pMissionEval
Mission utility tests for pMissionEval. These cases verify pass/fail grading, built-in finish publication, staged logic sequences, fail-condition precedence, lead-only and fail-only evaluator shapes, flags and mailflags, numeric and mission-hash macros, literal numeric flags, CSP report output, clock macros, and no-hash report output.
Stem Mission
The stem mission runs a single shoreside MOOS community with pMissionHash, pMissionEval, uTimerScript, logging, and process-watch support. Case overlays change pMissionEval configuration and scripted MOOS mail while the harness starts uniquely aliased uMayFinish instances so completion remains observable.
Examples
These pMissionEval GIFs are generated mail-flow views instead of raw pMarineViewer captures because the meaningful evidence is mission grading mail, report rows, flags, and completion publication. The visuals call out the evidence that makes each case pass.
Case Matrix
eval_baseline_pass
Posts TEST_EVAL_READY=true and PASS_INPUT=true, testing the basic one-lead, one-pass-condition evaluator; passes when pMissionEval writes exactly one result row with subject_grade=pass and the mission launcher exits successfully.
eval_false_condition_fail
Posts TEST_EVAL_READY=true with PASS_INPUT=false, testing a false pass condition after the lead condition opens evaluation; the harness passes when pMissionEval writes subject_grade=fail with pass_input=false.
eval_builtin_finish_pass
Removes the explicit MISSION_EVALUATED result flag while satisfying the normal lead and pass conditions, testing pMissionEval's built-in completion post; passes when the subject grades pass, the .alog contains MISSION_EVALUATED, no EVAL_RESULT_FLAG line contains result:, and the mission launcher exits successfully.
eval_numeric_multi_pass
Requires both TEST_EVAL_READY=true and CASE_PHASE=ready, plus PASS_INPUT=true and NUM_SCORE>40, testing multiple lead/pass conditions and numeric macro expansion; passes with subject_grade=pass, phase=ready, score=42.5, and EVAL_SCORE_FLAG=score:42.5 in the .alog.
eval_sequence_two_stage_pass
Defines consecutive stages requiring STAGE1_READY/STAGE1_PASS and then STAGE2_READY/STAGE2_PASS, testing ordered LogicAspect evaluation; passes when both ready posts appear and the result row records subject_grade=pass, stage1=true, and stage2=true.
eval_sequence_first_stage_fail
Posts STAGE1_READY=true with STAGE1_PASS=false and never posts stage-two mail, testing failure at the first LogicAspect; the harness passes when the result records subject_grade=fail and stage1=false, EVAL_FAIL_FLAG=stage1_failed appears, and no STAGE2_READY mail appears.
eval_sequence_second_stage_fail
Passes stage one, then posts STAGE2_READY=true with STAGE2_PASS=false, testing failure after advancing to the second LogicAspect; the harness passes when the result records subject_grade=fail, stage1=true, and stage2=false, and the .alog contains both STAGE2_READY=true and EVAL_FAIL_FLAG=stage2_failed.
eval_numeric_literal_flag_pass
Configures result_flag=EVAL_LITERAL_NUM=17.25, testing numeric-literal publication through the non-string flag path; passes when the subject grades pass with case_value=literalnum and the .alog contains EVAL_LITERAL_NUM=17.25.
eval_lead_only_pass
Configures only lead_condition=TEST_EVAL_READY=true, testing an evaluator with no pass or fail conditions; passes when the ready post produces subject_grade=pass, eval=true, and case_value=leadonly.
eval_numeric_partial_fail
Satisfies both lead conditions and PASS_INPUT=true but posts NUM_SCORE=39 against NUM_SCORE>40, testing partial failure in a multi-condition evaluator; the harness passes when the result records subject_grade=fail, phase=ready, and score=39, and the .alog contains EVAL_FAIL_FLAG=partial_numeric_fail.
eval_fail_only_condition_fail
Configures no pass conditions and posts FORCE_FAIL=true after the lead condition, testing a fail-only evaluator; the harness passes when the result records subject_grade=fail, eval=true, and force_fail=true, and the .alog contains EVAL_FAIL_FLAG=fail_only.
eval_fail_condition_fail
Posts PASS_INPUT=true and FORCE_FAIL=true, testing that an explicit fail condition overrides a satisfied pass condition; the harness passes when the result records subject_grade=fail and force_fail=true, and the .alog contains EVAL_FAIL_FLAG.
eval_flags_and_mail_pass
Posts CASE_VALUE=alpha and CASE_MAIL=baseline, testing macro expansion in configured result_flag, pass_flag, and mailflag publications; passes when the subject grades pass and the .alog records exactly EVAL_RESULT_FLAG=result:alpha, EVAL_PASS_FLAG=pass:alpha, and EVAL_MAIL_SEEN=true.
eval_csp_report_pass
Sets report_line_format=csp and satisfies the baseline conditions, testing comma-and-space report formatting; passes when the subject grades pass and the result row contains the ordered comma-separated fields form, mmod, grade=pass, and eval=true.
eval_mhash_macros_pass
Places $[MHASH_SHORT], $[MHASH_LSHORT], $[MHASH_UTC], $[MHASH], and $[MISSION_HASH] in report columns, testing mission-hash macro expansion; passes when the row matches the expected short-word, timestamp, and full-hash formats for every macro.
eval_clock_macros_pass
Places $[MONTH], $[DAY], $[HOUR], $[MIN], $[SEC], $[DATE], and $[TIME] in report columns, testing clock macro expansion; passes when every field has its expected numeric format and no raw $[ token remains.
eval_no_hash_report_pass
Removes pMissionHash from ANTLER and writes a report with no hash columns, testing evaluation without mission-hash mail; passes when the subject grades pass with eval=true and case_value=nohash, and no MISSION_HASH publication appears in the .alog.