Мини-Ганҷур — Tajik proof of concept

▶ Live demo

A single static file, no build step, no server — proof that ganjoor-tajik-data is directly usable, and specifically that its deliberately thin poem files (just vOrder + tajikText, no duplicated metre/rhyme/section data) work exactly as documented: cross-referenced against ganjoor-data by matching verse order.

• Two data sources, one page. Every poem fetch pulls from both ganjoor-tajik-data (for the Tajik text) and ganjoor-data (for hemistich structure — which verse is the right/left half of a couplet). They're merged client-side by matching vOrder, exactly as ganjoor-tajik-data's own API.md describes. If the structure lookup fails for any reason, it falls back to showing the Tajik lines unpaired rather than breaking.
• Left-to-right, not right-to-left. Tajik written in Cyrillic reads LTR, unlike Persian — so unlike mini-ganjoor this page is dir="ltr" throughout, with a serif that actually supports Cyrillic well (PT Serif) instead of a Perso-Arabic display face.
• Sparse by nature. Most poets/categories/poems have no Tajik translation at all — 404s while browsing are expected and explained in the UI, not treated as errors.
• Same id-index jump-to-poem feature as mini-ganjoor, same underlying design (bucketed index/poems-by-id/{id / shardSize}.json).

1. Copy index.html into the root of a repo (this one, ganjoor/tjmini, already has it).
2. Push.
3. Repo → Settings → Pages → Source: Deploy from a branch → main / (root) → Save.

No build step, no dependencies beyond what's already in the file.

Same as mini-ganjoor: no search, no offline support, and uncommon poem structures fall back to one line per verse rather than reproducing every classical form exactly. This is a proof of concept for the data design, not a production reading app.