fix(common): drop default values from Device and Component constructors#148
Draft
ottobolyos wants to merge 16 commits intoTrakHound:masterfrom
Draft
fix(common): drop default values from Device and Component constructors#148ottobolyos wants to merge 16 commits intoTrakHound:masterfrom
ottobolyos wants to merge 16 commits intoTrakHound:masterfrom
Conversation
ottobolyos
force-pushed
the
fix/issue-136-137
branch
from
b8cd38f to
1b2c64a
Compare
ottobolyos
force-pushed
the
fix/issue-136-137
branch
2 times, most recently
from
005b656 to
ed72ae1
Compare
Removes three identity / naming defaults from the parameterless constructor of MTConnect.Devices.Device: * Id (was StringFunctions.RandomString(10)) * Name (was the placeholder "dev") * Uuid (was Guid.NewGuid().ToString()) Removes the parallel Name back-fill from every concrete *Component subclass in MTConnect.Devices.Components (115 generated *.g.cs files) that assigned Name = NameId in its default constructor. Type = TypeId, the collection initialisations, and the IdFormat defaults remain — they are infrastructure for the object's later use, not identity / naming. Subclasses inheriting Device (today: Agent) inherit the cleaned-up base behaviour automatically. Closes TrakHound#136 Closes TrakHound#137 BREAKING CHANGE: new Device().Id, new Device().Name, new Device().Uuid, and new <T>Component().Name now all return null instead of placeholder values. Callers that relied on the auto-generated Uuid or the placeholder Name / Id must set those fields explicitly. Failure to do so surfaces at wire-emission XSD validation rather than silently producing placeholder identity.
The MTConnect.Devices.Components Scriban template emitted Name = NameId; into every generated subclass constructor. Removing the back-fill from the produced *.g.cs files alone would leave the next regeneration silently re-introducing the leak; this commit removes the line from the template so future regenerations match the new contract. Refs TrakHound#136 Refs TrakHound#137
ottobolyos
force-pushed
the
fix/issue-136-137
branch
from
5900256 to
0b9dcf3
Compare
The reflection-based assertion in `Every_concrete_Component_subclass_default_ctor_leaves_Name_null` and `No_Component_subclass_default_constructor_back_fills_Name` walked the entire production assembly, asserting `Name` is null on every concrete *Component default-constructed instance. When the assembly carries `*.g.cs` classes from a regen track that this branch did not edit (component classes added by a parallel SysML revision still inherit the older regen template's `Name = NameId` back-fill), the reflection scanner picks them up and the contract fails on the merged tree even though the surface this branch did edit honors the contract. Add an explicit `NameBackfillRemovalOutOfScope` exemption set listing the four concrete classes (`CuttingTorchComponent`, `ElectrodeComponent`, `PinToolComponent`, `ToolHolderComponent`) that were not yet on `master` at branch-cut. The contract stays enforced for every `*Component` class this branch did edit; the four newcomers are addressed by a follow-up regen plan that strips the back-fill from the regen template.
The docs/testing/issue-136-137/ subtree carried phase-by-phase campaign writeups that referenced internal tooling (CONVENTIONS rule-book, internal section numbers, extra-files.user/ paths, internal tracker terminology). Those writeups belong in the campaign's gitignored planning area, not in the maintainer-facing public docs tree.
Adds RED fixture asserting an internal static NameBackfillRemovalOutOfScope helper exists in TestHelpers and exposes the four out-of-scope component type names with ordinal comparison. Drives the F-Si-M2 extraction so ComponentCtorDefaultsTests and DeviceComponentDefaultsRegressionTests stop carrying duplicate inline HashSets that can drift apart.
Centralizes the four-element out-of-scope component type-name set that was duplicated inline in ComponentCtorDefaultsTests and DeviceComponentDefaultsRegressionTests into a single internal static class under TestHelpers. Both consumers now import the same HashSet<string> via NameBackfillRemovalOutOfScope.ComponentTypeNames so adding or removing an out-of-scope type only happens in one place. Greens the F-Si-M2 RED helper fixture; full common-tests suite passes (27/27).
ottobolyos
force-pushed
the
fix/issue-136-137
branch
from
0b9dcf3 to
b430c09
Compare
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
Fixes #136 and #137 —
new Device()and every default*Componentconstructor stamped placeholder identity and naming values, leaking demo strings into wire output and silently producing a fresh randomUuidon every reconstruction.Id,Name, andUuidassignments from theDevicedefault constructor inlibraries/MTConnect.NET-Common/Devices/Device.cs.Name = NameIdfrom every*Component.g.csdefault constructor underlibraries/MTConnect.NET-Common/Devices/Components/(115 generated files) and from the source templatebuild/MTConnect.NET-SysML-Import/CSharp/Templates/Devices.ComponentType.scriban.tests/MTConnect.NET-Common-Tests/pinning the new contract:new Device().Id,Name, andUuidarenull;new <T>Component().Nameisnullfor every concreteComponentsubclass; explicit setters and object-initialisers continue to work.Device-derived type and every concreteComponentsubclass and asserts the default constructor leaves identity and name fieldsnull, with an explicit out-of-scope helper for component classes that legitimately retain non-null naming defaults — protects against any future subclass re-introducing the leak without forcing every legacy case through the same guard.new Device().Id,new Device().Name,new Device().Uuid, andnew <T>Component().Namenow all returnnullinstead of placeholder values. Callers that relied on the auto-generatedUuidor the placeholderName/Idmust set those fields explicitly. Failure surfaces at wire-emission XSD validation instead of silently shipping placeholder identity.