_TOWN_ANCHORS_SEED covered only 29 hand-picked towns, while the live
CT108 town_anchors table (GUI-curated since) has grown to 186. Any
fresh deploy (fresh volume, or the pre-v19 sqlite path) would seed
only the original 29 and silently lose anchor-resolution coverage for
the other 157 real, already-in-production towns.
Regenerated the dict from a live dump of CT108's town_anchors
(name, lat, lon, state; all 186 rows confirmed enabled=1), alphabetized,
same dict-of-dicts shape and column-aligned brace formatting as before
(alignment column widened to fit the longest name, "mountain home
afb"). seed_town_anchors() is unchanged and still INSERT OR IGNORE, so
it stays idempotent against the existing production rows.
Expanding the seed changes which town resolves as "nearest" for a few
existing test fixtures whose incident/work-zone coordinates are
genuinely closer to a newly-added real town (e.g. Oakley, Wilder) than
to the old 29-town subset's nearest match -- those goldens and
assertions are updated to the new (correct) nearest-town result.
test_api_post_add_town's probe town is renamed from the now-real
"Bellevue" to a fictitious "Testopolis" to avoid a duplicate-name 400.
- Move all application source (meshai/, dashboard-frontend/, tests/,
config/, docs/, Dockerfile, etc.) into work/ directory
- Add Node.js multi-stage build to Dockerfile for frontend compilation;
remove compiled static assets from git tracking
- Fix satpass missing time windows: consolidation was splitting wire on
newline and only putting line 1 in event.title, dropping the time
window line that the composer uses for precomposed broadcasts
- Fix satpass burst flooding: stagger consolidation timers (+60s per
pending pass) so Central batch publishes don't blast the mesh
- Update CI workflow build context to work/
- Anchor lib/ and data/ gitignore patterns to repo root to prevent
false matches on nested directories
- Add dashboard-frontend/node_modules/ to .dockerignore
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
