Purge tracked dataset_1d + overview datasets behind auto-simulate guards - #18
Merged
Merged
Conversation
…te guards Second leg of the dataset purge (first leg: imaging_ci, PR #12). Untracks the two remaining committed simulated-dataset trees — dataset/dataset_1d/** (239 files) and the simulated parts of dataset/overview/** (43 files), ~5.2 MB total — after adding `ac.util.dataset.should_simulate` auto-simulate guards to every consumer script, so the datasets regenerate at runtime. Provenance (verified by reading each simulator's output-path code): - dataset/dataset_1d/simple <- scripts/dataset_1d/simulators/start_here.py - dataset/dataset_1d/species_x1_continuum_0 <- .../simulators/examples/species_x1_continuum.py - dataset/dataset_1d/species_x3 <- .../simulators/examples/species_x3.py - dataset/dataset_1d/temporal (time_0..time_4) <- .../simulators/examples/temporal.py - dataset/overview/dataset_1d <- .../simulators/examples/overview.py - dataset/overview/imaging_ci/calibrate <- scripts/imaging_ci/simulators/overview/calibrate.py Guards added (21 guard blocks across 18 scripts): all dataset_1d modeling / correction / extract / results / database / temporal consumers, the plot consumers (plot/start_here, Dataset1dPlotter, FitDataset1dPlotter), and overview_4/5/6 (overview_4's uniform + non_uniform_cosmic_rays datasets were already untracked but had no guards; they get the same treatment). Regeneration proven per dataset before purging: each tree was moved away and a guarded consumer re-created and read it back (correction/start_here, species_x3, temporal/fit, overview_5 for both overview trees; species_x1_continuum_0 has no consumer scripts, so its simulator was run directly as the witness). Post-purge witnesses re-ran clean with `git status --porcelain` showing no untracked dataset files. Kept committed (with reasons): - dataset/overview/{ccd.gif, ccd_schematic.png, cti.gif, cti_time_evolution.png, what_is_cti.png}: hand-made illustrations displayed by overview_1/2/4 and the notebooks; no script writes them, so they cannot regenerate. - dataset/.gitignore: the per-directory ignore file, now carrying the media allowlist (the deeper file wins over the root .gitignore for dataset/ paths). Also: simulate_datasets_missing.sh now runs every simulator (all datasets are runtime-simulated now); simulate_datasets_include.sh is deleted — its job was to force-re-add (`git add -f`) the committed datasets, which would silently revert this purge. Note: species_x3's tracked cti.json was stale (the current simulator no longer writes it); no consumer reads it. Notebook counterparts of the 18 edited scripts regenerate via the release pipeline. Co-Authored-By: Claude Fable 5 <[email protected]> Claude-Session: https://claude.ai/code/session_01Vu1p9rcDQGRrYddB6VWcxN
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Completes the dataset-bulk purge for autocti_workspace (leg-1 recipe; follows July's imaging_ci purge #12 and today's
should_simulatemigration #17): 282 tracked dataset files (~5.3 MB) untracked — all ofdataset/dataset_1d/(simple, species_x1_continuum_0, species_x3, temporal) and the simulated parts ofdataset/overview/(dataset_1d, imaging_ci/calibrate) — each behind proven auto-simulate guards. The workspace now matches the other workspaces' ideal setup:dataset/**ignored by default, everything simulated on demand.Kept committed (no simulator write-site — judged by write site, not name): the five illustration assets
dataset/overview/{ccd.gif, ccd_schematic.png, cti.gif, cti_time_evolution.png, what_is_cti.png}(only evermpimg.imread/markdown-referenced), plusdataset/.gitignore.What changed
scripts/dataset_1d/**,scripts/plot/**,scripts/overview/**), all using the standardac.util.dataset.should_simulateidiom; CRLF preserved in the 6 CRLF files.overview_4_charge_injection_data.pyalso gained guards for its already-untracked uniform/non_uniform datasets. Deliberately unguarded:plot/fpa/dataset_1d/fit.py(itssimplepath is dead code) and the database/temporal scripts that read onlyoutput/.species_x1_continuum_0has zero consumers (regen witnessed via its simulator directly). Stale oddity noted: trackedspecies_x3/cti.jsonis no longer written by the current simulator and has no readers..gitignore: allowlist pins for the purged trees dropped; authoritative allowlist moved todataset/.gitignorewith per-file media pins — verified withgit check-ignore(media not ignored, purged paths ignored).simulate_datasets_include.shdeleted — itsgit add -f dataset/dataset_1d dataset/overviewwould have silently re-committed this purge (same trap the July leg hit).simulate_datasets_missing.shnow runs all 15 simulators.Validation
Per-dataset clean-tree regeneration proven (tree moved away → guarded consumer run → files re-created and read back): simple via
correction/start_here.py, temporal viatemporal/fit.py, overview trees viaoverview_5_fitting.py(pre and post purge). Post-purge runs ofmodeling/start_here.py(TEST_MODE=1, real reduced sampler),extract.py,Dataset1dPlotter.pyall exit 0;git statusstays clean after every regeneration (ignore rules absorb the products);py_compileclean on all 18 edited scripts.species_x3.pyregenerates and loads its data, then hits the documented TEST_MODE=2 tied-median trap-ordering artifact (known, per AGENTS.md — not a script bug). autocti_workspace_test has no consumers of these paths.Notebooks
18 notebook counterparts now trail their scripts; regeneration happens via the PyAutoHands release pipeline (autocti is not a generate target), same as #17.
Pre-purge main SHA for the PyAutoGut condemnation ledger:
50b701d.🤖 Generated with Claude Code
https://claude.ai/code/session_01Vu1p9rcDQGRrYddB6VWcxN
Generated by Claude Code