# htext knesset-laws-v2: Knesset laws as published, K1–8, OCR text layer (HeOCR)

The enacted laws of the first eight Knessets (1949–1977), as printed in Reshumot (Sefer HaChukim), cut into lines. Part of htext, HeOCR's public-domain Hebrew text for training Hebrew OCR.

- Version: `knesset-laws-v2`.
- Lines sharing any 6-word sequence with HeOCR's evaluation sets (12 private development, test and held-out targets) were removed, as were laws with long shared runs; HeOCR's evaluation ran its own independent check on this lines file and found 0 overlap. The files are byte-identical to the GitHub release knesset-laws-v2.
- Licence: CC0-1.0 (htext-cc0).
- Credit: The Knesset of Israel (Knesset database, laws as published in Reshumot), https://main.knesset.gov.il.
- Text quality: OCR layer. The lines are the Knesset's own OCR of the printed page, filtered against a lexicon and checked by eye; expect some OCR errors, and do not build word lists or lexicons from them.
- `card_*` files are the GitHub release's own corpus card, rights index and checksums.

| File | Size |
| --- | ---: |
| `SHA256SUMS` | 776 bytes |
| `card_CORPUS_CARD.md` | 11,367 bytes |
| `card_SHA256SUMS` | 669 bytes |
| `card_corpus_card.json` | 22,985 bytes |
| `card_knesset_laws_excluded.v1.json` | 340,039 bytes |
| `card_knesset_laws_eye_check_sample.v1.jsonl` | 24,521 bytes |
| `card_knesset_laws_rights_index.v1.jsonl.gz` | 280,902 bytes |
| `knesset_laws_line_index.v1.jsonl.gz` | 9,106,157 bytes |
| `knesset_laws_lines_plain_v2.txt` | 9,853,052 bytes |

- GitHub release: https://github.com/HeOCR/htext/releases/tag/knesset-laws-v2
- Hugging Face (the same files): https://huggingface.co/datasets/HeOCR/htext-knesset-laws/tree/knesset-laws-v2
- Documentation: https://github.com/HeOCR/htext/blob/main/docs/knesset_laws.md
- htext: https://heocr.org/htext

The rights fields are the project's reading of the law, not legal advice.
