Provenance

Built from sources you can inspect.

Quran content should never become anonymous data. Every Quran.ws dataset identifies its source, edition, riwayah, version, and verification status.

How a dataset is verified

01
Name the source

Every dataset starts from an identified printed edition or scholarly reference, never from an unattributed file.

02
Record every transformation

Extraction, normalization, and segmentation steps are listed so the path from source to data is reproducible.

03
Audit against the source

Text and geometry are checked against the source. Status stays "In review" until the audit completes.

04
Publish with a digest

Each version ships with a digest. Corrections create a new version; earlier versions remain available.

Dataset registry

DatasetRiwayahSourceVersionDigestStatus

Select a row to view its full record.

Full record

Quran Text · Ḥafṣ

Ḥafṣ text taken unedited from the KFGQPC digital package named above. Word identity is shared with all other riwayat in Quran Text.

Extracted from UthmanicHafs-v-3.0.zip, unedited
Every departure from the source package listed in the file itself
Rebuilt offline from the committed package by pipeline/build.py; the build fails if a letter moves
Word index and cross-riwayah numbering derived here — the source packages are ayah-level only
Quran Text · ḤafṣVerified
Building blockQuran Text
RiwayahḤafṣ ʿan ʿĀṣim
Mushaf editionKFGQPC Madani, 2026
SourceUthmanicHafs-v-3.0.zip
Count / extentKufi · 6,236 ayat
Version0.1.0
Digestsha256: cdec7341b7c684e7…
About Quran Text →

Maturity levels

Verified — checked against its named source and in production use.
Beta — verified, API may still change.
In review — under review; not for production Quran text.

Report an issue

Found a discrepancy between a dataset and its printed source? Open an issue with the dataset name, version, digest, and the page or ayah reference. Corrections ship as a new version with a new digest; earlier versions stay available.

Open an issue on GitHub →