docs: spec + implementation plan for --amount <quantity><unit> flag#23
Open
phitoduck wants to merge 3 commits into
Open
docs: spec + implementation plan for --amount <quantity><unit> flag#23phitoduck wants to merge 3 commits into
--amount <quantity><unit> flag#23phitoduck wants to merge 3 commits into
Conversation
Capture findings + implementation plan in docs/ so the work survives context loss between sessions. Backed by the user-provided HAR capture of the official UI logging the same Trader Joe's tomato soup at 490 mL. The spec covers: - Wire-level evidence: the HAR's `updateFoodLogEntry` payload and what each field means (FoodServingSize.f4 = conversion factor, f5 = raw user input, FoodMeasure.ordinal = user's chosen unit). - Concrete code citations for where to change `_build_log_payload`, the `log` CLI command, and the new `_units.py` module. - 4-phase implementation plan with effort estimates. - Edge case the user explicitly called out: foods that don't support the requested unit. Detection via `conversion_factor` returning None, fallback via `foods.find_supporting_alternatives` showing a numbered list of candidates that *do* support the unit. - Verification: unit tests pinning the wire bytes to the HAR snippet, dry-run grep check, live log + app rendering check. - Risks and follow-ups: diary readback semantics after #22 (separate fix), the partial-decoder concern when probing alternatives. Cited code: entries.py:35-181, cli.py:398-478, foods.py:268-288, _models.py:23-67. HAR quotes inlined.
Per maintainer policy, the evidence file is kept outside the repo. The wire-byte facts (cup→mL conversion factor 236.5882365, FoodServingSize field layout, per-serving HashMap pattern) are the protocol-level findings and remain in the spec verbatim — only the references to *where* the evidence file lives are removed. Affected: - docs/amount-flag-spec.md: rewrite "captured a HAR" → "observed in manual testing"; rewrite citation index entry; tweak verification steps that compared to "the HAR". - src/lose_it_utils/client/entries.py: rewrite the comment in _build_log_payload that mentioned the HAR capture; keep the per-serving vs pre-scaled explanation.
…od-id spec Per maintainer feedback: 1. The unit-conversion spec switches from a single `--amount 490mL` flag to a two-flag pair `--serving-amount 490 --serving-unit mL`. Rationale documented in the spec: each flag is a familiar `<flag> <value>` shape, the unit is a discrete enum that can be tab-completed, and unit/amount-only mistakes surface as flag validation errors rather than silent regex failures. Rename: docs/amount-flag-spec.md → docs/serving-unit-spec.md. The `_units` module no longer needs a `parse_amount` regex — `resolve_unit(str) → ord` is enough. 2. New spec: docs/food-id-flow-spec.md covers the broader CLI flow problem. Today `lose-it log "1 tortilla" --pick N` relies on a natural-language query the CLI can't disambiguate plus a `--pick` index that drifts. Adds `--food-id <32-char hex>` that bypasses search entirely by calling a new `getFood` RPC wrapper (wire shape already observed in manual UI testing — quoted verbatim in the spec) and using the food's stable 16-byte PK as the identifier. Includes the encoding helpers (_ids.py), the search-output changes to surface the ID, edge-case behaviour for invalid/missing IDs, a 5-phase implementation plan, and how the two specs compose (`--food-id <hex> --serving-amount 490 --serving-unit mL` is the fully-disambiguated form).
6 tasks
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.
Summary
Captures the work to add a general
--amount <quantity><unit>flag (e.g.--amount 490mL,--amount 86g) tolose-it log. This is a SPEC ONLY — no code changes. The implementation is intended for a separate PR. Filed now so the work survives a context-loss boundary between sessions.What the spec covers
_schemas.json. Pins what eachFoodServingSize.f0..f5field means and confirms the conversion factor (f4=236.588) is the universal US cup→mL constant.entries._build_log_payload(lines 35-181, FoodServingSize block at 164-178);cli.log(--gramshandling at 398-409 and 455-478);foods.get_unsaved_food_log_entry(268-288);_models.UnsavedFoodLogEntry(23-34) andFoodLogEntry.calories(62-67)._units.pywithCONVERSIONStable,UNIT_ALIASES,parse_amount,conversion_factor. Universal volumetric constants;ozdeliberately rejected as ambiguous.--amount 50g). Detection viaconversion_factor()returningNone; fallback viafoods.find_supporting_alternativesshowing 2-5 candidates from the same search that DO support the unit, each annotated withserving_qtyso the catalog's known traps are visible.27|2.071…|1|28|11|1.0|236.588|490), dry-run grep check, live-log + app rendering check.FoodLogEntry.caloriesshows per-serving until we multiply byservingsin a follow-up); partial-decoder concern when probing alternatives.Why ship the spec separately
The session that produced PRs #18-#22 is approaching its context budget. Capturing the findings + plan to disk lets the next session land the feature without re-deriving the HAR analysis or hunting line numbers.
Test plan
prek run --files docs/amount-flag-spec.mdclean.