Data Sources & Licensing¶
bharat-courts does not host any court data of its own. It is a thin, open-source layer over public Indian court data — the official eCourts portals on one side, and a public AWS Open Data archive on the other. This page explains exactly where the data comes from, how current it is, how it is licensed, and how to attribute it when you republish it.
For the mechanics of how the two backends are wired together, see How it works.
Two kinds of source¶
bharat-courts reads from two complementary kinds of source:
- Live sources — the official eCourts and court websites. These reflect the current state of a case: status, next hearing, cause lists, freshly uploaded orders. They are CAPTCHA-gated and rate-limited.
- Historical archive — two public AWS Open Data S3 buckets mirroring decided judgments back to 1950. No CAPTCHA, no rate limits, but the data lags real time by a couple of months.
Which one answers my question?
Use the live sources for "what is happening now" — case status, next hearing date, today's cause list, an order uploaded this week. Use the archive for historical research and bulk retrieval of decided judgments. The Judgments facade picks between them for you on most queries.
Live sources¶
These are the official portals bharat-courts scrapes. Each has its own client in the SDK.
| Source | Portal | Client | What it provides |
|---|---|---|---|
| High Court Services | hcservices.ecourts.gov.in | HCServicesClient | High Court case status, orders, cause lists, bench/case-type discovery |
| District Courts | services.ecourts.gov.in | DistrictCourtClient | District court case status, orders, cause lists across 700+ court complexes |
| Judgment Search | judgments.ecourts.gov.in | JudgmentSearchClient | Full-text keyword search of High Court judgments, with PDF download |
| Supreme Court | www.sci.gov.in | SCIClient | The homepage "Latest Judgements / Orders" feed and PDF download |
| Calcutta High Court | calcuttahighcourt.gov.in | CalcuttaHCClient | Order/judgment search direct from the court's own site (better Calcutta HC PDF coverage from September 2020 onwards) |
Update cadence: live = real-time
Live sources are queried on demand and reflect whatever the portal shows at that moment. There is no caching delay on the data itself — if the court has updated a record, the live client sees it.
Known live-source limitations
A few flows are not yet wired up, and bharat-courts is honest about this rather than guessing:
- The Supreme Court client surfaces the recent-judgments feed and PDF download;
SCIClient.search_by_year(...)andsearch_by_party(...)raiseNotImplementedError(the legacy host they used is permanently offline and the replacement form is CAPTCHA-gated). - The HC Services
showRecordsendpoint does not return case status,registration_date,judges, ornext_hearing_date, so those fields stay empty on results from that path.
Historical archive (AWS Open Data)¶
For research and bulk work, ArchiveClient reads two public S3 buckets in the AWS Open Data programme. No CAPTCHA, no rate limits, no AWS account needed. They require the archive extra (pip install 'bharat-courts[archive]').
| Bucket | Contents | Registry |
|---|---|---|
indian-supreme-court-judgments | Supreme Court of India, 1950 → present | registry.opendata.aws |
indian-high-court-judgments | 25 High Courts | registry.opendata.aws |
Both buckets are:
- hosted in the
ap-south-1(Mumbai) region, - licensed CC-BY-4.0,
- maintained by Dattam Labs.
Judgment metadata lives in partitioned parquet files (queried with DuckDB); the PDFs are served as direct downloads (High Court) or per-year tar archives (Supreme Court). Both layers cache to disk under ~/.cache/bharat-courts/archive/.
Update cadence: archive lags by 2–3 months
- The Supreme Court bucket updates bi-monthly.
- The High Court buckets update quarterly.
A judgment delivered in the last 2–3 months may not be in the archive yet. If a structured archive query returns nothing for a recent year, retry on the live portal — for example Judgments().find(..., source="live"), or a direct JudgmentSearchClient / HCServicesClient lookup.
Coverage at a glance¶
| Court tier | Coverage | Available via |
|---|---|---|
| Supreme Court of India | Live recent-judgments feed; archive 1950 → present | SCIClient, ArchiveClient, Judgments |
| High Courts | 25 High Courts (live case data + archive) | HCServicesClient, JudgmentSearchClient, CalcuttaHCClient, ArchiveClient, Judgments |
| District Courts | 700+ court complexes across 36 states/UTs (live only) | DistrictCourtClient |
The full list of High Court codes is in the Courts reference.
Licensing¶
The bharat-courts software: MIT¶
The bharat-courts library itself — the SDK, CLI, and the bundled AI-agent skill — is released under the MIT License. You are free to use it commercially, modify it, and redistribute it, subject to the MIT terms.
The data: terms of the originating source¶
The software licence does not relicense the data. The data carries its own terms, which depend on where it came from:
- Archive data (both AWS Open Data buckets) is licensed CC-BY-4.0.
- Live portal data is retrieved directly from the official eCourts / court websites and is subject to those portals' own terms of use.
Attribution¶
When you redistribute or republish data obtained through bharat-courts, attribute the originating source — this is a condition of the CC-BY-4.0 licence on the archive, and good practice for the live portals.
- Archive (CC-BY-4.0): attribute Dattam Labs (the bucket maintainers) and the underlying eCourts platform. Retaining a link to the relevant AWS Open Data registry page is the simplest way to satisfy the attribution requirement.
- Live portals: credit the relevant eCourts portal or court website as the source.
Attributing in practice
A short credit line is enough — for example: "Judgment data via the Indian High Court Judgments / Supreme Court Judgments AWS Open Data buckets (CC-BY-4.0), maintained by Dattam Labs, sourced from the eCourts platform."
Disclaimer¶
Read this before relying on the data
- Not affiliated with the judiciary. bharat-courts is an independent, open-source tool. It is not affiliated with, endorsed by, or operated by the Supreme Court of India, any High Court, any District Court, the eCourts platform, or any government body.
- Provided as-is, from public sources. Data is surfaced as published by the official portals and the public archive. bharat-courts does not alter the substance of records; accuracy, completeness, and timeliness depend entirely on the upstream source.
- Verify mission-critical information against the official record. For anything that matters — limitation periods, hearing dates, the operative text of an order or judgment — confirm against the official court record or a certified copy before relying on it. The archive's freshness gap and the occasional empty field on live results make independent verification essential.
- Not legal advice. bharat-courts is a data-access tool. Nothing it returns, and nothing in this documentation, constitutes legal advice.
See also¶
- How it works — the architecture behind the live/archive split.
- Courts reference — every supported court and its codes.