Hidden enum values#628
Open
zaucker wants to merge 6 commits into
Open
Conversation
Two model versions can now share a dataset even when their enum option sets diverge. A foreign-version value is accepted by the validator, rewritten to the canonical local key when formulas evaluate, and surfaced in the GUI as orange (mapped) — informational only, never blocks calculation. Backend - LanguageParser: parse `accepts = a, b, c` inside an enum option as a comma-separated alias list, preserving underscores. - Model::Input: store %!enum-aliases; add canonical-enum-value, is-mapped-enum-value, and extend is-valid-enum-value to accept aliases. Expose `:enumAliases` via as-hash for the frontend. - Model::Module: canonicalize-input-hash rewrites alias values to the local key for formula evaluation only — storage stays raw, so re-opening in the other version still sees its own value. - Model + Inputs: call canonicalize-input-hash at the two formula-eval entry points (run-as-single, apply-defaults). - t/enum-aliases.rakutest + t/test-data/EnumAliases.nhd: 14 tests covering parser, validator, canonical lookup, and module rewrite. Frontend - NavFolder: track __mapped per folder and aggregate up the tree. Icon precedence: red (incomplete) > orange (mapped) > green (complete). Mapped does not gate the outputEnabled event. - PropTable: extend the cell-renderer replaceMap so a stored alias value renders with the canonical option's localized label. - cellrenderer/input/Replace: orange cell background (#f9d8a4 / #fbeacb) when the value is in enumAliases. - orange-dot.png + orange-circle.png: new 18x18 icons matching the existing red/green set. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The frontend's setData() guarded against unknown enum values in the loaded dataset by alerting "no match for >X< in >a,b,c<, should be removed from dataset" and dispatching deleteInstanceData. That gate ran before any of the alias display machinery added in 9508f66, so a foreign-version value declared as `accepts =` in the .nhd was purged at load time even though the backend validator and the cell renderer already understood it. Treat enumAliases membership as "found" — the value passes through to setValue unchanged (storage stays raw, preserving round-trip); the existing replaceMap entries handle the canonical-label render and the orange highlight. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Without SOURCE_MODE, Cro's static-content routes serve from public/ — the production build target, which lags behind any source change until a full minified rebuild runs. With SOURCE_MODE=1 the same routes serve from frontend/compiled/source/, so a single `npx qx compile` (or live `--watch`) is enough to see edits in the browser. Matches the dev workflow the rest of this branch was designed for. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
NavBar.__getInputVariables walks the JSON returned from getInputVariables and constructs each Variable's metaData. The enum branch copied only options and optionsLang, dropping the enumAliases map served alongside. Result: every downstream check (metaData.enumAliases.hasOwnProperty(value) in NavFolder.setData, NavFolder.isComplete, the Replace cell renderer, the PropTable replaceMap) saw `undefined` and short-circuited — so a stored alias value was rejected at load with "no match for >X< in >a,b,c<, should be removed from dataset" even though the rest of the alias machinery was wired up correctly. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The cell-editor factory iterates every metaData key and falls into "This should not happen: unknown metaData value: X" on anything not explicitly cased. enumAliases is consumed by the renderer and by NavFolder, so the editor itself has nothing to do — just acknowledge the key. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
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
1 participant
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.
To handle using the same dataset on both 6.5.2 and 7.0.0 model versions, we have to map free_range enums from Livestock::Poultry::Grazing
This PR marks mapped values in orange, but does not block calculation. Upon storing a new value in the currently opened model, the orange marker goes away.