From fdc6d396d68601c690427d53cf1a485d7dc35fbc Mon Sep 17 00:00:00 2001 From: Claude Date: Fri, 7 Aug 2026 01:42:19 +0000 Subject: [PATCH] record: purge-autocti-dataset-1d-overview shipped (autocti_workspace#18) + condemn entry 282 tracked simulated-dataset files (~5.35 MB) purged behind 21 new auto-simulate guards; workspace now on the standard dataset/** ignore regime. Condemned-ledger entry filed with pre-purge SHA 50b701d, sweep-after 2026-09-07. Index regenerated; lifecycle check OK. Co-Authored-By: Claude Fable 5 Claude-Session: https://claude.ai/code/session_01Vu1p9rcDQGRrYddB6VWcxN --- complete/2026/08/purge-autocti-dataset-1d-overview.md | 8 ++++++++ complete/index.md | 3 ++- condemned.md | 11 +++++++++++ 3 files changed, 21 insertions(+), 1 deletion(-) create mode 100644 complete/2026/08/purge-autocti-dataset-1d-overview.md diff --git a/complete/2026/08/purge-autocti-dataset-1d-overview.md b/complete/2026/08/purge-autocti-dataset-1d-overview.md new file mode 100644 index 00000000..e089ccf0 --- /dev/null +++ b/complete/2026/08/purge-autocti-dataset-1d-overview.md @@ -0,0 +1,8 @@ +- issue: none (no GitHub issue — the leftover from `autocti-util-dataset-export`, executed on direct human instruction "do the leftover" in the same cloud session; no draft prompt existed, only the candidates note in that record) +- completed: 2026-08-07 +- pr: https://github.com/PyAutoLabs/autocti_workspace/pull/18 (MERGED, merge af5b7edc) +- summary: The final dataset-bulk purge for autocti_workspace, by the leg-1 recipe: 282 tracked simulated-dataset files (~5.35 MB) untracked — all of `dataset/dataset_1d/` (simple, species_x1_continuum_0, species_x3, temporal) and the simulated `dataset/overview/` subtrees (dataset_1d, imaging_ci/calibrate) — behind 21 new auto-simulate guards across 18 consumer scripts using `ac.util.dataset.should_simulate` (exported the same day). The workspace now matches the other workspaces' regime: `dataset/**` ignored by default, allowlist authoritative in `dataset/.gitignore`, everything simulated on demand. Kept committed by write-site judgment: the 5 overview media files (ccd.gif, ccd_schematic.png, cti.gif, cti_time_evolution.png, what_is_cti.png — only ever read, no writer) + `dataset/.gitignore`. Condemned-ledger entry `release-datasets/autocti-dataset-1d-overview` filed (pre-purge SHA 50b701d, sweep-after 2026-09-07). +- evidence: per-dataset provenance by reading simulator write-sites (table in commit 679a0a6 body); clean-tree regeneration proven per dataset (tree moved away → guarded consumer run → files re-created and read back): simple via correction/start_here.py, temporal via advanced/temporal/fit.py, both overview trees via overview_5_fitting.py (pre and post purge); species_x1_continuum_0 has zero consumers (witnessed via its simulator directly). Post-purge validation: modeling/start_here.py under PYAUTO_TEST_MODE=1 (real reduced sampler), extract.py, Dataset1dPlotter.py all exit 0; `git status` clean after every regeneration; py_compile clean on all 18 edited scripts. +- traps: (1) `simulate_datasets_include.sh` carried a `git add -f dataset/dataset_1d dataset/overview` that would have silently re-committed the purge — deleted (the exact trap leg 6 hit and narrowed for imaging_ci); `simulate_datasets_missing.sh` now runs all 15 simulators. (2) gitignore re-include mechanics: pins inside an ignored directory need the parent unignored — the authoritative allowlist moved into `dataset/.gitignore` (its `*` overrides the root file), verified with `git check-ignore` both ways. (3) `overview_4_charge_injection_data.py` consumed two already-untracked datasets with NO guards — guards added in the same pass. (4) `overview.py` simulator builds its path as `join("dataset", name, type)` — inverted order; write-site reading caught what name-matching would have missed. (5) Stale tracked `species_x3/cti.json`: the current simulator no longer writes it and nothing reads it — purged with the rest. +- deliberately-unguarded: `plot/fpa/dataset_1d/fit.py` (its `simple` path is dead code — all `from_fits` use self-simulated fpa_plot paths); `advanced/database/examples/data_fitting.py` and `advanced/temporal/{plot,database}.py` (read only `output/`/sqlite). autocti_workspace_test has no consumers of these paths; `legacy/` untouched. +- merge-context: merged 2026-08-07 on the session's standing human authorization; autocti_workspace has no CI workflows (resurrection Phase 3 pending) — merge evidence is the validation above. Notebook drift: 18 notebook counterparts trail their scripts until the release-pipeline regen (autocti is not a PyAutoHands generate target). diff --git a/complete/index.md b/complete/index.md index 244c0d23..96e41a96 100644 --- a/complete/index.md +++ b/complete/index.md @@ -6,7 +6,7 @@ Token-light navigation over the finished-work records (schema: only then grep a dated bucket. Curators: edit the band between the CURATED markers; everything below GENERATED is rebuilt. -926 records across 7 buckets. +927 records across 7 buckets. ## Highlights @@ -49,6 +49,7 @@ _(curate hard-won records here — survives regeneration.)_ - [potential-correction-env-declaration](2026/08/potential-correction-env-declaration.md) - [potential-correction-validation](2026/08/potential-correction-validation.md) - [pr-ci-for-own-test-suite](2026/08/pr-ci-for-own-test-suite.md) +- [purge-autocti-dataset-1d-overview](2026/08/purge-autocti-dataset-1d-overview.md) — no GitHub issue — the leftover from `autocti-util-dataset-export`, executed on direct human instruction "do th… - [pyautobrain-pr-test-ci](2026/08/pyautobrain-pr-test-ci.md) — auto-closed by the merge - [pyautogalaxy-mge-sigma-test](2026/08/pyautogalaxy-mge-sigma-test.md) - [simulator-util-to-af-ex](2026/08/simulator-util-to-af-ex.md) — moved the four 1D-Gaussian simulator helpers out of the duplicated diff --git a/condemned.md b/condemned.md index a585a111..4d5a1695 100644 --- a/condemned.md +++ b/condemned.md @@ -175,6 +175,17 @@ One `##` block per item. Fields: - breaks-if-wrong: loses the exact historical noise realizations only; regenerated copies are equivalent by construction - archive-ref: n/a — committed deletion; pre-purge SHA 272bae3 VOIDED by the 2026-07-27 leg-7 autocti rewrite — bytes recoverable ONLY from the local mirror backup ~/Code/PyAutoLabs-backups/autocti_workspace-pre-rewrite-2026-07-27.git +## release-datasets/autocti-dataset-1d-overview +- type: file +- locator: autocti_workspace dataset/dataset_1d/{simple,species_x1_continuum_0,species_x3,temporal} + dataset/overview/{dataset_1d,imaging_ci}/ (282 files, 5.35 MB, PR#18) +- confidence: 0.98 +- reason: committed simulated 1D CTI calibration + overview datasets, each regenerated in seconds by its sole write-site simulator (scripts/dataset_1d/simulators/** and scripts/imaging_ci/simulators/overview/calibrate.py); clean-tree regeneration proven per dataset pre-purge and re-witnessed post-purge with clean git status; 21 auto-simulate guards added to 18 consumers. The dataset-bulk leftover recorded in complete/2026/08/autocti-util-dataset-export.md. The 5 non-simulated overview media files (ccd.gif, ccd_schematic.png, cti.gif, cti_time_evolution.png, what_is_cti.png) stay committed — no write site. +- merged: no +- condemned: 2026-08-07 +- sweep-after: 2026-09-07 +- breaks-if-wrong: loses the exact historical noise realizations only; regenerated copies are equivalent by construction +- archive-ref: n/a — committed deletion; pre-purge SHA 50b701d on autocti_workspace main (post-rewrite history — reachable on the remote) + ## pyautogalaxy/docs-cite-prodigy - type: branch - locator: docs/cite-prodigy (PyAutoGalaxy, tip e06a527)