linguist

AIGX on GitHub (Linguist)

This kit registers AIGX as a language in github/linguist, so .aigx files get a language label and proper syntax highlighting on GitHub (blobs, diffs, search).

Two ways to highlight .aigx

1. Now — .gitattributes (interim)

This repo's .gitattributes already does:

*.aigx linguist-language=XML linguist-detectable=false

GitHub renders .aigx with XML highlighting today (AIGX is XML-style) and keeps it out of the repository language bar. Any repo adopting AIGX can copy that one line. Remove it once AIGX lands in Linguist proper.

2. Proper — a Linguist pull request

Registering AIGX in Linguist gives it its own label, color, and grammar (tm_scope: source.aigx).

Heads-up on the bar: Linguist's policy is that a new language should be in use across a few hundred repositories before it's merged. Land adoption first; this kit makes the PR mechanical when you're ready.

Steps:

1. Fork and clone github-linguist/linguist.
2. Add the language — paste languages.yml into lib/linguist/languages.yml (alphabetical order). Replace the placeholder language_id with the value Linguist's tests tell you to use (bundle exec rake test reports collisions; ids are otherwise stable random integers).
3. Add the grammar — AIGX's canonical grammar lives at editors/textmate/aigx.tmLanguage.json with scopeName: source.aigx. Register it:

   script/add-grammar https://github.com/Lolner95/AIGX

   (Linguist scans the repo for a grammar whose scope is source.aigx and vendors it into vendor/grammars + grammars.yml.)
4. Add samples — copy sample.aigx (and ideally one or two more real genome files, e.g. from examples/sourcing-app/.aigx/) into samples/AIGX/.
5. Verify:

   bundle exec rake test
   script/licensed   # license metadata for the vendored grammar

6. Open the PR against github-linguist/linguist describing AIGX, linking the spec (https://github.com/Lolner95/AIGX/blob/main/standard/AIGX-1.1.md) and adoption.

Why type: data?

AIGX is structured markup consumed as configuration/context (not a programming language and not prose), so it is typed data — the same family as XML, YAML, and JSON — with XML editor modes for highlighting.