Supreme Court (SCIClient)¶

SCIClient is the live client for the Supreme Court of India. It talks to the official portal at www.sci.gov.in and does two things reliably today:

1. list_recent_judgments() — scrapes the homepage "Latest Judgements / Orders" feed (the 50 most recent items), no CAPTCHA.
2. download_pdf() — fetches the PDF bytes for one of those judgments.

That is the whole live surface. Read the scope note below before you build on it.

Scope: what this client does and does not do¶

If you only need recent Supreme Court items — for a digest, a watch list, or a "what landed this week" check — list_recent_judgments() is exactly the right tool, and it needs no CAPTCHA and no credentials.

Quick example¶

import asyncio
from bharat_courts.sci import SCIClient


async def main():
    async with SCIClient() as client:
        recent = await client.list_recent_judgments(limit=10)
        for j in recent:
            print(f"{j.judgment_date}  {j.case_number}  {j.title}")

        # Download the PDF for the first item.
        if recent:
            j = await client.download_pdf(recent[0])
            with open("sci_latest.pdf", "wb") as f:
                f.write(j.pdf_bytes)

asyncio.run(main())

SCIClient is an async context manager; async with opens and closes the underlying HTTP session for you.

Methods¶

list_recent_judgments(*, limit=50)¶

Scrapes the homepage feed of the 50 most recent items. limit is keyword-only and capped at 50 by the portal; passing a smaller value truncates the list locally. Returns a list of JudgmentResult objects.

download_pdf(judgment)¶

Takes a JudgmentResult (typically one returned by list_recent_judgments()) and downloads its PDF. The judgment's pdf_url is the portal's /sci-get-pdf/?diary_no=... link that the in-page viewer iframe uses.

• On success, the PDF bytes are stored on judgment.pdf_bytes and the same object is returned.
• If the judgment has no pdf_url, it raises RuntimeError.
• If the response is not a valid PDF (it does not start with the %PDF magic bytes), it raises RuntimeError.

What you get back: JudgmentResult¶

Each item is a JudgmentResult. The fields most useful for the SCI feed:

See the models reference for the full field list.

SCI live feed vs. the historical archive¶

When to reach for something else¶

• Structured SC search ("all Justice Chandrachud judgments 2018–2024", "judgments citing X") — use the Judgments facade, which routes these to the archive automatically:

  import asyncio
  from bharat_courts import Judgments
  
  
  async def main():
      async with Judgments() as j:
          results = await j.find(
              court="sci", judge="chandrachud", year=(2018, 2024), limit=10,
          )
          for r in results:
              print(f"{r.decision_date}  {r.case_id}  {r.title}")
  
  asyncio.run(main())
  

• A specific SC judgment by CNR — Judgments().find(cnr=...) routes the CNR prefix straight to the archive partition. SCI CNRs start with the ESCR prefix.

• Full-text search across SC reports — the judgment search portal (JudgmentSearchClient) supports keyword search with court_type="3" (Supreme Court Reports).

See also¶

• Historical archive guide — SCI 1950 to present, structured and bulk search.
• The Judgments facade — the recommended default entry point for finding any judgment.
• Judgment search portal — keyword full-text search.
• Clients API reference — full signatures for SCIClient and the other live clients.