endomorphosis/ipfs_yemen_laws

Yemen NIC Legislation DB and MoJ LawsMD

Research snapshot of official national legislation from National Information Center laws DB + Ministry of Justice LawsMD.

Not legal advice. The official gazette / authentic source prevails over this corpus.

Snapshot

Field	Value
Snapshot date	2026-10-14 (batch 1; earlier rows 2026-09-10)
Coverage	snapshot
Source	National Information Center laws DB + Ministry of Justice LawsMD
Collector	scrapers/collect_ye.py (earlier rows); scrapers/collect_ye_nic_v2.py (batch 1)
Laws / instruments	1,269
Articles	21,126
Language	ar
Jurisdiction	Yemen
License	ye-nic-moj-official-texts

Contents

• data/laws.parquet — one row per instrument.
• data/articles.parquet — article/section rows for the same snapshot.
• scrapers/collect_ye.py — research collector used to build this snapshot.
• scrapers/collect_ye_nic_v2.py, run_slot.sh, restage_ye.py, dedupe_ye.py, pii_scan.py, finalize_ye.py, robots_guard.py, requirements.txt (pinned), README.md, RUN_NOTE.md — batch 1 collector, staging and QA code (see scrapers/README.md).
• manifests/ye_code.json — batch 1 code record (file sha256s, upstream commit); ye_stage_result.json, ye_dedupe.jsonl (23 inherited duplicate article ids renamed -dupN), ye_drops.jsonl (11 dropped laws), ye_redactions.json (0), ye_known_issues.jsonl.
• bounds.json — batch 1 scope, hosts contacted and limits; pii_scan.json — counts-only PII/secret scan.

License / reuse

Official NIC / MoJ texts; authentic official publication prevails; not legal advice.

This dataset is not legal advice. The official source prevails.

Source

• http://www.yemen-nic.info/db/laws_ye/

Known gaps

NIC legislation DB HTML + MoJ LawsMD PDFs. Aden Official Gazette e-portal deferred.

Update 2026-10-14 (batch 1: NIC delta)

• +927 laws, +16,604 articles from the National Information Center laws DB (https://www.yemen-nic.info/db/laws_ye/, sections 541 قانون, 542 قانون معدل بقانون, 543 قرار بقانون): 342 → 1,269 laws, 4,522 → 21,126 articles. New ids are ye-nic-<ID> (laws) and ye-nic-<ID>-<n> (articles), source_type yemen_nic.
• Earlier law rows are unchanged. Earlier article rows are unchanged except the id of 23 rows in 19 inherited duplicate-id groups, renamed -dupN (first occurrence keeps its id; logged in manifests/ye_dedupe.jsonl). No duplicate ids remain.
• 11 laws dropped (manifests/ye_drops.jsonl): 4 with no text besides the title, 7 that re-post a law already in the dataset under another NIC ID. NIC ID 69381 is listed but has no detail page and is not included.
• Network: only the official host www.yemen-nic.info; robots.txt checked on every hop, TLS verified, ≥2 s between requests, no archive or Wayback sources.
• Personal data: names in published decrees (e.g. citizenship decrees) are part of the official text and kept. The Ministry of Justice footer contact (Info@moj.gov.ye and its official number) in 4 earlier rows is official public-body contact data. No private ID numbers found; 0 redactions.

How to run (batch 1)

pip install -r scrapers/requirements.txt, then scrapers/run_slot.sh (Fleet v2 slot wrapper: memory/disk start gate held 2 min, 300 s stagger, live-tip check). It runs collect_ye_nic_v2.py (resumable; skips NIC IDs already in the dataset) and then finalize_ye.py; restage_ye.py and dedupe_ye.py do the offline restage. Details and the run record are in scrapers/README.md and scrapers/RUN_NOTE.md.

Licence basis

Yemen Law No. 15 of 2012 on Copyright and Related Rights, Art. 5(2): protection does not cover official documents such as texts of laws, regulations, government orders and judicial decisions (WIPO Lex: https://www.wipo.int/wipolex/en/legislation/details/19994). Authentic official text prevails; not legal advice.

Known issues (batch 1)

• 272 earlier law rows and 169 earlier article rows still end with the yemen-nic.info navigation menu (kept verbatim); batch 1 rows are clean. (Superseded: see 'Fix batch: nav-menu strip' below.)
• MANIFEST.md describes the 2026-09-10 snapshot and is not updated by batch 1; this README and manifests/ are current.

Fix batch: nav-menu strip (2026-10-15)

• Parent: endomorphosis/ipfs_yemen_laws @ f86999b325372d69f3e30e09873ea09f670e77bb. Offline only: no source site was contacted.
• The yemen-nic.info site navigation menu (≈650 chars captured after the law body) is removed from text with the same rule already applied to the rows added in the previous batch (NAV.sub("", text).rstrip(), see scrapers/build_ye_nav.py).
• data/laws.parquet: 181 of 1269 rows changed (176,029 chars removed). data/articles.parquet: 169 of 21126 rows changed (154,644 chars removed).
• Pattern: same anchors and span as the previous batch (law-type list قانون / قانون معدل بقانون / قرار بقانون / دستور الجمهورية اليمنية through تواصل معنا), whitespace-tolerant between the header lines because the older rows keep the page's \r\n and blank lines. The Copyright © National Information Center footer line after the menu is left in place, as in the previous batch.
• ids, all other columns, row order and row counts are unchanged. Rows without the menu are untouched.
• Known issue: 91 law rows (Common Crawl captures of NIC news/portal pages) carry a different NIC page layout without the law-type menu header; they are not touched (manifests/fix_navmenu_other_chrome.jsonl, ids only).
• Manifest: manifests/fix_navmenu.jsonl (one line per changed row: table, id, field, chars_removed; no text).
• PII: pii_scan.json (scanner v2, counts only); no class count rose versus the parent.