feat(ci): add compare workflow for PR code quality diff#9
Merged
Conversation
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Code Quality: PR Comparison13 files compared (5 added, 8 modified) %%{init: {'theme': 'neutral'}}%%
xychart-beta
title "Code Health After PR"
x-axis ["Readability", "Complexity", "Structure", "Duplication", "Naming", "Magic Numbers"]
y-axis "Score" 0 --> 100
bar [100, 17, 69, 57, 90, 81]
File changes — no changes — all metrics stableNo metric data available. Aggregate metricsAggregate Metrics
|
🟡 Code Health: B+ (78/100)
%%{init: {'theme': 'neutral'}}%%
xychart-beta
title "Code Health Scores"
x-axis ["Readability", "Complexity", "Structure", "Duplication", "Naming", "Magic Numbers"]
y-axis "Score" 0 --> 100
bar [100, 32, 84, 72, 94, 90]
🟢 Readability — A (100/100)Codebase averages: flesch_adapted=103.06, fog_adapted=2.31, avg_tokens_per_line=3.93, avg_line_length=34.30
Worst Offenders
🔴 Complexity — D- (32/100)Codebase averages: difficulty=36.52, effort=297806.01, volume=4860.41, estimated_bugs=1.62
Worst Offenders
🟡 Structure — B+ (84/100)Codebase averages: branching_density=0.14, mean_depth=4.14, avg_function_lines=11.96, max_depth=9.67, max_function_lines=26.20, variance=5.99, avg_param_count=1.12, max_param_count=2.02
Worst Offenders
🟡 Duplication — B (72/100)Codebase averages: redundancy=0.56, bigram_repetition_rate=0.20, trigram_repetition_rate=0.10
Worst Offenders
🟢 Naming — A (94/100)Codebase averages: entropy=1.28, mean=6.33, variance=14.48, avg_sub_words_per_id=1.18
Worst Offenders
🟢 Magic Numbers — A- (90/100)Codebase averages: density=0.02
Worst Offenders
|
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Extends the structure health category with five new metrics beyond indentation depth: - branching_density: control-flow keywords / non-blank lines, a language-agnostic proxy for cyclomatic complexity - avg_function_lines / max_function_lines: estimated via function-keyword detection at line start, catches god-function smells - avg_param_count / max_param_count: comma-count in function signatures, signals missing abstractions Covers Python, Ruby, JavaScript, Elixir, C#, Java, C++, Go, Rust, PHP, Swift, Shell, and Kotlin via keyword patterns and an access-modifier regex for C#/Java. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Updates the keyword blocklist in Pipeline and the operator keyword set in Halstead to cover Python, Ruby, JavaScript, Elixir, C#, Java, C++, Go, Rust, PHP, Swift, Shell, and Kotlin. Previously the list was Python-focused with a mix of Rust/Go words that leaked into identifier counts for the five primary target languages. Now keywords are organised by language with clear comments, and language-specific words (synchronized, defer, suspend, willSet, etc.) no longer pollute identifier or operand analysis. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Tests for Branching and FunctionMetrics are generated directly from the modules' public keyword lists, so adding a keyword automatically produces a new test case on the next run. - branching_keywords/0 exposes the MapSet for test generation - func_keywords/0 and access_modifiers/0 expose the function-detection keyword and C#/Java access modifier lists 54 tests total: 31 per-keyword branching tests, 10 func-keyword tests, 4 access-modifier tests, and 9 hand-written behaviour tests. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Based on observed codebase averages, the A thresholds were too generous — most metrics scored A despite averages that indicate room for improvement. Changes (A threshold → new A threshold): branching_density 0.10 → 0.08 mean_depth 4 → 3.5 avg_function_lines 10 → 8 max_depth 12 → 8 max_function_lines 25 → 20 variance 10 → 7 B/C/D thresholds tightened proportionally. avg_param_count and max_param_count unchanged — codebase average well within A. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
aspala
added a commit
that referenced
this pull request
Mar 15, 2026
* feat(ci): add compare workflow for PR code quality diff Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * refactor(grader): use pattern matching in score_metric/2 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix(formatter): guard against nil files in format_markdown/2 Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat(compare): add github format with mermaid chart and progress bars Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat(structure): add branching density and function metrics Extends the structure health category with five new metrics beyond indentation depth: - branching_density: control-flow keywords / non-blank lines, a language-agnostic proxy for cyclomatic complexity - avg_function_lines / max_function_lines: estimated via function-keyword detection at line start, catches god-function smells - avg_param_count / max_param_count: comma-count in function signatures, signals missing abstractions Covers Python, Ruby, JavaScript, Elixir, C#, Java, C++, Go, Rust, PHP, Swift, Shell, and Kotlin via keyword patterns and an access-modifier regex for C#/Java. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * chore(keywords): expand keyword coverage to 13 languages Updates the keyword blocklist in Pipeline and the operator keyword set in Halstead to cover Python, Ruby, JavaScript, Elixir, C#, Java, C++, Go, Rust, PHP, Swift, Shell, and Kotlin. Previously the list was Python-focused with a mix of Rust/Go words that leaked into identifier counts for the five primary target languages. Now keywords are organised by language with clear comments, and language-specific words (synchronized, defer, suspend, willSet, etc.) no longer pollute identifier or operand analysis. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * test(metrics): add auto-generated keyword coverage tests Tests for Branching and FunctionMetrics are generated directly from the modules' public keyword lists, so adding a keyword automatically produces a new test case on the next run. - branching_keywords/0 exposes the MapSet for test generation - func_keywords/0 and access_modifiers/0 expose the function-detection keyword and C#/Java access modifier lists 54 tests total: 31 per-keyword branching tests, 10 func-keyword tests, 4 access-modifier tests, and 9 hand-written behaviour tests. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * feat(compare): add 🟢/🔴 direction emoji to aggregate metric deltas Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * chore(structure): tighten structure category thresholds Based on observed codebase averages, the A thresholds were too generous — most metrics scored A despite averages that indicate room for improvement. Changes (A threshold → new A threshold): branching_density 0.10 → 0.08 mean_depth 4 → 3.5 avg_function_lines 10 → 8 max_depth 12 → 8 max_function_lines 25 → 20 variance 10 → 7 B/C/D thresholds tightened proportionally. avg_param_count and max_param_count unchanged — codebase average well within A. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
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
.github/workflows/compare.ymlto runcodeqa compareon every PRgit merge-base HEAD <base.sha>and compares it against PR HEADTest plan