Match attribute groups by whole name, not by word, when looking up valid terms

オープン
  • 何か問題がありますか？
#299 コメント 0 件 リアクション 0 件 担当者 0 名 GitHub で見る

メンテナーはふだん 1 日以内に返信

まだ誰も着手していません。

評価

難易度
4/5
見積もり時間
3〜5日
初心者へのやさしさ
57/100
issue の種類
バグ
明瞭さ
明確に書かれている
活発さ
活発
技術スタック
r
領域
data

調査の方向性

Start with .validAttributes() in R/physiologies.R and the multistate Attribute_value check in tests/testthat/test-importBugphyzz.R; then review inst/scripts/stop_on_unmapped.R and taxPPro's filterDataDiscrete() so the checks and export filter agree. Check whether any shape rows in the sheets or sources use endospore, branched, or cell chain. Done when group matching uses exact names across all four locations, tests confirm shape and halophily exclude unrelated terms, and the export still handles current source rows.

索引モデルが issue の本文から書いたものです。

説明

The valid terms for a physiology are looked up in inst/extdata/attributes.tsv by searching for the physiology name as a whole word in attribute_group (\b<name>\b). A physiology whose name is a whole word inside another group name therefore also gets that group's terms.

Two cases exist now:

  • shape also accepts three terms that belong only to spore shape: endospore, branched, cell chain.
  • halophily also accepts the terms of halophily class (halophilic, halotolerant, non-halophilic), added in #298.

Neither does harm today. halophily is a range attribute, so it is never checked against the vocabulary, and no shape source uses those three terms. The risk is a later discrete physiology whose wrong values pass the vocabulary check unnoticed.

Proposed fix

Match the whole group name: split attribute_group on ; and test for equality, in place of the regex.

Where the match is made

  1. bugphyzz, .validAttributes() in R/physiologies.R.
  2. bugphyzz, tests/testthat/test-importBugphyzz.R, the Attribute_value check for multistate attributes (grepl(unique(x$Attribute), attribute_group), with no word boundary at all).
  3. bugphyzzExports, stop_on_unmapped() in inst/scripts/stop_on_unmapped.R.
  4. taxPPro, filterDataDiscrete(), which decides the rows the export keeps.

The four should change together, so that the check in bugphyzz and the filter in the export keep agreeing. A test that .validAttributes("halophily") excludes the class terms and that .validAttributes("shape") excludes endospore would pin it.

Check before changing

With an exact match, shape loses the three terms above. Confirm that no shape row in the sheets or sources uses them, otherwise the export would stop on unmapped values. On the current attributes.tsv these two physiologies are the only ones whose valid terms change.

Raised in the review of #298.

Written with Claude Code.

主要言語
R
スター
6
フォーク
6
平均マージ
3時間 17分
マージ済み PR（30日）
20

環境構築

はじめの一歩

  1. issue を最後まで読み、次にプロジェクトのコントリビューションガイドを読みます。
  2. 着手することを issue にコメントします — 二人が同じ作業をするのを防げます。
  3. リポジトリをフォークし、ブランチを切って変更します。
  4. issue 番号を参照したプルリクエストを送ります。

waldronlab/bugphyzz のほかの issue

waldronlab/bugphyzz の issue をすべて見る

似ている issue

R の issue をもっと見る

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。