What this is, and how it is made

Every Central Act of India that has section-level text in the official corpus, with each section on its own permanent URL, and the Acts of every State and Union Territory alongside them. No paywall, no registration, no advertisement over the text, and no interstitial between you and the provision you came to read.

Alongside the text we publish original matter of our own: an answer-first summary on every section, the statutory cross-reference graph, the administering Ministry, commencement information, and the questions people actually ask about a provision. That is partly a product decision and partly a legal one — section 52(1)(q)(ii) of the Copyright Act, 1957 permits reproducing an Act of a Legislature only when it is published together with commentary or other original matter.

Most of the corpus is harvested from India Code, the Government of India's official repository, through its public API. The harvest is a deliberate, reviewed step run on a machine we control. The published site makes no external calls at all: every value on every page comes out of our own database.

Some of it comes from elsewhere, because India Code does not hold it, and each part says so on the page itself:

• The income-tax corpus. India Code has no record of the Income-tax Act, 1961, the Finance Acts, or the Income-tax Rules. Those are taken from the Income Tax Department's own consolidated text.
• 7,765 State Acts read off a scan. For most States, India Code holds the Act as a record and a scanned document with no sections behind it. Those were recovered by machine from the image, not transcribed by the Government, and every one of them is marked text_source: ocr in the API and labelled on the page, with the Government's own document linked beside it. A further 586 are listed without text at all, because the scan would not yield any we would be willing to set out as the law.
• 2,628 provisions from a State's own consolidated reprint. Across 35 Acts, India Code lists a provision's number and heading and publishes no text for it — the field the text would come in is not among the ones it serves for those Acts, and it holds no document for them either. Where the State itself publishes what India Code does not — the Act as one document carrying every amendment, current to a stated date — that is used instead. Those provisions are marked text_source: state-reprint in the API and say on the page which document they came from and to what date its text runs, because a consolidation is current to a day and that day may be some years ago. 1,368 provisions still have no text here, and their pages say so rather than pretending otherwise.

We do not edit statutory text. Where the source publishes text we normalise its markup and do not touch the words. Where the source publishes only an image, the words are a machine's reading of it, which is a different kind of artefact and is never presented as the other.

Statutory text is dense with internal citation: "as provided in section 12", "notwithstanding section 4 of the Companies Act, 2013". We extract those references at build time and turn each one into a link, which yields a graph of 1,73,699 edges across the corpus.

A link is only ever emitted when the destination section actually exists in our corpus. A reference we cannot resolve is left as plain text rather than pointed somewhere plausible.

• We do not redirect repealed provisions to their successors. An offence committed before 1 July 2024 is still tried under the Indian Penal Code. Those pages have independent legal value.
• We do not publish a mapping we cannot source. See the note on IPC to BNS for where that stands.
• We do not machine-translate operative statutory text. Hindi, when it comes, will be the official Hindi text under the Authoritative Texts (Central Laws) Act, 1973.
• We do not block AI crawlers. A public statute corpus whose value is being the reference has nothing to gain from being invisible to the tools people now ask first.
• We do not gate anything behind registration.

If a provision here differs from the Gazette, the Gazette is right and we want to know. Write to hi@ecourtsindia.com with the URL and what is wrong.

This site is not a substitute for legal advice, and nothing on it is advice. For the authoritative text of any enactment, consult the Gazette of India.