Repository navigation
Markup-driven lipid CID spectrum generator (experimental) - #806
Draft
YukiMatsuzawa wants to merge 93 commits into
Draft
YukiMatsuzawa wants to merge 93 commits into
YukiMatsuzawa wants to merge 93 commits into
Conversation
Added the `FormulaStringParser` class to parse chemical formula strings into a dictionary representation, with methods for validation and conversion. Introduced corresponding tests in `FormulaStringParserTests` to ensure functionality. Updated `CompMs.Common.Lipidomics.SourceGenerator.csproj` with an `InternalsVisibleTo` directive for test project access.
Refactored `ToIon` method in `LipidSpectrumGeneratorTypeGenerator.cs` to use a new `ParseElements` helper method, improving modularity and readability. Added `ParseElements` as a private static method to encapsulate element parsing logic. Updated `MsdialLipidCidSpectrumModel.xml` to include new `<NL>` and `<PI>` elements with detailed metadata and commented-out alternative representations for future extensibility.
Updated `_peakElements` to include `Charge` and `Formula`, enabling dynamic `MZ` calculation when `MZ` is `0`. Reused `ParseElements` for `Formula` processing. Added a test case in `MsdialLipidCidSpectrumModel.xml` to validate the new functionality. Enhances flexibility and accuracy in lipidomics data analysis.
Introduced the `FormulaSentenceParser` class in the `CompMs.Common.Lipidomics.SourceGenerator` namespace to parse chemical formula sentences into structured data. Key features include regex-based parsing, support for terms and factors, and internal classes (`Sentence`, `Context`, `SubVar`) for data representation. Added unit tests in `FormulaSentenceParserTest` to validate parsing functionality with multiple test cases.
Updated the metadata key in `LipidSpectrumGeneratorTypeGenerator.cs` to use `build_metadata.AdditionalFiles.Category` for accurate retrieval. Enhanced `CompMs.Common.Lipidomics.SourceGeneratorTests.csproj` by: - Adding an `<Import>` for `SourceGenerator.props` to integrate build configurations. - Including `MsdialLipidCidSpectrumModel.xml` with a `<Category>` of `Cid` for better test resource management.
The `Context` class was refactored into its own file in the `CompMs.Common.Lipidomics.SourceGenerator` namespace, improving modularity and readability. It now resolves terms using constants and regex-based rules. A new `ContextTests` class was added to validate its functionality with unit tests. The `Context` class was removed from `FormulaSentenceParser.cs`.
Added a `_vars` dictionary and `RegisterHandle` to `Context` for scoped `SubVar` management. Introduced `Register` and `Resolve` methods to handle dynamic variable resolution. Updated `Sentence` and `SubVar` to support `Sentence`-based values and caching. Refactored `FormulaSentenceParser` to use `SubVar` objects and removed unused methods and dependencies. General improvements to enable dynamic and scoped variable handling.
Refactored the source generator to dynamically handle constants defined in an external `Constants.xml` file, generating a `ConstantFormulas` class at build time. Replaced the `EmitStandardFormulas` method with the new `EmitConstants` method and updated related namespaces. Adjusted project files to support the new `IsConstants` metadata for better build integration. Enhanced test coverage by parameterizing tests using `DynamicData` and improving test data management. Refactored test assertions to validate generated spectra against reference data. Improved code clarity, maintainability, and consistency across the project.
Updated `Resolve` methods in the `Context` class to simplify output by omitting multipliers when they equal `1`. Adjusted existing tests and added new test cases in `ContextTests` to validate the updated behavior, including complex formulas.
A new static method `IsMarkupFormula` was added to the `FormulaStringParser` class to validate markup formula strings using a regex pattern. The method takes a `rawMarkup` string and an array of `constants` as input and returns a boolean indicating whether the input matches the expected pattern. A corresponding test method, `IsMarkupFormulaTest`, was added to the `FormulaStringParserTests` class. It uses `[DataTestMethod]` and `[DataRow]` attributes to test various scenarios, including valid and invalid markup strings. The test ensures the method works correctly with constants like `["Carbon", "Hydrogen", "Oxygen"]`.
Added a static `IsMarkupFormula(string rawMarkup)` method in `FormulaStringParser` to validate markup formulas using regex. Refactored the existing `IsMarkupFormula` method to use `Regex.IsMatch` for better performance. Updated tests in `FormulaSentenceParserTest` to cover new functionality.
Added the `ParseMarkupFormula` method to `FormulaStringParser` for parsing markup strings into dictionaries using `XDocument` and LINQ. Enhanced `FormulaStringParserTests` with unit tests for this method, including cases for valid inputs, empty strings, and invalid values. Fixed a typo in `IsMarkupFormulaTest` test data. Added `using System.Xml.Linq;` to support XML parsing.
Refactored the `Context` class to improve formula parsing: - Extracted formula construction logic into a new `ConstructFormula` method for modularity and reusability. - Updated `Resolve` to use `ConstructFormula` and handle markup formulas via `FormulaStringParser`. - Added a `GetConstants` method to expose `_constants` keys. Added a new test case in `ContextTests` to validate markup formula resolution. Improved code readability and maintainability.
Added a check in the `Parse` method of `FormulaSentenceParser` to identify markup formulas using `FormulaStringParser.IsMarkupFormula`. If identified, the method now returns a `Sentence` object with the `Raw` property set to the input string and the `Terms` property initialized with a single `Term` containing the same `Raw` value and a count of `1`.
Refactored `FormulaSentenceParser` to handle nullable `SubVar` and added a `LoadElements` method for nested structures. Improved regex patterns in `FormulaStringParser` for stricter XML-like validation and updated related test cases. Enhanced `LipidSpectrumGeneratorTypeGenerator` with a `Context` object for formula resolution and constant handling. Updated `Constants.xml` for shorthand constants and consistency. Modernized code with C# features and improved readability.
# Conflicts: # src/Common/CommonMVVM/ChemView/ChemOntologyColor.cs # src/Common/CommonStandard/Algorithm/Scoring/MsScanMatching.cs # src/Common/CommonStandard/Database/LbmQuery.cs # src/Common/CommonStandard/Enum/LbmQuery.cs # src/Common/CommonStandard/Lipidomics/Lipidomics.cs # src/Common/CommonStandard/Lipidomics/LipidomicsConverter.cs # src/Common/CommonStandard/Lipidomics/MsmsCharacterization.cs # src/Common/CommonStandard/Molecular spectrum networking/MolecularSpectrumNetworkingCommon.cs # src/Common/CommonStandard/Ontology/ChemOntologyColorConverter.cs # src/Common/CommonStandard/Resources/LbmQueries.txt # src/Common/CommonStandard/Resources/LipidClassProperties.csv # src/MSFINDER/MsfinderCommonStandard/Resources/LipidQueryMaster.txt
…ectrum_generator_by_markup
Drop a 90MB temporary resource file and three Visual Studio wpftmp scratch project files that were committed by mistake. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
…ication-catchup-21d98d
Parse_FactorIsCorrectTest compared an int[] expectation against the double factors FormulaSentenceParser actually produces, so CollectionAssert.AreEqual failed on boxed int vs double even when the values matched. Switch expectations to double and correct the "H2O/2" case, whose expected factor was 1 instead of the correct 0.5. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
YukiMatsuzawa
added this pull request to stack #807
September 24, 2026 09:20
…ster-8a243c # Conflicts: # tests/MSDIAL5/MsdialCoreTestApp/Properties/Resources.Designer.cs
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
Long-running branch implementing a markup-XML-driven lipid CID spectrum generator: a Roslyn source generator that reads
MsdialLipidCidSpectrumModel.xml(precursor/class-ion/chain-ion rules per lipid class) plusConstants.xml, and emits one{Class}CidLipidSpectrumGeneratorper lipid class - intended as a declarative alternative/successor to the hand-written{Class}SpectrumGeneratorclasses inFacadeLipidSpectrumGenerator.Opening as draft to track this against
masterand keep it visible/reviewable rather than parked. #805 (claude/lipid-annotation-design-e016f8-> this branch) adds a SpectrumViewer tool for authoring/validating these rules and will land here once ready.Status: draft / WIP
Not yet wired into any production annotation path - see #805 for current known gaps (an unrelated pre-existing
TotalChainplasmalogen-rendering issue, and lipid classes with no matchingLbmClass).Co-Authored-By: Claude Sonnet 5 noreply@anthropic.com