Skip to content

Replay console failures on the BDSP sim from fault JSON - #447

Closed
dayfine wants to merge 6 commits into
bdsp/wanderer-zone-standfrom
nav/leg-replay
Closed

dayfine wants to merge 6 commits into
bdsp/wanderer-zone-standfrom
nav/leg-replay

Conversation

@dayfine

@dayfine dayfine commented Nov 13, 2026 •

Copy link
Copy Markdown
Owner

Depends on #432

A console failure becomes a small fault JSON (start, injected faults, failed step, frame); one generic test runs every embedded file on the sim world, with trainers, NPCs and walls taken from the map. Adds LegReplay (parser, Pokemon/Navigation), the hold_slides and scene_slides knobs (a recovery hold that runs out a door unseen; a scene step that ends a tile further than counted), and devtools/maps/leg_replay.py to collect the facts from a console log. Three cases are embedded: 18:36 east-mouth slid, 20:23 hold out the east door, 21:02 camper drift; the last two fail on the code that failed on the console (73f04b6, 8a570a6) and pass on #432. The replay test discovers JSONs on its own, so it cannot satisfy code-health's test-case table rule (documented in the test; the one remaining failure). Two small touches outside this PR's files, approved: StorySim (knobs) and GridSim (SimWanderer::goes_home, so a talked-to person moves off the path). The FRLG walker's [Leg] records are not here (the FRLG chain has no LegTrace).

Testing: SerialProgramsTests --filter "Console failure" 42 passed, 0 failed.

🤖 Generated with Claude Code

https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss

dayfine and others added 4 commits November 12, 2026 21:43
LegReplay parses a small per-failure fault JSON (start, injected faults,
failed step, frame). One generic test runs every embedded file on the sim
world. SimWorldKnobs gains hold_slides for a recovery hold that runs out a
door. devtools/maps/leg_replay.py builds the facts from a console log.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss
Add the scene_ends_further fault (scene-step slide knob) and make a
hold that crosses an unseen warp stop with TRANSITION as the walker
does; person_stands now also covers non-wanderer people. Both pending
fault JSONs are active; they fail on 73f04b6 / 8a570a6 and pass on
the #432 tip. The data-driven replay test keeps discovering JSONs, so it
documents its exception to the test-case table rule.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss
dayfine and others added 2 commits November 12, 2026 23:55
… sim); the BDSP sim rejects it

Pokemon_LegReplay is owned here; the kind is added so the FRLG leg-records PR (#451) carries an identical copy and the two merge without a conflict in this file.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013VUubhHeYSMGSSDXZkKZss
@dayfine

dayfine commented Nov 13, 2026

Copy link
Copy Markdown
Owner Author

Split and moved, so closing this one. The shared part (Pokemon_LegReplay, with LegTrace and the leg_trace/leg_replay tools) is now #464 on nav/grid-story-steps, merged into the FRLG chain (#451, #458) and into bdsp/rebuild-base and the rebuild stack (#455 to #462). The BDSP part (the three 01.02 fault files, the replay test and the GridSim/StorySim hooks) moved unchanged onto the rebuild stack as #465 on bdsp/rebuild-seg01, where all three replay tests pass. The branch nav/leg-replay is left in place.

@dayfine dayfine closed this Nov 13, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant