small-scripts/tests/test-ci-triage-minimax-run.sh

#!/usr/bin/env bash
# test-ci-triage-minimax-run.sh — Verify ci-triage-minimax-run against its spec.
#
# Strategy: CI_TRIAGE_FAKE_LOGS bypasses gac, CI_TRIAGE_FAKE_JSON /
# CI_TRIAGE_FAKE_TRIAGE drive the sub-session paths, so exits 0/2/3,
# redaction, and the failed_step handling are all exercised offline.

set -uo pipefail

SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
REPO_ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
SCRIPT="$REPO_ROOT/scripts/ci-triage-minimax-run"

GREEN='\033[0;32m'
RED='\033[0;31m'
NC='\033[0m'
PASS=0
FAIL=0

assert_contains() {
    local label=$1 needle=$2 haystack=$3
    if [[ "$haystack" == *"$needle"* ]]; then
        printf "  ${GREEN}PASS${NC}  %s\n" "$label"; PASS=$((PASS + 1))
    else
        printf "  ${RED}FAIL${NC}  %s\n" "$label"
        printf "        expected to contain: %s\n" "$needle"; FAIL=$((FAIL + 1))
    fi
}

assert_not_contains() {
    local label=$1 needle=$2 haystack=$3
    if [[ "$haystack" != *"$needle"* ]]; then
        printf "  ${GREEN}PASS${NC}  %s\n" "$label"; PASS=$((PASS + 1))
    else
        printf "  ${RED}FAIL${NC}  %s\n" "$label"
        printf "        expected NOT to contain: %s\n" "$needle"; FAIL=$((FAIL + 1))
    fi
}

assert_eq() {
    local label=$1 expected=$2 actual=$3
    if [[ "$expected" == "$actual" ]]; then
        printf "  ${GREEN}PASS${NC}  %s\n" "$label"; PASS=$((PASS + 1))
    else
        printf "  ${RED}FAIL${NC}  %s\n" "$label"
        printf "        expected: %s  actual: %s\n" "$expected" "$actual"; FAIL=$((FAIL + 1))
    fi
}

TMP=$(mktemp -d /tmp/test-ci-triage.XXXXXX)
trap 'rm -rf "$TMP"' EXIT

# gac-shaped NDJSON envelope + log lines, including a credential-shaped URL
FAKE_LOGS="$TMP/logs.ndjson"
cat > "$FAKE_LOGS" <<'EOF'
{"job_id":1,"conclusion":"failure","failed_step":"Run tests","sliced":true,"note":"","markers":[],"lines":["cloning https://x-access-token:ghp_abcdEFGH1234@gitea.example/r.git","FAILED tests/test_x.py::test_y - AssertionError","exit code 1"]}
EOF
FAKE_LOGS_NOSTEP="$TMP/logs-nostep.ndjson"
echo '{"job_id":1,"conclusion":"cancelled","failed_step":null,"sliced":false,"note":"","markers":[],"lines":["cancelled"]}' > "$FAKE_LOGS_NOSTEP"

FAKE_OK_JSON="$TMP/ok.json"
echo '{"is_error":false,"result":"done"}' > "$FAKE_OK_JSON"
FAKE_ERR_JSON="$TMP/err.json"
echo '{"is_error":true,"result":"boom"}' > "$FAKE_ERR_JSON"
FAKE_TRIAGE="$TMP/triage.md"
printf '## Failing step\nRun tests\n## Error excerpt\nFAILED tests/test_x.py\n## Probable cause\nAssertion failure in test_y.\n## Suggested next action\nRe-run that test.\n## Confidence\nhigh\n' > "$FAKE_TRIAGE"
FAKE_TRIAGE_EMPTY_CAUSE="$TMP/triage-empty.md"
printf '## Failing step\nRun tests\n## Probable cause\n## Confidence\nlow\n' > "$FAKE_TRIAGE_EMPTY_CAUSE"

echo "== ci-triage-minimax-run: --help =="
HELP_OUT="$("$SCRIPT" --help 2>&1)"
assert_contains "help shows usage" "--repo" "$HELP_OUT"

echo "== usage errors (exit 1) =="
"$SCRIPT" >/dev/null 2>&1; assert_eq "no args exits 1" 1 $?
"$SCRIPT" --repo badformat --run 1 >/dev/null 2>&1; assert_eq "bad repo format exits 1" 1 $?
"$SCRIPT" --repo o/r --run abc >/dev/null 2>&1; assert_eq "non-integer run exits 1" 1 $?

echo "== dryrun =="
DRY_OUT="$("$SCRIPT" --repo skynet/gac-scratch --run 42 --dryrun 2>&1)"
assert_eq "dryrun exits 0" 0 $?
assert_contains "dryrun shows gac command" "inspect skynet/gac-scratch --run 42" "$DRY_OUT"
assert_contains "dryrun defaults to Flash" "MiniMax-M3.1-Flash-Preview" "$DRY_OUT"
assert_contains "dryrun default effort low" "--effort low" "$DRY_OUT"
assert_contains "dryrun defers fetch" "not fetched in dryrun" "$DRY_OUT"
assert_contains "dryrun shows job passthrough default" "job: auto" "$DRY_OUT"

echo "== dryrun --m3/--effort =="
M3_OUT="$("$SCRIPT" --repo skynet/gac-scratch --run 42 --m3 --dryrun 2>&1)"
assert_eq "m3 dryrun exits 0" 0 $?
assert_contains "m3 dryrun shows M3 model" "MiniMax-M3[1m]" "$M3_OUT"
assert_not_contains "m3 dryrun sends no effort" "--effort" "$M3_OUT"
EF_OUT="$("$SCRIPT" --repo skynet/gac-scratch --run 42 --effort high --dryrun 2>&1)"
assert_contains "explicit effort overrides default" "--effort high" "$EF_OUT"
"$SCRIPT" --repo o/r --run 1 --effort bogus --dryrun >/dev/null 2>&1; assert_eq "bad effort exits 1" 1 $?

echo "== fake logs: redaction + failing step in prompt, exit 0 =="
OUT0="$TMP/out0.md"
RUN_OUT="$(CI_TRIAGE_FAKE_LOGS="$FAKE_LOGS" CI_TRIAGE_FAKE_JSON="$FAKE_OK_JSON" CI_TRIAGE_FAKE_TRIAGE="$FAKE_TRIAGE" \
    "$SCRIPT" --repo o/r --run 7 --out "$OUT0" 2>&1)"
assert_eq "happy path exits 0" 0 $?
assert_contains "triage file written" "ok: triage written" "$RUN_OUT"

echo "== fake logs: null failed_step still succeeds =="
CI_TRIAGE_FAKE_LOGS="$FAKE_LOGS_NOSTEP" CI_TRIAGE_FAKE_JSON="$FAKE_OK_JSON" CI_TRIAGE_FAKE_TRIAGE="$FAKE_TRIAGE" \
    "$SCRIPT" --repo o/r --run 8 --out "$TMP/out-nostep.md" >/dev/null 2>&1
assert_eq "null failed_step exits 0" 0 $?

echo "== gate: empty probable cause -> exit 3 =="
CI_TRIAGE_FAKE_LOGS="$FAKE_LOGS" CI_TRIAGE_FAKE_JSON="$FAKE_OK_JSON" CI_TRIAGE_FAKE_TRIAGE="$FAKE_TRIAGE_EMPTY_CAUSE" \
    "$SCRIPT" --repo o/r --run 9 --out "$TMP/out3.md" >/dev/null 2>&1
assert_eq "empty probable-cause exits 3" 3 $?

echo "== gate: no triage file -> exit 3 =="
CI_TRIAGE_FAKE_LOGS="$FAKE_LOGS" CI_TRIAGE_FAKE_JSON="$FAKE_OK_JSON" \
    "$SCRIPT" --repo o/r --run 10 --out "$TMP/out3b.md" >/dev/null 2>&1
assert_eq "missing triage file exits 3" 3 $?

echo "== is_error envelope -> exit 2 =="
CI_TRIAGE_FAKE_LOGS="$FAKE_LOGS" CI_TRIAGE_FAKE_JSON="$FAKE_ERR_JSON" \
    "$SCRIPT" --repo o/r --run 11 --out "$TMP/out2.md" >/dev/null 2>&1
assert_eq "is_error envelope exits 2" 2 $?

echo ""
echo "ci-triage-minimax-run: $PASS passed, $FAIL failed"
[[ "$FAIL" -eq 0 ]]