test(ai): expand unit and functional coverage for /ai feature

Unit tests (tests/unittests/test_ai_client.c): +50 cases on top of the
existing 50 — total 100. Covers:
  - chat response parser (ai_parse_response): OpenAI content + Perplexity
    text formats, escape decoding, empty/null/missing inputs, format-string
    safety, multiline content
  - error envelope parser (ai_parse_error_message): standard envelope,
    nested escapes, missing fields, null/empty
  - extended JSON escape: \b, \f, \r, all-specials, UTF-8 pass-through
  - provider autocomplete cycling with >=2 matches and wrap-around
  - session edge cases: NULL args, 100-message order preservation,
    set_model(NULL), ref/unref(NULL)
  - provider edge cases: get/remove with NULL, double-remove, survival
    via session ref after ai_remove_provider
  - settings: multi-key independence, missing key, cross-provider
    isolation
  - model parsing edges: data not array, empty data, "id" outside data,
    multiple models
  - prefs round-trip: set token -> shutdown -> init -> token reloaded
    from disk (uses load_preferences fixture)

Functional tests:
  - Tier A (test_ai.c): 15 cases for the /ai command surface that don't
    need HTTP. Covers /ai help, providers list, set provider/token,
    start with/without key/unknown provider, clear, remove, default
    provider/model, switch without window, bad subcommand.
  - Tier B (test_ai_http.c + ai_http_stub.{c,h,py}): 5 cases that
    exercise the libcurl path against a local Python HTTP stub. The
    stub serves canned bodies in five modes (ok, openai, 401, 500,
    models) so each chat/models/error path is hit end-to-end without
    network.

Infrastructure:
  - TEST_GROUPS bumped to 5 in proftest.c; AI tests live in their own
    Group 5 because mixing them with stabber-driven tests in Group 4
    poisoned stbbr_stop() teardown.
  - PROF_FUNC_TEST_AI macro in functionaltests.c registers AI tests
    with ai_init_test() which wraps init_prof_test with a prof_connect()
    so stabber sees a graceful disconnect at teardown.
  - dist_check_DATA ships ai_http_stub.py with the source tarball.

Source-level changes to enable testing:
  - src/ai/ai_client.{c,h}: dropped 'static' from _parse_ai_response
    (renamed ai_parse_response) and _parse_error_response (renamed
    ai_parse_error_message); declared under a "Parsing helpers (exposed
    for testing)" section. Same approach as ai_parse_models_from_json.

Results in Docker (cproof-debian image):
  - 595/595 unit tests pass
  - 20/20 AI functional tests pass (Group 5)
This commit is contained in:
2026-05-11 16:21:45 +03:00
parent 010b2062a4
commit e67ae4e7a5
15 changed files with 1484 additions and 10 deletions

View File

@@ -59,6 +59,8 @@
#ifdef HAVE_SQLITE
#include "test_export_import.h"
#endif
#include "test_ai.h"
#include "test_ai_http.h"
/* Macro to wrap each test with setup/teardown functions */
#define PROF_FUNC_TEST(test) \
@@ -66,6 +68,14 @@
#test, test, init_prof_test, close_prof_test, #test \
}
/* AI tests use a custom init that primes stabber via prof_connect so
* stbbr_stop() in close_prof_test() doesn't hang on a never-connected
* server thread. See ai_init_test() in test_ai.c. */
#define PROF_FUNC_TEST_AI(test) \
{ \
#test, test, ai_init_test, close_prof_test, #test \
}
#ifdef HAVE_SQLITE
/* Custom init that sets a non-UTC timezone (POSIX TST-3 = UTC+3) */
static int
@@ -84,9 +94,9 @@ main(int argc, char* argv[])
int group = 0; /* 0 = all groups */
if (argc > 1) {
group = atoi(argv[1]);
if (group < 1 || group > 4) {
if (group < 1 || group > 5) {
fprintf(stderr, "Usage: %s [group]\n", argv[0]);
fprintf(stderr, " group: 1-4 to run specific group, or omit for all\n");
fprintf(stderr, " group: 1-5 to run specific group, or omit for all\n");
return 1;
}
}
@@ -307,6 +317,39 @@ main(int argc, char* argv[])
#endif
};
/* ============================================================
* GROUP 5: AI command surface (console + HTTP stub)
* Standalone because AI tests don't use stabber's XMPP path; mixing
* them with XMPP-bound tests in the same group poisons stabber state
* between fixture init/teardown cycles and hangs subsequent
* prof_connect() calls.
* ============================================================ */
const struct CMUnitTest group5_tests[] = {
/* Tier A — no network calls */
PROF_FUNC_TEST_AI(ai_no_args_shows_help),
PROF_FUNC_TEST_AI(ai_providers_lists_defaults),
PROF_FUNC_TEST_AI(ai_providers_list_shows_key_status),
PROF_FUNC_TEST_AI(ai_set_provider_adds_custom),
PROF_FUNC_TEST_AI(ai_set_token_marks_key_set),
PROF_FUNC_TEST_AI(ai_start_unknown_provider_errors),
PROF_FUNC_TEST_AI(ai_start_without_key_errors),
PROF_FUNC_TEST_AI(ai_start_with_key_opens_window),
PROF_FUNC_TEST_AI(ai_clear_without_window_errors),
PROF_FUNC_TEST_AI(ai_remove_provider_works),
PROF_FUNC_TEST_AI(ai_remove_provider_unknown_errors),
PROF_FUNC_TEST_AI(ai_set_default_provider_unknown_errors),
PROF_FUNC_TEST_AI(ai_set_default_model_updates_provider),
PROF_FUNC_TEST_AI(ai_switch_without_window_errors),
PROF_FUNC_TEST_AI(ai_bad_subcommand_shows_usage),
/* Tier B — full request/response through Python HTTP stub */
PROF_FUNC_TEST_AI(ai_http_prompt_returns_ok),
PROF_FUNC_TEST_AI(ai_http_prompt_openai_format),
PROF_FUNC_TEST_AI(ai_http_prompt_401_shows_error),
PROF_FUNC_TEST_AI(ai_http_prompt_500_surfaced),
PROF_FUNC_TEST_AI(ai_http_models_fetch_lists_models),
};
/* Test group registry for easy extension */
struct
{
@@ -318,6 +361,7 @@ main(int argc, char* argv[])
{ "Group 2: ExportImport/Receipts", group2_tests, ARRAY_SIZE(group2_tests) },
{ "Group 3: Presence/Disconnect/DBHistory/Message/MUC-DB", group3_tests, ARRAY_SIZE(group3_tests) },
{ "Group 4: Session/MUC/Carbons/MigrationStress", group4_tests, ARRAY_SIZE(group4_tests) },
{ "Group 5: AI command surface (Tier A + Tier B)", group5_tests, ARRAY_SIZE(group5_tests) },
};
const int num_groups = ARRAY_SIZE(groups);