Divine Discourses

Coverage: what this site's data actually covers

Every number on this page is measured directly from the data files bundled with this site at generation time. None is carried over from documentation or memory. Where a fact is not available in the repository, this page says so explicitly instead of estimating.

Generated by scripts/compute-coverage.mjs, which writes data/coverage/report.json. Run it twice and the output is byte-identical. See the public repository for the script.

Morphology fields ●

Of 77,429 word-tokens in the Leeds Quranic Arabic Corpus v0.4, how many carry each morphological field.

Root gloss coverage ●

0 of 1,642 roots (0%) have an English gloss sourced from a verified local file.

Per-word gloss coverage ●

A different question from root gloss coverage above: 6 of 114 surahs (5.26%) have a per-word English gloss, transcribed from Khan (2011), behind the Meaning column on the reading page.

Khan 2005 Reflections coverage ○

Khan's 2005 volume is a full tafsir of the two most-read surahs on this site, al-Fatihah and al-Baqarah. 0 of 2 surahs have any transcribed excerpt from it: data/khan-interpretations.json draws only on the 2011 volume. The count stays zero until someone brings a page-cited transcription; see Sources.

QurSim (Mishkat) cross-reference coverage ●

110 of 114 surahs have a data/qursim/{surah}.json cross-reference file.

Surahs not covered: 104, 105, 106, 110.

Counting-rule sensitivity ~

Roots where this site's own pages already document more than one defensible count for the same concept. Only entries already documented elsewhere on this site are listed here; this table does not introduce new claims.

Wanted: surah outlines to transcribe

Dr. Khan's 2013 volume publishes sectional outlines for surahs 85 to 114 ●. The transcribed ones power the outline exercises; the rest are this project's open work queue. Claiming one means transcribing from the book under the review checklist in CONTRIBUTING.md, or submitting your own structural hypothesis for it, which needs no book at all.

Wanted: ring and structural analyses

Ring composition, where a surah's opening and closing mirror each other around a centre, is documented in the scholarly literature. This site names the phenomenon and asserts no ring outline for any surah, because a real one has to be transcribed from a published, page-cited analysis, never computed. Inventing verse ranges under a scholar's name would misattribute work they did not do, so the count below is zero and stays zero until someone brings a citation.

Claiming one means naming the scholar, the volume and the page, and giving the verse ranges exactly as published. Cuypers, Farrin● and Islahi● are the obvious starting points. The Patterns page carries the one narrow mechanical proxy the site did test corpus-wide, and its null result, which says nothing either way about that literature.

Source registry

For each dataset in use: name, version, license, and source URL, exactly as recorded in data/sources.json. A field reading "not stated in repository" is not a claim of any particular license or absence of one; it means the repository's own records do not state it, and this page does not infer one.

Source-id usage across the site

A source can be fully cited in data/sources.json and still be reachable from no badge anywhere: nothing checked for that until this measurement. 34 of 36 sources appear in at least one data-source-ids attribute on the site, static or built at render time. The rest are cited in the bibliography on Sources but not yet attached to any specific claim a reader can badge-click through to.

For downloadable CSV/JSON tables of the underlying corpus statistics, see Export. For every dataset this site ships, with its own license and generator script, see Datasets.