fix: metric value semantics, MI formula, and tool-bound doc comments

.bca-baseline.toml

@@ -11,6 +11,7 @@ version = 6
 tier = "soft"
 headroom = 0.95
 mi_charges_docs = false
+mi_formula = 2
 
 [[entry]]
 path = "big-code-analysis-ast/src/alterator.rs"
@@ -226,49 +227,13 @@ value = 5.0
 path = "big-code-analysis-ast/src/checker.rs"
 qualified = "<file>"
 metric = "loc.ploc"
-value = 568.0
+value = 584.0
 
 [[entry]]
 path = "big-code-analysis-ast/src/checker.rs"
 qualified = "<file>"
 metric = "loc.sloc"
-value = 1373.0
-
-[[entry]]
-path = "big-code-analysis-ast/src/checker.rs"
-qualified = "Checker::holds_doc_comment_in_body"
-metric = "nargs"
-value = 5.0
-
-[[entry]]
-path = "big-code-analysis-ast/src/checker/groovy.rs"
-qualified = "GroovyCode::holds_doc_comment_in_body"
-metric = "nargs"
-value = 5.0
-
-[[entry]]
-path = "big-code-analysis-ast/src/checker/java.rs"
-qualified = "JavaCode::holds_doc_comment_in_body"
-metric = "nargs"
-value = 5.0
-
-[[entry]]
-path = "big-code-analysis-ast/src/checker/kotlin.rs"
-qualified = "KotlinCode::holds_doc_comment_in_body"
-metric = "nargs"
-value = 5.0
-
-[[entry]]
-path = "big-code-analysis-ast/src/checker/php.rs"
-qualified = "PhpCode::holds_doc_comment_in_body"
-metric = "nargs"
-value = 5.0
-
-[[entry]]
-path = "big-code-analysis-ast/src/checker/rust.rs"
-qualified = "RustCode::holds_doc_comment_in_body"
-metric = "nargs"
-value = 5.0
+value = 1411.0
 
 [[entry]]
 path = "big-code-analysis-ast/src/getter.rs"
@@ -312,6 +277,48 @@ qualified = "RubyCode::get_op_type"
 metric = "halstead.effort"
 value = 120642.89096069236
 
+[[entry]]
+path = "big-code-analysis-ast/src/lang_helpers/doc_comment.rs"
+qualified = "<file>"
+metric = "loc.ploc"
+value = 632.0
+
+[[entry]]
+path = "big-code-analysis-ast/src/lang_helpers/doc_comment.rs"
+qualified = "bind_leading"
+metric = "nargs"
+value = 5.0
+
+[[entry]]
+path = "big-code-analysis-ast/src/lang_helpers/doc_comment.rs"
+qualified = "bound_doc_run"
+metric = "halstead.effort"
+value = 74870.34845140431
+
+[[entry]]
+path = "big-code-analysis-ast/src/lang_helpers/doc_comment.rs"
+qualified = "bound_doc_run"
+metric = "nexits"
+value = 5.0
+
+[[entry]]
+path = "big-code-analysis-ast/src/lang_helpers/doc_comment.rs"
+qualified = "joined_run"
+metric = "nargs"
+value = 5.0
+
+[[entry]]
+path = "big-code-analysis-ast/src/lang_helpers/doxygen.rs"
+qualified = "literal_end"
+metric = "cyclomatic"
+value = 19.0
+
+[[entry]]
+path = "big-code-analysis-ast/src/lang_helpers/doxygen.rs"
+qualified = "literal_end"
+metric = "nexits"
+value = 6.0
+
 [[entry]]
 path = "big-code-analysis-ast/src/node.rs"
 qualified = "Node<'a>"
@@ -364,13 +371,13 @@ value = 7.0
 path = "big-code-analysis-cli/src/baseline.rs"
 qualified = "Baseline::from_str"
 metric = "halstead.effort"
-value = 61984.213363755734
+value = 56959.04904936
 
 [[entry]]
 path = "big-code-analysis-cli/src/baseline_diff.rs"
 qualified = "BaselineDiff::compute"
 metric = "halstead.effort"
-value = 118241.28751025953
+value = 96116.62210735636
 
 [[entry]]
 path = "big-code-analysis-cli/src/check_format.rs"
@@ -606,12 +613,6 @@ qualified = "write_table_with_tooltips"
 metric = "nargs"
 value = 6.0
 
-[[entry]]
-path = "big-code-analysis-cli/src/html_report/sections.rs"
-qualified = "<file>"
-metric = "loc.ploc"
-value = 529.0
-
 [[entry]]
 path = "big-code-analysis-cli/src/html_report/sections.rs"
 qualified = "write_language_section"
@@ -634,7 +635,7 @@ value = 49443.19551903411
 path = "big-code-analysis-cli/src/markdown_report.rs"
 qualified = "<file>"
 metric = "loc.ploc"
-value = 575.0
+value = 633.0
 
 [[entry]]
 path = "big-code-analysis-cli/src/markdown_report.rs"
@@ -646,7 +647,7 @@ value = 5.0
 path = "big-code-analysis-cli/src/markdown_report.rs"
 qualified = "extract_summaries_inner"
 metric = "abc"
-value = 39.153543900903784
+value = 40.124805295477756
 
 [[entry]]
 path = "big-code-analysis-cli/src/markdown_report.rs"
@@ -684,30 +685,24 @@ qualified = "select_for"
 metric = "nargs"
 value = 7.0
 
-[[entry]]
-path = "big-code-analysis-cli/src/threshold_suggestion.rs"
-qualified = "closest_names"
-metric = "halstead.effort"
-value = 59984.93899693509
-
-[[entry]]
-path = "big-code-analysis-cli/src/threshold_suggestion.rs"
-qualified = "edit_distance_with_cutoff"
-metric = "halstead.effort"
-value = 70177.90308128356
-
 [[entry]]
 path = "big-code-analysis-cli/src/thresholds.rs"
 qualified = "<file>"
 metric = "loc.ploc"
-value = 554.0
+value = 557.0
 
 [[entry]]
 path = "big-code-analysis-cli/src/thresholds.rs"
 qualified = "ThresholdSet::evaluate_with_policy"
 metric = "nargs"
 value = 5.0
 
+[[entry]]
+path = "big-code-analysis-cli/src/thresholds.rs"
+qualified = "parse_cli_threshold"
+metric = "nexits"
+value = 5.0
+
 [[entry]]
 path = "big-code-analysis-cli/src/vcs_command.rs"
 qualified = "<file>"
@@ -886,7 +881,7 @@ value = 6.0
 path = "big-code-analysis-py/src/sarif.rs"
 qualified = "resolve_thresholds"
 metric = "nexits"
-value = 6.0
+value = 7.0
 
 [[entry]]
 path = "big-code-analysis-py/src/types_codegen.rs"
@@ -1210,7 +1205,7 @@ value = 5.0
 path = "src/metrics/loc/perl.rs"
 qualified = "PerlCode::compute"
 metric = "halstead.effort"
-value = 59310.25237096207
+value = 62449.60005820913
 
 [[entry]]
 path = "src/metrics/loc/perl.rs"
@@ -1282,7 +1277,7 @@ value = 6.0
 path = "src/ops.rs"
 qualified = "ops_inner"
 metric = "halstead.effort"
-value = 84452.11438842113
+value = 86380.09140562497
 
 [[entry]]
 path = "src/output/checkstyle.rs"
@@ -1416,6 +1411,18 @@ qualified = "open_func_space"
 metric = "nargs"
 value = 6.0
 
+[[entry]]
+path = "src/suggestion.rs"
+qualified = "closest_names"
+metric = "halstead.effort"
+value = 59984.93899693509
+
+[[entry]]
+path = "src/suggestion.rs"
+qualified = "edit_distance_with_cutoff"
+metric = "halstead.effort"
+value = 67009.44460326429
+
 [[entry]]
 path = "src/suppression.rs"
 qualified = "parse_native"

.claude/rules/testing.md

@@ -53,9 +53,26 @@ gets reached for, and it destroyed an entire uncommitted rewrite a third
 time during #1219. Read the file once into a variable, write each
 perturbation from that string, and restore from it in a `finally` — the
 backup then lives in memory and cannot be defeated by the working tree
-changing underneath. Put the driver in a file rather than nesting quotes
-through zsh into `python3 -c`, which is how those edits came to be
-applied inconsistently in the first place.
+changing underneath. Write that same copy to the driver's scratch
+directory before the first edit, too, for the case below. Put the
+driver in a file rather than nesting quotes through zsh into
+`python3 -c`, which is how those edits came to be applied
+inconsistently in the first place.
+
+**An interrupted sweep leaves a mutant in the tree, and it looks like
+a fix in progress.** A driver killed between a perturbation and its
+restore — a server restart, a cancelled agent — never reaches its
+`finally`, and production code keeps the mutant. What the next reader
+sees is an uncommitted edit that compiles with a warning
+(`unused variable: index`, for the dropped `index > 0 &&`) or fails one
+test, which is exactly the shape of a half-finished fix, and the
+natural response is to finish it. Before trusting partial edits from an
+interrupted session, `cmp` each touched file against the on-disk copy
+the driver wrote; where the agent that ran the driver can be resumed,
+have it do the classification, since it knows which hunks were its own
+edits and which its mutant. During #1613 a restart left one mutant in
+`doxygen.rs` beside real edits; the resumed agent's `cmp` is what told
+them apart.
 
 **A broken harness reports a plausible failure count, not an error.**
 In that same sweep the restore silently reverted the gate to its `main`

.rustfmt-bail-baseline.txt

@@ -148,6 +148,5 @@ src/metrics/cyclomatic/irules.rs 2
 src/metrics/cyclomatic/perl.rs 2
 src/metrics/cyclomatic/php.rs 4
 src/metrics/cyclomatic/ruby.rs 5
-src/metrics/loc/perl.rs 6
 src/vcs/error.rs 12
 tests/common/mod.rs 2