYYY-MM-DD` or full ISO 8601." }, "until": { "title": "Drop articles newer than (ISO date)", "type": "string", "description": "Optional. Skip articles with `publishedAt > until`. Format: `YYYY-MM-DD` or full ISO 8601." }, "proxy": { "title": "Proxy configuration", "type": "object", "description": "Apify Residential US is the default and is sufficient — Google News + most publisher sites are open to standard residential traffic.", "default": { "useApifyProxy": true, "apifyProxyGroups": [ "RESIDENTIAL" ] } }, "maxRequestRetries": { "title": "Max request retries", "minimum": 0, "type": "integer", "description": "Reserved. Maximum HTTP retry attempts per URL.", "default": 4 } } }, "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 } } } } } } } } } }