Hacktoberfest 2026：メンテナが10月に向けて印を付けた、オープンで初心者向けの issue。 Hacktoberfest の issue を見る

allocate accepts two definitions as its ends (ReferenceSubsetting::referencedFeature must be a Feature)

クローズ
  • 何か問題がありますか？
#646 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
68/100
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
go
領域
compilers

調査の方向性

Start by reproducing the two allocate examples from the issue, then trace allocate relationship endpoint resolution and the ReferenceSubsetting validation that types referencedFeature as Feature. Done means definition endpoints produce a diagnostic like sysml-toolkit, while the usage form remains accepted; check the neighboring perform validation for comparison.

索引モデルが issue の本文から書いたものです。

説明

Summary

allocate accepts two definitions as its ends (rather than two features/usages) with no diagnostic, even though the ends resolve through ReferenceSubsetting, whose referencedFeature is typed Feature — a Definition is not a Feature. sysml-toolkit rejects the same source.

Version: OpenSysML v0.9.0. sysml-toolkit v0.9.1 with the standard library rejects the same source (see below).

Observed

package P { action def A; part def H; allocate A to H; }

Loads with ok=True and no diagnostic in OpenSysML v0.9.0.

With usages instead of definitions:

package P { part def S { action a : A; part h : H; allocate a to h; } }

loads cleanly in both tools, as expected.

sysml-toolkit v0.9.1 (check --lib <sysml.library>) rejects the definition form:

error: ReferenceSubsetting::referencedFeature must refer to a Feature [relationship-endpoint-metaclass]

OpenSysML already distinguishes definitions from usages in the neighboring perform case — perform A; naming an action definition is (correctly) rejected, since perform must reference a usage — so this looks like an inconsistency in how the same distinction is applied to allocate.

Reference

KerML 1.1 Beta 2, 8.3.3.3.9 ReferenceSubsetting: referencedFeature : Feature {redefines subsettedFeature} — the attribute's declared type is Feature. This section's own Constraints subsection is empty ("None"): there's no separate OCL rule needed, because the requirement is the attribute's own metamodel typing. A part def/action def is a Definition-kind element, not a Feature, so it cannot satisfy referencedFeature's type at all.

SysML v2.0 (formal/2026-03-02), 7.15.2 Allocation Definitions and Usages: the section's own worked example only ever allocates usages (allocate logical.component to physical.assembly inside an allocation definition, and allocate logical ::> system to physical ::> device at usage level) — never two bare definitions.

Note on #95

This is a different claim from #95 ("Subsetting type conformance is not checked"), which your team investigated and correctly closed as not-a-bug for general Subsetting (no type-conformance constraint exists there per KerML 8.3.3.3.10). ReferenceSubsetting's referencedFeature attribute is a different, more basic thing: it isn't asking whether two feature types conform to each other, it's that the referenced element must be a Feature at all, which a Definition structurally is not. We think this survives #95's reasoning, but wanted to flag the connection explicitly rather than have it look like the same question re-asked.

Request

Diagnose an allocate (or any ReferenceSubsetting-based relationship) whose ends are not features, the way sysml-toolkit already does.

主要言語
Go
スター
33
フォーク
8
平均マージ
10時間 12分
マージ済み PR（30日）
598

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。

Open-MBEE/OpenSysML のほかの issue

Open-MBEE/OpenSysML の issue をすべて見る

似ている issue

Go 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。