EPS
← All tasks·#1078Completed

daily-fix: paper_palette crashes at >8 colors - extend grace

kind: infra#daily-auto-filed
track:

Highlight text on any card (body, plan, or an event) to anchor a comment, or leave a whole-task comment here. Mention @claude to summon a reply.

No comments yet.

Overview / Motivation

Auto-filed by /daily 2026-07-05 (route 2: behavior/logic change -> independent review), from the nightly transcript problem sweep.

Goal

Make paper_palette(n) degrade gracefully for n>8: extend with a second colorblind-safe cycle differentiated by linestyle/marker (per the /paper-plots conventions), or fall back to a perceptually-uniform continuous colormap sampling for large n - never a hard ValueError at render time. Add a test for n=16.

Workflow gap

  • Bug observed: #833's on-instance lane G1 tail plotting crashed in ALL 9 behavior x layer cells (rc=1 each): 'ValueError: paper_palette supports at most 8 colors; requested 16' (paper_plots.py:180) - the plot wanted one color per source (16 sources). Numeric outputs survived; figures had to be regenerated locally by the analyzer, so the whole remote plotting leg was dead weight. Session d8a9ed84 (issue-833 worktree), 2026-07-05.
  • Why it is a workflow gap: the failure originates in the workflow surface / shared helper named below, not in any one experiment.
  • Confidence (emitter): medium

Scope / surfaces

  • Primary target: src/explore_persona_space/analysis/paper_plots.py
  • Analysis-code bug (not workflow surface): drop the wf-fix tags, keep daily-auto-filed.

Constraints / invariants

  • scripts/workflow_lint.py --check-asks + --check-references stay green; ruff clean on touched files; relevant tests pass.

Provenance

  • workflow_fix_target: src/explore_persona_space/analysis/paper_plots.py
  • source: /daily 2026-07-05 problem sweep (transcript-mined)
Activity