A small parser reads the UCD Blocks.txt file line by line. Each line gives a start codepoint, an end codepoint and a block name. The parser produces a JSON row with slug, block, start, end, count, script and a sample-character array. The full array is written to data/unicode-blocks.json.

The parser walks the range and selects a small set of representative codepoints, prioritizing assigned characters from the UnicodeData.txt file. The result is a sample-character array stored in each row, which the template renders as a clickable grid on the block page.

Most blocks have a clear dominant script, like Latin, Hangul or Bopomofo. The parser reads Scripts.txt from the UCD and tags each block with its dominant script. Edge cases like symbols and dingbats receive a generic script value the index page filters on like any other.

Re-run the parser against the new Blocks.txt file. New blocks appear as new rows, existing blocks may have their counts adjusted, and the catalog ships an updated dataset on the next SleekRank sync. The URL slugs stay stable for blocks that did not change.

Yes. Add a plane field on each row, like BMP, SMP or SIP, and the index page exposes a plane filter. The detail pages can include a sibling-blocks section that lists other blocks in the same plane, all derived from the dataset.

The sample-character grid stays bounded, but the row also stores the full count and a link to a paginated character index, like /unicode-block/hangul-syllables/page/3/. SleekRank routes the pagination through the same data source, so even big blocks remain navigable.

Yes. The sample-character grid links each codepoint to /unicode-character/{slug}/, which is rendered by another SleekRank dataset of codepoints. The two catalogs share script and plane fields, so navigation between block and character feels natural.

Yes. Each row can carry a proposal_url or scripts_introduced_in field. The template surfaces the metadata in a sidebar, which is useful for blocks tied to specific Unicode releases. The data lives in the row, so any update flows to the matching page on the next sync.