OrbTop

China CNIPA Patent Scraper

BUSINESSEDUCATIONAUTOMATION
Try on Apify

China CNIPA Patent Scraper

Scrape Chinese patent records from Google Patents with the CN jurisdiction filter applied. Returns bilingual data — English-translated titles and abstracts alongside the original Chinese — plus filing dates, assignees, inventors, CPC/IPC codes, legal event timelines, and patent family members.

China CNIPA Patent Scraper Features

  • Extracts 20 fields per patent including bilingual title and abstract
  • Search by keyword, IPC/CPC classification code, assignee, or inventor
  • Filter by filing date range with dateFrom and dateTo
  • Direct URL mode — paste specific Google Patents links and skip the search step entirely
  • Returns legal event timeline (filings, grants, status changes) as structured JSON
  • Includes patent family members so you can cross-reference equivalent filings in other jurisdictions
  • Residential proxy preconfigured — Google Patents serves a soft-block page to datacenter IPs

Who Uses Chinese Patent Data?

  • IP analysts — Track competitor filings, monitor classification clusters, build prior-art datasets
  • Corporate R&D teams — Map who's filing what in China, and when. Useful before you spend a year on the same idea.
  • Patent attorneys — Pull family members and legal-event histories for freedom-to-operate analysis
  • Market intelligence — Detect filing surges in specific CPC codes as a leading indicator of where Chinese R&D money is going
  • Academic researchers — Build longitudinal patent corpora filtered by date, assignee, or technology area

How the China CNIPA Patent Scraper Works

  1. Pick a mode — Provide a search query, or paste specific patent URLs into startUrls. URLs win when both are set.
  2. Search pagination — In search mode, the scraper hits Google Patents' JSON XHR endpoint with the country=CN filter and pages until it has enough patent IDs or hits Google's 1,000-result hard cap.
  3. Detail extraction — Each patent's detail page is fetched through residential proxy and parsed for bibliographic data, classification codes, legal events, and the Chinese-source text.
  4. Export — Records land in your Apify dataset as clean JSON, one record per patent.

Input

Search by keyword and date range

{
  "query": "battery electric vehicle",
  "dateFrom": "2022-01-01",
  "dateTo": "2024-12-31",
  "maxItems": 50
}

Search by classification code

{
  "query": "H04L",
  "maxItems": 100
}

Search by assignee

{
  "query": "Huawei",
  "maxItems": 200
}

Direct URL mode

{
  "startUrls": [
    "https://patents.google.com/patent/CN115982415A/en",
    "https://patents.google.com/patent/CN108081978B/en"
  ],
  "maxItems": 2
}
Field Type Default Description
query string battery electric vehicle Keywords, IPC/CPC code, assignee, or inventor. Empty string runs a broad CN patent search.
dateFrom string — Earliest priority date (YYYY-MM-DD).
dateTo string — Latest priority date (YYYY-MM-DD).
startUrls array [] Specific Google Patents URLs. When provided, the search API is skipped.
maxItems integer 10 Cap on patents returned. Google Patents enforces a 1,000-result limit per query.
proxyConfiguration object Apify Residential Proxy settings. Residential is required.

China CNIPA Patent Scraper Output Fields

{
  "publication_number": "CN115982415A",
  "application_number": "CN:202310093204.3A",
  "title_en": "Incremental graph division method, device, equipment, medium and product",
  "abstract_en": "The application discloses an incremental graph division method...",
  "abstract_cn": "本申请公开了一种增量图划分方法、装置、设备、介质及产品...",
  "filing_date": "2023-02-06",
  "publication_date": "2023-04-18",
  "grant_date": "",
  "priority_date": "2023-02-06",
  "status": "Pending",
  "inventors_cn": "汤韬, 高鹏飞, 孙权, 潘婧, 赵金涛, 郑建宾, 艾博轩, 庞悦",
  "assignees_en": "China Unionpay Co Ltd",
  "assignees_cn": "China Unionpay Co Ltd",
  "cpc_codes": "G06F16/174, G06F16/901, G06F16/9536, G06Q50/00, Y02D10/00",
  "ipc_codes": "",
  "legal_events": "[{\"date\":\"2023-02-06\",\"title\":\"Application filed by China Unionpay Co Ltd\",\"type\":\"filed\"}]",
  "family_members": "WO2024164667A1",
  "pdf_url": "https://patentimages.storage.googleapis.com/64/48/49/800ddc17fec1fc/CN115982415A.pdf",
  "google_patents_url": "https://patents.google.com/patent/CN115982415A/en",
  "scraped_at": "2026-05-11T04:22:20.537Z"
}
Field Type Description
publication_number string CNIPA publication number (e.g. CN114547329B)
application_number string Application number with country prefix
title_en string Patent title in English (Google-translated)
abstract_en string Full abstract in English (Google-translated)
abstract_cn string Full abstract in original Chinese
filing_date string Filing date (YYYY-MM-DD)
publication_date string Publication date (YYYY-MM-DD)
grant_date string Grant date (YYYY-MM-DD). Empty if not yet granted.
priority_date string Priority date (YYYY-MM-DD)
status string Legal status (Pending, Active, Expired, etc.)
inventors_cn string Inventors in Chinese characters, comma-separated
assignees_en string Current assignees in English, comma-separated
assignees_cn string Original assignees in Chinese, comma-separated
cpc_codes string CPC classification codes (leaf-level only), comma-separated
ipc_codes string IPC classification codes, comma-separated
legal_events string JSON-encoded array of {date, title, type} events
family_members string Patent family member publication numbers, comma-separated
pdf_url string URL to patent PDF
google_patents_url string Source URL of the Google Patents detail page
scraped_at string ISO timestamp when the record was scraped

FAQ

How do I scrape Chinese patents from CNIPA?

China CNIPA Patent Scraper handles it. Plug in a query, classification code, or assignee, set maxItems, and run. The scraper hits Google Patents with a CN-jurisdiction filter, which is the cleanest public surface for CNIPA data — and the only one that ships translated abstracts alongside the originals.

How much does this actor cost to run?

China CNIPA Patent Scraper uses pay-per-event pricing on the default_2603_basic profile at a 1.5x price coefficient. Residential proxy usage is included in the proxy tier. A 200-patent run typically costs a few cents in platform fees.

What data can I get for each patent?

China CNIPA Patent Scraper returns 20 fields per record: bilingual title and abstract, four key dates (filing, publication, grant, priority), inventors and assignees in both languages, CPC and IPC codes, legal event timeline, family members, and a direct PDF link. Enough to build prior-art datasets without scraping a second source.

Can I filter by date or classification code?

China CNIPA Patent Scraper accepts both. Use dateFrom/dateTo for priority date windows, and put a CPC or IPC code in the query field (e.g. H04L or G06F16/174). You can combine a classification code with a keyword if you want a narrower slice.

Does this actor need proxies?

China CNIPA Patent Scraper requires residential proxy. Google Patents returns a soft-block "Sorry" page to datacenter IPs, even on HTTP 200. The actor is preconfigured with Apify Residential — leave it alone unless you have a specific reason to override it.

Need More Features?

Need additional patent fields, different jurisdictions, or full-text claim extraction? Open an issue or get in touch.

Why Use China CNIPA Patent Scraper?

  • Bilingual output — Returns both the Google-translated English and the original Chinese for titles and abstracts, so downstream tools can pick whichever they need.
  • Search or seed — Run a broad keyword search across all CN patents, or paste a list of specific URLs when you already know which records you want. Both modes share the same parser.
  • Structured legal events — Most patent scrapers give you a status string and call it done. This one ships the full event timeline (filings, grants, expirations) as JSON, which is what you actually need for due diligence.
Try on Apify