jabber.developer2
37218d2e07
All checks were successful
CI Code / Check spelling (pull_request) Successful in 16s
CI Code / Check coding style (pull_request) Successful in 26s
CI Code / Code Coverage (pull_request) Successful in 9m12s
CI Code / Linux (ubuntu) (pull_request) Successful in 9m18s
CI Code / Linux (debian) (pull_request) Successful in 9m29s
CI Code / Linux (arch) (pull_request) Successful in 13m13s
- bound _parse_responses scans to the output_text part's object and to
the content array, so a later sibling item's "text" (e.g. a reasoning
summary) can never be returned as the assistant reply
- do not retry the other flavour on a curl timeout: the request likely
reached the server and may still be generating, so a re-POST of the
conversation could trigger a second billed generation
- remember an unparseable 2xx from the first AUTO attempt and surface
stashed first-attempt errors in both error paths, instead of showing
only the final attempt's transport or HTTP error
- recognize Ollama's model-not-found wording in _names_model so a model
typo is not misread as a missing endpoint
- match model errors case-insensitively and recognize the
'model "x" is not supported' wording, so a wrong-model 400 (seen live
on Perplexity /v1/responses) fails fast instead of probing the other
flavour with a doomed request
- say "empty response body" instead of "Unknown error" when an HTTP
error arrives without a body (e.g. Perplexity's bare 404)
- drop the dead, racy provider-lookup fallback in the generic request
thread: a missing provider ref is a caller bug and now fails loudly
- fix ai_providers_lists_defaults to expect the header the command
actually prints ("Configured providers:"); the test was broken since
its introduction but CI never ran it
- add functional test group 5 (AI command surface) to FUNC_TEST_GROUPS
so the CI parallel target runs it; proftest.c port ranges already
account for five groups