Decide fate of the _preload cache layer after the data drop

coilysiren commented

2026-06-12 09:52:56 +00:00

Owner

The 20M species_profiles.json + ecopedia.json offline caches and their build scripts were dropped during the inbox#101 consolidation (decision 4: the cached info was not useful). src/eco_mcp_app/_preload.py degrades gracefully - lookups return None and the tools fall back to live Wikidata/iNaturalist/Wikipedia fetches - but the module, its tests, and the ECO_MCP_PRELOAD_DISABLE escape hatch are now dead weight guarding empty dicts.

Options:

Remove _preload.py and its call sites entirely, simplifying species.py / server.py / wikidata.py.
Keep it as the future cache slot if rate limits bite (the live paths now carry full Wikidata/iNaturalist exposure per request).

Origin: coilysiren/inbox#101

The 20M species_profiles.json + ecopedia.json offline caches and their build scripts were dropped during the inbox#101 consolidation (decision 4: the cached info was not useful). `src/eco_mcp_app/_preload.py` degrades gracefully - lookups return None and the tools fall back to live Wikidata/iNaturalist/Wikipedia fetches - but the module, its tests, and the `ECO_MCP_PRELOAD_DISABLE` escape hatch are now dead weight guarding empty dicts. Options: - Remove `_preload.py` and its call sites entirely, simplifying species.py / server.py / wikidata.py. - Keep it as the future cache slot if rate limits bite (the live paths now carry full Wikidata/iNaturalist exposure per request). Origin: https://forgejo.coilysiren.me/coilysiren/inbox/issues/101

coilysiren referenced this issue from a commit

2026-06-12 12:08:36 +00:00

refactor: remove the _preload cache layer

coilysiren closed this issue

2026-06-12 12:08:36 +00:00

coilysiren referenced this issue

2026-07-03 05:19:34 +00:00

Bring eco-gnome onto eco-app: self-host the MIT calculator (Phase 1), restyle to house style + wire Sirens data (Phase 2) #40