code-gauge 10.0.0 → 10.1.0

package/README.md

@@ -94,19 +94,19 @@ ignore passes as a check of zero files when it holds no git-visible source file,
 its only sources are ignored build output. Outside a git repository it walks the target directory
 like the ranking command. The thresholds:
 
-| Config key                        | Violation                                         | Warning | Error |
-| --------------------------------- | ------------------------------------------------- | ------- | ----- |
-| `maxFunctionCognitiveComplexity`  | a function's cognitive complexity is above it     | 15      | 30    |
-| `maxFunctionCyclomaticComplexity` | a function's cyclomatic complexity is above it    | off     | off   |
-| `maxFunctionNcss`                 | a function's NCSS is above it                     | 30      | 60    |
-| `maxFunctionNestingDepth`         | a function's nesting depth is above it            | off     | off   |
-| `maxFunctionParameterCount`       | a function's parameter count is above it          | 6       | off   |
-| `maxFunctionHalsteadVolume`       | a function's Halstead volume is above it          | off     | off   |
-| `maxFunctionHalsteadDifficulty`   | a function's Halstead difficulty is above it      | off     | off   |
-| `maxFunctionHalsteadEffort`       | a function's Halstead effort is above it          | off     | off   |
-| `maxFunctionDepDegree`            | a function's DepDegree is above it                | off     | off   |
-| `maxFileNcss`                     | a file's NCSS is above it                         | 400     | 1000  |
-| `minDuplicateLines`               | a duplicated block holds at least this many lines | 15      | off   |
+| Config key                        | Violation                                                           | Warning | Error |
+| --------------------------------- | ------------------------------------------------------------------- | ------- | ----- |
+| `maxFunctionCognitiveComplexity`  | a function's cognitive complexity is above it                       | 15      | 30    |
+| `maxFunctionCyclomaticComplexity` | a function's cyclomatic complexity is above it                      | off     | off   |
+| `maxFunctionNcss`                 | a function's NCSS is above it                                       | 30      | 60    |
+| `maxFunctionNestingDepth`         | a function's nesting depth is above it                              | off     | off   |
+| `maxFunctionParameterCount`       | a function's parameter count is above it                            | 6       | off   |
+| `maxFunctionHalsteadVolume`       | a function's Halstead volume is above it                            | off     | off   |
+| `maxFunctionHalsteadDifficulty`   | a function's Halstead difficulty is above it                        | off     | off   |
+| `maxFunctionHalsteadEffort`       | a function's Halstead effort is above it                            | off     | off   |
+| `maxFunctionDepDegree`            | a function's DepDegree is above it                                  | off     | off   |
+| `maxFileNcss`                     | a file's NCSS is above it                                           | 400     | 1000  |
+| `minDuplicateLines`               | a duplicated block holds at least this many lines matched in a copy | 15      | off   |
 
 Every threshold has a limit per level, and a value is reported at the most severe level whose limit
 it violates. The warning limits mark code worth simplifying when it is touched; the error limits
@@ -132,10 +132,13 @@ for each function.
 
 A duplicated block is one occurrence of a within-file or cross-file clone, found with the
 [duplication detection settings](#duplication-detection-settings); its span runs from its first to
-its last line, and its length is the number of lines of that span that are neither blank nor
-comment-only, so that a copy has the same length whether or not it is commented. Occurrences of at least the warning or the error `minDuplicateLines` that overlap in a
-file are reported as one block covering all of them, at the level the code lines of its whole span
-reach.
+its last line, and its length is the number of lines of that span matched in another copy. Lines
+that are blank or comment-only do not count, so that a copy has the same length whether or not it
+is commented, and neither do the lines between the matched parts of an edited copy, which the copies do not
+share. A block that resembles another as a whole (a near-miss clone, at least
+`minSimilarityPercent` similar) counts all its code lines, its edited ones included. Occurrences of at least the warning or the error
+`minDuplicateLines` that overlap in a file are reported as one block covering all of them, at the
+level its duplicated lines reach, each line counted once.
 
 Each limit is set in the `warning` or `error` part of the
 [`thresholds` config section](#configuration), where `null` disables it and `languages` overrides
@@ -184,7 +187,7 @@ branching construct, which add to the score without being such a part), largest
 and what it adds: the outermost branching constructs and nested functions of the body or, where one
 of them merely wraps most of the function (a loop or an unnamed callback around the whole body),
 the ones inside it. A file-level violation prints the path without a line span, and a duplicated block lists
-up to three of its other copies, those with the most code lines first.
+up to three of its other copies, the largest first.
 Paths are relative to the repository root, or to the target directory outside a git repository.
 
 | Exit code | Meaning                                                                                                                          |
@@ -267,15 +270,14 @@ ranking command.
   - `exceeded`: every violated threshold with its `metric` (the config key without its `max` or
     `min` prefix, e.g. `functionNcss` for `maxFunctionNcss`), the measured `value`, the most severe
     `level` whose limit the value violates, and that `limit`. A value violates when it is above the
-    limit; `duplicateLines`, the code lines of the block, violates from the limit on;
+    limit; `duplicateLines`, the lines of the block matched in another copy, violates from the limit on;
   - `largestBlocks`: only for a function exceeding a cognitive-complexity limit that has a
     branching construct or a nested function adding to it; the parts the
     text report names, each with its line span, the `cognitiveComplexity` it adds to the function,
     and its `name` when it is a named nested function;
-  - `partners`: only for kind `duplication`; every other copy of the block, most code lines
-    first: the
-    block merges every clone overlapping it, so its copies range from whole copies of it to a few
-    lines matching one fragment.
+  - `partners`: only for kind `duplication`; every other copy of the block, largest first. A
+    place repeating only a part of the block (less than `minSimilarityPercent` of its tokens) is
+    not a copy of it, and is listed only when no copy is known.
 - `errors`: the files the check covers that could not be measured (exit code 2); unrelated to the
   `error` level of a violation.
 - `warnings`: files measured without cross-file duplication data, and unmeasured files of the
@@ -420,10 +422,11 @@ The `duplication` section tunes how clones are detected:
   match; dependency declarations such as imports, package clauses, `#include`s, re-exports, and
   `require`s carry no tokens, since every module must spell out its own), with adjacent matches around a small edit merged into gapped (Type-3) clone groups and
   near-miss (Type-3) clones matched by token-LCS similarity (tolerating reordered statements and
-  copies embedded in added code), plus duplicated line count and ratio
+  copies embedded in added code), each copy with its line span, its matched lines, and its matched token count,
+  plus duplicated line count and ratio
 - Cross-file duplication (via `measureCrossFileDuplication`): copy-pasted blocks shared between
   files, matched with the same normalization (exact, gapped, and near-miss clones) and reported as
-  groups with their file locations
+  groups with the file location, the matched lines, and the matched token count of each copy
 - Halstead base counts, vocabulary, length, volume, difficulty (half the distinct operators times
   the total operands per distinct operand; 0 without operands), and effort (difficulty times
   volume), per function and per file

package/dist/checkCommand.cjs

@@ -1,5 +1,5 @@
-"use strict";const e=require("./_virtual/_rolldown/runtime.cjs"),t=require("./thresholds.cjs"),n=require("./cliConfig.cjs"),r=require("./git.cjs"),i=require("./exclusion.cjs"),a=require("./scan.cjs"),o=require("./thresholdCheck.cjs");let s=require("@willbooster/shared-lib"),c=require("node:fs/promises"),l=require("node:path");l=e.__toESM(l,1);let u=require("node:os");u=e.__toESM(u,1);async function d(e,r,i){try{let s=a.resolveTarget(e),c=await n.loadConfig(r.config,await a.configSearchDirectory(s)),l=n.resolveOptions(r,c),u=await m(s,r.base,l),d=o.checkThresholds(u.files,u.crossFileDuplication,t.resolveLimits(i,c.config.thresholds),u.codeLineNumbersByFile);await p(d.violations,u.root),l.json?j(r,u,d):S(r,u,d),u.errors.length>0?process.exitCode=2:f(d.violations)&&(process.exitCode=1)}catch(e){a.writeStderr(`Error: ${(0,s.getErrorMessage)(e)}\n`),process.exitCode=2}}function f(e){return e.some(({level:e})=>e===`error`)}async function p(e,t){for(let n of e.filter(({kind:e})=>e===`file`)){let e=await(0,c.readFile)(l.default.join(t,n.file),`utf8`);/[\n\r]$/u.test(e)&&(n.endLine=Math.max(n.endLine-1,1))}}async function m(e,t,n){let i=await(0,c.realpath)(e),o=(await(0,c.stat)(i)).isFile()?i:void 0;if(o&&!a.getLanguage(o,n,!0))throw Error(`${l.default.basename(o)}: unsupported file type`);let s;try{s=await(0,c.realpath)(await r.resolveRepoRoot(o?l.default.dirname(i):i))}catch(e){if(t!==void 0)throw e;return await h(i,n)}return await g({repoRoot:s,canonicalTarget:i,targetFile:o},t,n)}async function h(e,t){let n=await a.configSearchDirectory(e),r=await a.scanTarget(e,{...t,loadExclusion:e=>i.loadExclusion(n,t.exclude,e)}),o=x(r,t);return{files:r.files.map(({file:e,metrics:t})=>({file:a.formatPath(e,r.displayRoot),metrics:t})),codeLineNumbersByFile:_(r,r.displayRoot),crossFileDuplication:o,errors:r.errors,warnings:r.warnings,root:r.displayRoot}}async function g({repoRoot:e,canonicalTarget:t,targetFile:n},o,s){if(n===void 0&&t!==e&&await r.isIgnored(e,t))throw Error(`${a.formatPath(t,e)}: git ignores this directory, so no file in it is checked; target a file to check it regardless`);let c=await r.listRepositoryFiles(e),u=new Set(n===void 0?[]:[n]),d=n!==void 0&&!c.has(a.formatPath(n,e))?a.formatPath(n,e):void 0;d!==void 0&&c.add(d);let f=i.keepPaths(await i.loadRepositoryExclusion(e,c,s.exclude),u),p=a.formatPath(t,e),m=await a.scanListedFiles(e,c,{...s,loadExclusion:()=>Promise.resolve(f)},u,n===void 0&&t!==e?p:``),h=x(m,s),g=n=>a.isWithinDirectory(l.default.join(e,n),t),b=m.files.map(({file:t,metrics:n})=>({file:a.formatPath(t,e),metrics:n})),S=b.filter(({file:e})=>g(e)),C=n=>t===e||n.startsWith(`${p}/`)||n.startsWith(`${p}:`),w;if(o!==void 0){w=await r.resolveMergeBase(e,o);let t=await r.listChangedFiles(e,w);d!==void 0&&t.push({status:`added`,headPath:d});let n=t.filter(({status:e,headPath:t})=>e!==`deleted`&&g(t));S=await y(b,n,e,w),C=e=>n.some(({headPath:t})=>e.startsWith(`${t}:`))}return{files:S,crossFileDuplication:h,codeLineNumbersByFile:_(m,e),...v(m,C),root:e,mergeBase:w}}function _(e,t){return new Map(e.files.map(({file:e,metrics:n})=>[a.formatPath(e,t),new Set(n.codeLineNumbers)]))}function v(e,t){return{errors:e.errors.filter(e=>t(e)),warnings:[...e.warnings,...e.errors.filter(e=>!t(e))]}}async function y(e,t,n,i){let a=new Map(e.map(e=>[e.file,e]));return(await(0,s.mapConcurrently)(t,u.default.availableParallelism()*2,async e=>{let t=a.get(e.headPath);if(t===void 0)return[];let{metrics:o}=t,s={headStart:1,headCount:o.lines.total},u=e.basePath===void 0?[s]:await r.listLineHunks(n,i,e.basePath,e.headPath);return o.language===`python`&&u.some(({headCount:e})=>e===0)&&b(u,await(0,c.readFile)(l.default.join(n,e.headPath),`utf8`)),[{...t,hunks:u}]})).flat()}function b(e,t){let n=t.split(/\r\n|\n|\r/u);for(let t of e)if(!(t.headCount>0))for(;t.headStart>0&&/^\s*(?:#.*)?$/u.test(n[t.headStart-1]??``);)--t.headStart}function x(e,t){if(e.fatalError)throw Error(e.fatalError);return a.addCrossFileDuplication(e,t),e.crossFileDuplication}function S(e,n,r){let{violations:i}=r,o=`${n.files.length} ${e.base===void 0?``:`changed `}files, ${r.checkedFunctionCount} functions checked${C(e,n)}`;n.errors.length>0?a.writeStdout(`Check could not complete: ${n.errors.length} measurement failures (details on stderr); ${w(i)} in the measured files (${o}).\n`):i.length===0?a.writeStdout(`No threshold violations: ${o}.\n`):a.writeStdout(`Threshold violations: ${w(i)} (${o}).\n`);for(let e of i)a.writeStdout(`${E(e)}\n`);let s=new Set(i.flatMap(({exceeded:e})=>e.map(({metric:e})=>e))),c=t.thresholds.filter(e=>s.has(t.metricNameOf(e)));if(c.length>0){a.writeStdout(`
+"use strict";const e=require("./_virtual/_rolldown/runtime.cjs"),t=require("./thresholds.cjs"),n=require("./cliConfig.cjs"),r=require("./git.cjs"),i=require("./exclusion.cjs"),a=require("./scan.cjs"),o=require("./thresholdCheck.cjs");let s=require("@willbooster/shared-lib"),c=require("node:fs/promises"),l=require("node:path");l=e.__toESM(l,1);let u=require("node:os");u=e.__toESM(u,1);async function d(e,r,i){try{let s=a.resolveTarget(e),c=await n.loadConfig(r.config,await a.configSearchDirectory(s)),l=n.resolveOptions(r,c),u=await m(s,r.base,l),d=o.checkThresholds(u.files,u.duplication,t.resolveLimits(i,c.config.thresholds));await p(d.violations,u.root),l.json?A(r,u,d):x(r,u,d),u.errors.length>0?process.exitCode=2:f(d.violations)&&(process.exitCode=1)}catch(e){a.writeStderr(`Error: ${(0,s.getErrorMessage)(e)}\n`),process.exitCode=2}}function f(e){return e.some(({level:e})=>e===`error`)}async function p(e,t){for(let n of e.filter(({kind:e})=>e===`file`)){let e=await(0,c.readFile)(l.default.join(t,n.file),`utf8`);/[\n\r]$/u.test(e)&&(n.endLine=Math.max(n.endLine-1,1))}}async function m(e,t,n){let i=await(0,c.realpath)(e),o=(await(0,c.stat)(i)).isFile()?i:void 0;if(o&&!a.getLanguage(o,n,!0))throw Error(`${l.default.basename(o)}: unsupported file type`);let s;try{s=await(0,c.realpath)(await r.resolveRepoRoot(o?l.default.dirname(i):i))}catch(e){if(t!==void 0)throw e;return await h(i,n)}return await g({repoRoot:s,canonicalTarget:i,targetFile:o},t,n)}async function h(e,t){let n=await a.configSearchDirectory(e),r=await a.scanTarget(e,{...t,loadExclusion:e=>i.loadExclusion(n,t.exclude,e)});return{files:r.files.map(({file:e,metrics:t})=>({file:a.formatPath(e,r.displayRoot),metrics:t})),duplication:b(r,t),errors:r.errors,warnings:r.warnings,root:r.displayRoot}}async function g({repoRoot:e,canonicalTarget:t,targetFile:n},o,s){if(n===void 0&&t!==e&&await r.isIgnored(e,t))throw Error(`${a.formatPath(t,e)}: git ignores this directory, so no file in it is checked; target a file to check it regardless`);let c=await r.listRepositoryFiles(e),u=new Set(n===void 0?[]:[n]),d=n!==void 0&&!c.has(a.formatPath(n,e))?a.formatPath(n,e):void 0;d!==void 0&&c.add(d);let f=i.keepPaths(await i.loadRepositoryExclusion(e,c,s.exclude),u),p=a.formatPath(t,e),m=await a.scanListedFiles(e,c,{...s,loadExclusion:()=>Promise.resolve(f)},u,n===void 0&&t!==e?p:``),h=b(m,s),g=n=>a.isWithinDirectory(l.default.join(e,n),t),y=m.files.map(({file:t,metrics:n})=>({file:a.formatPath(t,e),metrics:n})),x=y.filter(({file:e})=>g(e)),S=n=>t===e||n.startsWith(`${p}/`)||n.startsWith(`${p}:`),C;if(o!==void 0){C=await r.resolveMergeBase(e,o);let t=await r.listChangedFiles(e,C);d!==void 0&&t.push({status:`added`,headPath:d});let n=t.filter(({status:e,headPath:t})=>e!==`deleted`&&g(t));x=await v(y,n,e,C),S=e=>n.some(({headPath:t})=>e.startsWith(`${t}:`))}return{files:x,duplication:h,..._(m,S),root:e,mergeBase:C}}function _(e,t){return{errors:e.errors.filter(e=>t(e)),warnings:[...e.warnings,...e.errors.filter(e=>!t(e))]}}async function v(e,t,n,i){let a=new Map(e.map(e=>[e.file,e]));return(await(0,s.mapConcurrently)(t,u.default.availableParallelism()*2,async e=>{let t=a.get(e.headPath);if(t===void 0)return[];let{metrics:o}=t,s={headStart:1,headCount:o.lines.total},u=e.basePath===void 0?[s]:await r.listLineHunks(n,i,e.basePath,e.headPath);return o.language===`python`&&u.some(({headCount:e})=>e===0)&&y(u,await(0,c.readFile)(l.default.join(n,e.headPath),`utf8`)),[{...t,hunks:u}]})).flat()}function y(e,t){let n=t.split(/\r\n|\n|\r/u);for(let t of e)if(!(t.headCount>0))for(;t.headStart>0&&/^\s*(?:#.*)?$/u.test(n[t.headStart-1]??``);)--t.headStart}function b(e,t){if(e.fatalError)throw Error(e.fatalError);return a.addCrossFileDuplication(e,t),{crossFile:e.crossFileDuplication,minSimilarityPercent:t.duplication.minSimilarityPercent}}function x(e,n,r){let{violations:i}=r,o=`${n.files.length} ${e.base===void 0?``:`changed `}files, ${r.checkedFunctionCount} functions checked${S(e,n)}`;n.errors.length>0?a.writeStdout(`Check could not complete: ${n.errors.length} measurement failures (details on stderr); ${C(i)} in the measured files (${o}).\n`):i.length===0?a.writeStdout(`No threshold violations: ${o}.\n`):a.writeStdout(`Threshold violations: ${C(i)} (${o}).\n`);for(let e of i)a.writeStdout(`${T(e)}\n`);let s=new Set(i.flatMap(({exceeded:e})=>e.map(({metric:e})=>e))),c=t.thresholds.filter(e=>s.has(t.metricNameOf(e)));if(c.length>0){a.writeStdout(`
 How to fix:
-`);for(let{label:e,hint:t}of c)a.writeStdout(`- ${e}: ${t}\n`)}for(let e of n.warnings)a.writeStderr(`Warning: ${e}\n`);for(let e of n.errors)a.writeStderr(`Error: ${e}\n`)}function C(e,t){return t.mergeBase===void 0?``:`; base ${e.base}, merge-base ${t.mergeBase.slice(0,12)}`}function w(e){let t=T(e);return`${t.errorViolationCount} errors, ${t.warningViolationCount} warnings (${t.functionViolationCount} functions, ${t.fileViolationCount} files, ${t.duplicationViolationCount} duplicated blocks)`}function T(e){let t=t=>e.filter(e=>e.kind===t).length,n=e.filter(({level:e})=>e===`error`).length;return{violationCount:e.length,errorViolationCount:n,warningViolationCount:e.length-n,functionViolationCount:t(`function`),fileViolationCount:t(`file`),duplicationViolationCount:t(`duplication`)}}function E(e){let n=new Map(t.thresholds.map(e=>[t.metricNameOf(e),e.label])),r=e.exceeded.map(({metric:t,value:r,level:i,limit:a})=>{let o=Math.ceil(r*10)/10,s=i===e.level?``:`${i} `,c=e.kind===`duplication`?Math.max(Math.ceil(a)-1,0):a,l=t===`functionCognitiveComplexity`?D(e):``;return`${n.get(t)} ${o} (${s}max ${c}${l})`}).join(`, `);return`${e.level}: ${O(e,r)}`}function D({largestBlocks:e=[]}){let t=e.map(({startLine:e,endLine:t,name:n,cognitiveComplexity:r})=>`L${e}${t===e?``:`-${t}`}${n?` ${k(n)}`:``} ${r}`);return t.length>0?`; largest parts ${t.join(`, `)}`:``}function O(e,t){switch(e.kind){case`file`:return`${e.file}: ${t}`;case`function`:return`${A(e)} ${k(e.name??``)}: ${t}`;case`duplication`:{let n=(e.partners??[]).map(e=>A(e)),r=[...n.slice(0,3),...n.length>3?[`...`]:[]];return`${A(e)}: ${t}${r.length>0?`, also at ${r.join(`, `)}`:``}`}}}function k(e){return e.replaceAll(/\s*[\n\r]\s*/gu,` `)}function A({file:e,startLine:t,endLine:n}){return`${e}:${t}-${n}`}function j(e,t,n){a.writeStdout(JSON.stringify({passed:!f(n.violations)&&t.errors.length===0,base:e.base,mergeBase:t.mergeBase,summary:{...T(n.violations),checkedFileCount:t.files.length,checkedFunctionCount:n.checkedFunctionCount},violations:n.violations,errors:t.errors,warnings:t.warnings},void 0,2)+`
+`);for(let{label:e,hint:t}of c)a.writeStdout(`- ${e}: ${t}\n`)}for(let e of n.warnings)a.writeStderr(`Warning: ${e}\n`);for(let e of n.errors)a.writeStderr(`Error: ${e}\n`)}function S(e,t){return t.mergeBase===void 0?``:`; base ${e.base}, merge-base ${t.mergeBase.slice(0,12)}`}function C(e){let t=w(e);return`${t.errorViolationCount} errors, ${t.warningViolationCount} warnings (${t.functionViolationCount} functions, ${t.fileViolationCount} files, ${t.duplicationViolationCount} duplicated blocks)`}function w(e){let t=t=>e.filter(e=>e.kind===t).length,n=e.filter(({level:e})=>e===`error`).length;return{violationCount:e.length,errorViolationCount:n,warningViolationCount:e.length-n,functionViolationCount:t(`function`),fileViolationCount:t(`file`),duplicationViolationCount:t(`duplication`)}}function T(e){let n=new Map(t.thresholds.map(e=>[t.metricNameOf(e),e.label])),r=e.exceeded.map(({metric:t,value:r,level:i,limit:a})=>{let o=Math.ceil(r*10)/10,s=i===e.level?``:`${i} `,c=e.kind===`duplication`?Math.max(Math.ceil(a)-1,0):a,l=t===`functionCognitiveComplexity`?E(e):``;return`${n.get(t)} ${o} (${s}max ${c}${l})`}).join(`, `);return`${e.level}: ${D(e,r)}`}function E({largestBlocks:e=[]}){let t=e.map(({startLine:e,endLine:t,name:n,cognitiveComplexity:r})=>`L${e}${t===e?``:`-${t}`}${n?` ${O(n)}`:``} ${r}`);return t.length>0?`; largest parts ${t.join(`, `)}`:``}function D(e,t){switch(e.kind){case`file`:return`${e.file}: ${t}`;case`function`:return`${k(e)} ${O(e.name??``)}: ${t}`;case`duplication`:{let n=(e.partners??[]).map(e=>k(e)),r=[...n.slice(0,3),...n.length>3?[`...`]:[]];return`${k(e)}: ${t}${r.length>0?`, also at ${r.join(`, `)}`:``}`}}}function O(e){return e.replaceAll(/\s*[\n\r]\s*/gu,` `)}function k({file:e,startLine:t,endLine:n}){return`${e}:${t}-${n}`}function A(e,t,n){a.writeStdout(JSON.stringify({passed:!f(n.violations)&&t.errors.length===0,base:e.base,mergeBase:t.mergeBase,summary:{...w(n.violations),checkedFileCount:t.files.length,checkedFunctionCount:n.checkedFunctionCount},violations:n.violations,errors:t.errors,warnings:t.warnings},void 0,2)+`
 `)}exports.runCheckCommand=d,exports.scanCheckScope=m;
 //# sourceMappingURL=checkCommand.cjs.map

package/dist/checkCommand.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"checkCommand.cjs","names":["resolveTarget","loadConfig","configSearchDirectory","resolveOptions","checkThresholds","resolveLimits","getErrorMessage","readFile","path","realpath","stat","getLanguage","resolveRepoRoot","scanTarget","loadExclusion","formatPath","isIgnored","listRepositoryFiles","keepPaths","loadRepositoryExclusion","scanListedFiles","isWithinDirectory","resolveMergeBase","listChangedFiles","mapConcurrently","os","listLineHunks","thresholds","metricNameOf"],"sources":["../src/checkCommand.ts"],"sourcesContent":["import { readFile, realpath, stat } from 'node:fs/promises';\nimport os from 'node:os';\nimport path from 'node:path';\nimport { getErrorMessage, mapConcurrently } from '@willbooster/shared-lib';\nimport { loadConfig, resolveOptions, type CliOptions, type ResolvedOptions } from './cliConfig.js';\nimport type { CrossFileDuplicationMetrics } from './crossFileDuplication.js';\nimport { keepPaths, loadExclusion, loadRepositoryExclusion } from './exclusion.js';\nimport {\n  isIgnored,\n  listChangedFiles,\n  listLineHunks,\n  listRepositoryFiles,\n  resolveMergeBase,\n  resolveRepoRoot,\n  type ChangedFile,\n  type LineHunk,\n} from './git.js';\nimport {\n  addCrossFileDuplication,\n  configSearchDirectory,\n  formatPath,\n  getLanguage,\n  isWithinDirectory,\n  resolveTarget,\n  scanListedFiles,\n  scanTarget,\n  writeStderr,\n  writeStdout,\n  type ScanResult,\n} from './scan.js';\nimport {\n  checkThresholds,\n  type BlockLocation,\n  type CheckedFile,\n  type CodeLineNumbersByFile,\n  type CheckResult,\n  type Violation,\n} from './thresholdCheck.js';\nimport { metricNameOf, resolveLimits, thresholds, type LimitsByLevel } from './thresholds.js';\n\n/** Raw options of the `check` subcommand; every field is undefined unless the flag was passed. */\nexport interface CheckCliOptions extends CliOptions {\n  base?: string;\n}\n\n/** The measured files a check covers and the duplication they take part in. */\nexport interface CheckScope {\n  files: CheckedFile[];\n  crossFileDuplication?: CrossFileDuplicationMetrics;\n  /** Of every measured file, so that a copy outside the files the check covers is sized like one inside. */\n  codeLineNumbersByFile: CodeLineNumbersByFile;\n  /** Measurement failures of files the check covers. */\n  errors: string[];\n  warnings: string[];\n  /** The directory the files' paths are relative to. */\n  root: string;\n  mergeBase?: string;\n}\n\n/** Caps how many partner locations a duplicated block lists so the report stays scannable. */\nconst maxListedPartners = 3;\n\n/**\n * Reports every exceeded threshold of the target or, with `base`, of what the working tree changed\n * under it since the merge-base with that ref. Exit codes: 0 no error-level violations, 1 error-level\n * violations, 2 files the check covers could not be measured.\n */\nexport async function runCheckCommand(\n  target: string,\n  cliOptions: CheckCliOptions,\n  cliLimits: LimitsByLevel\n): Promise<void> {\n  try {\n    const resolvedTarget = resolveTarget(target);\n    const config = await loadConfig(cliOptions.config, await configSearchDirectory(resolvedTarget));\n    const options = resolveOptions(cliOptions, config);\n    const scope = await scanCheckScope(resolvedTarget, cliOptions.base, options);\n    const result = checkThresholds(\n      scope.files,\n      scope.crossFileDuplication,\n      resolveLimits(cliLimits, config.config.thresholds),\n      scope.codeLineNumbersByFile\n    );\n    await endFileViolationsAtLastLine(result.violations, scope.root);\n\n    if (options.json) {\n      printJsonReport(cliOptions, scope, result);\n    } else {\n      printTextReport(cliOptions, scope, result);\n    }\n    if (scope.errors.length > 0) {\n      process.exitCode = 2;\n    } else if (hasErrors(result.violations)) {\n      process.exitCode = 1;\n    }\n  } catch (error) {\n    writeStderr(`Error: ${getErrorMessage(error)}\\n`);\n    process.exitCode = 2;\n  }\n}\n\nfunction hasErrors(violations: readonly Violation[]): boolean {\n  return violations.some(({ level }) => level === 'error');\n}\n\n/**\n * A file's measured line count includes the empty line after a final line terminator, which is\n * not a line of the file, so file-level violations end at the last line that exists.\n */\nasync function endFileViolationsAtLastLine(violations: readonly Violation[], root: string): Promise<void> {\n  for (const violation of violations.filter(({ kind }) => kind === 'file')) {\n    const content = await readFile(path.join(root, violation.file), 'utf8');\n    if (/[\\n\\r]$/u.test(content)) {\n      violation.endLine = Math.max(violation.endLine - 1, 1);\n    }\n  }\n}\n\n/**\n * Inside a git repository, the scope is the repository's git-visible files under the target, and\n * duplication is detected against all of them; `base` only narrows the scope to what changed.\n * Outside a repository, where there is no change to narrow to, the target directory is walked.\n */\nexport async function scanCheckScope(\n  resolvedTarget: string,\n  base: string | undefined,\n  options: ResolvedOptions\n): Promise<CheckScope> {\n  const canonicalTarget = await realpath(resolvedTarget);\n  const targetStat = await stat(canonicalTarget);\n  const targetFile = targetStat.isFile() ? canonicalTarget : undefined;\n  if (targetFile && !getLanguage(targetFile, options, true)) {\n    throw new Error(`${path.basename(targetFile)}: unsupported file type`);\n  }\n  let repoRoot;\n  try {\n    repoRoot = await realpath(await resolveRepoRoot(targetFile ? path.dirname(canonicalTarget) : canonicalTarget));\n  } catch (error) {\n    if (base !== undefined) {\n      throw error;\n    }\n    return await scanDirectoryWalk(canonicalTarget, options);\n  }\n  return await scanRepository({ repoRoot, canonicalTarget, targetFile }, base, options);\n}\n\nasync function scanDirectoryWalk(canonicalTarget: string, options: ResolvedOptions): Promise<CheckScope> {\n  const searchDirectory = await configSearchDirectory(canonicalTarget);\n  const scan = await scanTarget(canonicalTarget, {\n    ...options,\n    loadExclusion: (absolutePaths) => loadExclusion(searchDirectory, options.exclude, absolutePaths),\n  });\n  const crossFileDuplication = measureDuplication(scan, options);\n  return {\n    files: scan.files.map(({ file, metrics }) => ({ file: formatPath(file, scan.displayRoot), metrics })),\n    codeLineNumbersByFile: indexCodeLineNumbers(scan, scan.displayRoot),\n    crossFileDuplication,\n    errors: scan.errors,\n    warnings: scan.warnings,\n    root: scan.displayRoot,\n  };\n}\n\ninterface RepositoryTarget {\n  repoRoot: string;\n  canonicalTarget: string;\n  /** Set when the target is a file, which is measured even when git ignores it or it is excluded. */\n  targetFile?: string;\n}\n\n/**\n * Measures every git-visible file (tracked or untracked non-ignored) of the repository, so that\n * code copied from outside the target is detected as duplication, and keeps the files under the\n * target: all of them or, with `base`, the changed ones with their line hunks.\n */\nasync function scanRepository(\n  { repoRoot, canonicalTarget, targetFile }: RepositoryTarget,\n  base: string | undefined,\n  options: ResolvedOptions\n): Promise<CheckScope> {\n  // Only git-visible files are measured, so a directory git ignores would pass as a check of no\n  // files. A directory git lists but whose sources it ignores is not told apart from a repository\n  // or package whose only sources are ignored build output, which must pass.\n  if (targetFile === undefined && canonicalTarget !== repoRoot && (await isIgnored(repoRoot, canonicalTarget))) {\n    throw new Error(\n      `${formatPath(canonicalTarget, repoRoot)}: git ignores this directory, so no file in it is checked; target a file to check it regardless`\n    );\n  }\n  const repositoryFiles = await listRepositoryFiles(repoRoot);\n  const explicitFiles = new Set(targetFile === undefined ? [] : [targetFile]);\n  const unlistedTarget =\n    targetFile !== undefined && !repositoryFiles.has(formatPath(targetFile, repoRoot))\n      ? formatPath(targetFile, repoRoot)\n      : undefined;\n  if (unlistedTarget !== undefined) {\n    repositoryFiles.add(unlistedTarget);\n  }\n  const exclusion = keepPaths(await loadRepositoryExclusion(repoRoot, repositoryFiles, options.exclude), explicitFiles);\n  // Scan errors start with the file or directory they concern.\n  const targetPath = formatPath(canonicalTarget, repoRoot);\n  const scan = await scanListedFiles(\n    repoRoot,\n    repositoryFiles,\n    { ...options, loadExclusion: () => Promise.resolve(exclusion) },\n    explicitFiles,\n    targetFile === undefined && canonicalTarget !== repoRoot ? targetPath : ''\n  );\n  const crossFileDuplication = measureDuplication(scan, options);\n  const isInTarget = (relativePath: string): boolean =>\n    isWithinDirectory(path.join(repoRoot, relativePath), canonicalTarget);\n  const measuredFiles = scan.files.map(({ file, metrics }) => ({ file: formatPath(file, repoRoot), metrics }));\n  let files = measuredFiles.filter(({ file }) => isInTarget(file));\n  let isCovered = (error: string): boolean =>\n    canonicalTarget === repoRoot || error.startsWith(`${targetPath}/`) || error.startsWith(`${targetPath}:`);\n  let mergeBase;\n  if (base !== undefined) {\n    mergeBase = await resolveMergeBase(repoRoot, base);\n    const changedFiles = await listChangedFiles(repoRoot, mergeBase);\n    // A targeted file git does not list has no diff; all of it counts as added.\n    if (unlistedTarget !== undefined) {\n      changedFiles.push({ status: 'added', headPath: unlistedTarget });\n    }\n    const changesInTarget = changedFiles.filter(({ status, headPath }) => status !== 'deleted' && isInTarget(headPath));\n    files = await attachHunks(measuredFiles, changesInTarget, repoRoot, mergeBase);\n    isCovered = (error) => changesInTarget.some(({ headPath }) => error.startsWith(`${headPath}:`));\n  }\n  return {\n    files,\n    crossFileDuplication,\n    codeLineNumbersByFile: indexCodeLineNumbers(scan, repoRoot),\n    ...partitionErrors(scan, isCovered),\n    root: repoRoot,\n    mergeBase,\n  };\n}\n\nfunction indexCodeLineNumbers(scan: ScanResult, root: string): CodeLineNumbersByFile {\n  return new Map(scan.files.map(({ file, metrics }) => [formatPath(file, root), new Set(metrics.codeLineNumbers)]));\n}\n\n/** Only a failure on a file the check covers leaves it incomplete; the others are warnings. */\nfunction partitionErrors(\n  scan: ScanResult,\n  isCovered: (error: string) => boolean\n): Pick<CheckScope, 'errors' | 'warnings'> {\n  return {\n    errors: scan.errors.filter((error) => isCovered(error)),\n    warnings: [...scan.warnings, ...scan.errors.filter((error) => !isCovered(error))],\n  };\n}\n\n/** The measured changed files with their line hunks against the merge-base. */\nasync function attachHunks(\n  measuredFiles: readonly CheckedFile[],\n  changedFiles: readonly ChangedFile[],\n  repoRoot: string,\n  mergeBase: string\n): Promise<CheckedFile[]> {\n  const measuredByPath = new Map(measuredFiles.map((measured) => [measured.file, measured]));\n  const files = await mapConcurrently(changedFiles, os.availableParallelism() * 2, async (changed) => {\n    const measured = measuredByPath.get(changed.headPath);\n    // A changed file the scan left out (unsupported, excluded, or failed) has nothing to check.\n    if (measured === undefined) {\n      return [];\n    }\n    const { metrics } = measured;\n    const wholeFile: LineHunk = { headStart: 1, headCount: metrics.lines.total };\n    const hunks =\n      changed.basePath === undefined\n        ? [wholeFile]\n        : await listLineHunks(repoRoot, mergeBase, changed.basePath, changed.headPath);\n    if (metrics.language === 'python' && hunks.some(({ headCount }) => headCount === 0)) {\n      anchorDeletionsToCode(hunks, await readFile(path.join(repoRoot, changed.headPath), 'utf8'));\n    }\n    return [{ ...measured, hunks }];\n  });\n  return files.flat();\n}\n\n/**\n * Moves each pure deletion up to the last code line before it. A Python function ends at its last\n * statement, so the blank and comment-only lines left above a deleted tail lie outside the span the\n * deletion shortened; anchored to that statement, the deletion counts as touching the function.\n */\nfunction anchorDeletionsToCode(hunks: LineHunk[], content: string): void {\n  const lines = content.split(/\\r\\n|\\n|\\r/u);\n  for (const hunk of hunks) {\n    if (hunk.headCount > 0) continue;\n    while (hunk.headStart > 0 && /^\\s*(?:#.*)?$/u.test(lines[hunk.headStart - 1] ?? '')) {\n      hunk.headStart -= 1;\n    }\n  }\n}\n\n/** A run-wide failure (a missing target or native addon) leaves nothing to check. */\nfunction measureDuplication(scan: ScanResult, options: ResolvedOptions): CrossFileDuplicationMetrics | undefined {\n  if (scan.fatalError) {\n    throw new Error(scan.fatalError);\n  }\n  addCrossFileDuplication(scan, options);\n  return scan.crossFileDuplication;\n}\n\nfunction printTextReport(cliOptions: CheckCliOptions, scope: CheckScope, result: CheckResult): void {\n  const { violations } = result;\n  const checked = `${scope.files.length} ${cliOptions.base === undefined ? '' : 'changed '}files, ${result.checkedFunctionCount} functions checked${describeBase(cliOptions, scope)}`;\n  if (scope.errors.length > 0) {\n    // Unmeasured files were not checked, so \"no violations\" would be vacuous; never claim a pass.\n    writeStdout(\n      `Check could not complete: ${scope.errors.length} measurement failures (details on stderr); ${describeCounts(violations)} in the measured files (${checked}).\\n`\n    );\n  } else if (violations.length === 0) {\n    writeStdout(`No threshold violations: ${checked}.\\n`);\n  } else {\n    writeStdout(`Threshold violations: ${describeCounts(violations)} (${checked}).\\n`);\n  }\n  for (const violation of violations) {\n    writeStdout(`${formatViolation(violation)}\\n`);\n  }\n\n  const violatedMetrics = new Set(violations.flatMap(({ exceeded }) => exceeded.map(({ metric }) => metric)));\n  const violatedThresholds = thresholds.filter((threshold) => violatedMetrics.has(metricNameOf(threshold)));\n  if (violatedThresholds.length > 0) {\n    writeStdout('\\nHow to fix:\\n');\n    for (const { label, hint } of violatedThresholds) {\n      writeStdout(`- ${label}: ${hint}\\n`);\n    }\n  }\n\n  for (const warning of scope.warnings) {\n    writeStderr(`Warning: ${warning}\\n`);\n  }\n  for (const error of scope.errors) {\n    writeStderr(`Error: ${error}\\n`);\n  }\n}\n\nfunction describeBase(cliOptions: CheckCliOptions, scope: CheckScope): string {\n  return scope.mergeBase === undefined ? '' : `; base ${cliOptions.base}, merge-base ${scope.mergeBase.slice(0, 12)}`;\n}\n\nfunction describeCounts(violations: readonly Violation[]): string {\n  const counts = countViolations(violations);\n  return `${counts.errorViolationCount} errors, ${counts.warningViolationCount} warnings (${counts.functionViolationCount} functions, ${counts.fileViolationCount} files, ${counts.duplicationViolationCount} duplicated blocks)`;\n}\n\nfunction countViolations(violations: readonly Violation[]): {\n  violationCount: number;\n  errorViolationCount: number;\n  warningViolationCount: number;\n  functionViolationCount: number;\n  fileViolationCount: number;\n  duplicationViolationCount: number;\n} {\n  const count = (kind: Violation['kind']): number => violations.filter((violation) => violation.kind === kind).length;\n  const errorViolationCount = violations.filter(({ level }) => level === 'error').length;\n  return {\n    violationCount: violations.length,\n    errorViolationCount,\n    warningViolationCount: violations.length - errorViolationCount,\n    functionViolationCount: count('function'),\n    fileViolationCount: count('file'),\n    duplicationViolationCount: count('duplication'),\n  };\n}\n\n/** One violation as a single line: its level, the location, then every limit it exceeds. */\nfunction formatViolation(violation: Violation): string {\n  const labelByMetric = new Map(thresholds.map((threshold) => [metricNameOf(threshold), threshold.label]));\n  const exceeded = violation.exceeded\n    .map(({ metric, value, level, limit }) => {\n      // Halstead values are fractional; rounding up keeps a violating value above the printed maximum.\n      const roundedValue = Math.ceil(value * 10) / 10;\n      const levelPrefix = level === violation.level ? '' : `${level} `;\n      // A duplicated block violates from its limit on, so the largest allowed code-line count is\n      // the last whole number below it; a limit of 0 allows none.\n      const maxAllowed = violation.kind === 'duplication' ? Math.max(Math.ceil(limit) - 1, 0) : limit;\n      const blocks = metric === 'functionCognitiveComplexity' ? describeLargestBlocks(violation) : '';\n      return `${labelByMetric.get(metric)} ${roundedValue} (${levelPrefix}max ${maxAllowed}${blocks})`;\n    })\n    .join(', ');\n  return `${violation.level}: ${describeViolation(violation, exceeded)}`;\n}\n\nfunction describeLargestBlocks({ largestBlocks = [] }: Violation): string {\n  const blocks = largestBlocks.map(\n    ({ startLine, endLine, name, cognitiveComplexity }) =>\n      `L${startLine}${endLine === startLine ? '' : `-${endLine}`}${name ? ` ${toOneLine(name)}` : ''} ${cognitiveComplexity}`\n  );\n  return blocks.length > 0 ? `; largest parts ${blocks.join(', ')}` : '';\n}\n\nfunction describeViolation(violation: Violation, exceeded: string): string {\n  switch (violation.kind) {\n    case 'file': {\n      return `${violation.file}: ${exceeded}`;\n    }\n    case 'function': {\n      return `${formatLocation(violation)} ${toOneLine(violation.name ?? '')}: ${exceeded}`;\n    }\n    case 'duplication': {\n      const partners = (violation.partners ?? []).map((partner) => formatLocation(partner));\n      const listed = [...partners.slice(0, maxListedPartners), ...(partners.length > maxListedPartners ? ['...'] : [])];\n      return `${formatLocation(violation)}: ${exceeded}${listed.length > 0 ? `, also at ${listed.join(', ')}` : ''}`;\n    }\n  }\n}\n\n/** A computed name can span lines in the source; the report keeps one line per function. */\nfunction toOneLine(name: string): string {\n  return name.replaceAll(/\\s*[\\n\\r]\\s*/gu, ' ');\n}\n\nfunction formatLocation({ file, startLine, endLine }: BlockLocation): string {\n  return `${file}:${startLine}-${endLine}`;\n}\n\nfunction printJsonReport(cliOptions: CheckCliOptions, scope: CheckScope, result: CheckResult): void {\n  writeStdout(\n    JSON.stringify(\n      {\n        passed: !hasErrors(result.violations) && scope.errors.length === 0,\n        base: cliOptions.base,\n        mergeBase: scope.mergeBase,\n        summary: {\n          ...countViolations(result.violations),\n          checkedFileCount: scope.files.length,\n          checkedFunctionCount: result.checkedFunctionCount,\n        },\n        violations: result.violations,\n        errors: scope.errors,\n        warnings: scope.warnings,\n      },\n      undefined,\n      2\n    ) + '\\n'\n  );\n}\n"],"mappings":"mYAmEA,eAAsB,EACpB,EACA,EACA,EACe,CACf,GAAI,CACF,IAAM,EAAiBA,EAAAA,cAAc,CAAM,EACrC,EAAS,MAAMC,EAAAA,WAAW,EAAW,OAAQ,MAAMC,EAAAA,sBAAsB,CAAc,CAAC,EACxF,EAAUC,EAAAA,eAAe,EAAY,CAAM,EAC3C,EAAQ,MAAM,EAAe,EAAgB,EAAW,KAAM,CAAO,EACrE,EAASC,EAAAA,gBACb,EAAM,MACN,EAAM,qBACNC,EAAAA,cAAc,EAAW,EAAO,OAAO,UAAU,EACjD,EAAM,qBACR,EACA,MAAM,EAA4B,EAAO,WAAY,EAAM,IAAI,EAE3D,EAAQ,KACV,EAAgB,EAAY,EAAO,CAAM,EAEzC,EAAgB,EAAY,EAAO,CAAM,EAEvC,EAAM,OAAO,OAAS,EACxB,QAAQ,SAAW,EACV,EAAU,EAAO,UAAU,IACpC,QAAQ,SAAW,EAEvB,OAAS,EAAO,CACd,EAAA,YAAY,WAAA,EAAUC,EAAAA,gBAAAA,CAAgB,CAAK,EAAE,GAAG,EAChD,QAAQ,SAAW,CACrB,CACF,CAEA,SAAS,EAAU,EAA2C,CAC5D,OAAO,EAAW,MAAM,CAAE,WAAY,IAAU,OAAO,CACzD,CAMA,eAAe,EAA4B,EAAkC,EAA6B,CACxG,IAAK,IAAM,KAAa,EAAW,QAAQ,CAAE,UAAW,IAAS,MAAM,EAAG,CACxE,IAAM,EAAU,MAAA,EAAMC,EAAAA,SAAAA,CAASC,EAAAA,QAAK,KAAK,EAAM,EAAU,IAAI,EAAG,MAAM,EAClE,WAAW,KAAK,CAAO,IACzB,EAAU,QAAU,KAAK,IAAI,EAAU,QAAU,EAAG,CAAC,EAEzD,CACF,CAOA,eAAsB,EACpB,EACA,EACA,EACqB,CACrB,IAAM,EAAkB,MAAA,EAAMC,EAAAA,SAAAA,CAAS,CAAc,EAE/C,GAAa,MAAA,EADMC,EAAAA,KAAAA,CAAK,CAAe,EAAA,CACf,OAAO,EAAI,EAAkB,IAAA,GAC3D,GAAI,GAAc,CAACC,EAAAA,YAAY,EAAY,EAAS,EAAI,EACtD,MAAU,MAAM,GAAGH,EAAAA,QAAK,SAAS,CAAU,EAAE,wBAAwB,EAEvE,IAAI,EACJ,GAAI,CACF,EAAW,MAAA,EAAMC,EAAAA,SAAAA,CAAS,MAAMG,EAAAA,gBAAgB,EAAaJ,EAAAA,QAAK,QAAQ,CAAe,EAAI,CAAe,CAAC,CAC/G,OAAS,EAAO,CACd,GAAI,IAAS,IAAA,GACX,MAAM,EAER,OAAO,MAAM,EAAkB,EAAiB,CAAO,CACzD,CACA,OAAO,MAAM,EAAe,CAAE,WAAU,kBAAiB,YAAW,EAAG,EAAM,CAAO,CACtF,CAEA,eAAe,EAAkB,EAAyB,EAA+C,CACvG,IAAM,EAAkB,MAAMN,EAAAA,sBAAsB,CAAe,EAC7D,EAAO,MAAMW,EAAAA,WAAW,EAAiB,CAC7C,GAAG,EACH,cAAgB,GAAkBC,EAAAA,cAAc,EAAiB,EAAQ,QAAS,CAAa,CACjG,CAAC,EACK,EAAuB,EAAmB,EAAM,CAAO,EAC7D,MAAO,CACL,MAAO,EAAK,MAAM,KAAK,CAAE,OAAM,cAAe,CAAE,KAAMC,EAAAA,WAAW,EAAM,EAAK,WAAW,EAAG,SAAQ,EAAE,EACpG,sBAAuB,EAAqB,EAAM,EAAK,WAAW,EAClE,uBACA,OAAQ,EAAK,OACb,SAAU,EAAK,SACf,KAAM,EAAK,WACb,CACF,CAcA,eAAe,EACb,CAAE,WAAU,kBAAiB,cAC7B,EACA,EACqB,CAIrB,GAAI,IAAe,IAAA,IAAa,IAAoB,GAAa,MAAMC,EAAAA,UAAU,EAAU,CAAe,EACxG,MAAU,MACR,GAAGD,EAAAA,WAAW,EAAiB,CAAQ,EAAE,gGAC3C,EAEF,IAAM,EAAkB,MAAME,EAAAA,oBAAoB,CAAQ,EACpD,EAAgB,IAAI,IAAI,IAAe,IAAA,GAAY,CAAC,EAAI,CAAC,CAAU,CAAC,EACpE,EACJ,IAAe,IAAA,IAAa,CAAC,EAAgB,IAAIF,EAAAA,WAAW,EAAY,CAAQ,CAAC,EAC7EA,EAAAA,WAAW,EAAY,CAAQ,EAC/B,IAAA,GACF,IAAmB,IAAA,IACrB,EAAgB,IAAI,CAAc,EAEpC,IAAM,EAAYG,EAAAA,UAAU,MAAMC,EAAAA,wBAAwB,EAAU,EAAiB,EAAQ,OAAO,EAAG,CAAa,EAE9G,EAAaJ,EAAAA,WAAW,EAAiB,CAAQ,EACjD,EAAO,MAAMK,EAAAA,gBACjB,EACA,EACA,CAAE,GAAG,EAAS,kBAAqB,QAAQ,QAAQ,CAAS,CAAE,EAC9D,EACA,IAAe,IAAA,IAAa,IAAoB,EAAW,EAAa,EAC1E,EACM,EAAuB,EAAmB,EAAM,CAAO,EACvD,EAAc,GAClBC,EAAAA,kBAAkBb,EAAAA,QAAK,KAAK,EAAU,CAAY,EAAG,CAAe,EAChE,EAAgB,EAAK,MAAM,KAAK,CAAE,OAAM,cAAe,CAAE,KAAMO,EAAAA,WAAW,EAAM,CAAQ,EAAG,SAAQ,EAAE,EACvG,EAAQ,EAAc,QAAQ,CAAE,UAAW,EAAW,CAAI,CAAC,EAC3D,EAAa,GACf,IAAoB,GAAY,EAAM,WAAW,GAAG,EAAW,EAAE,GAAK,EAAM,WAAW,GAAG,EAAW,EAAE,EACrG,EACJ,GAAI,IAAS,IAAA,GAAW,CACtB,EAAY,MAAMO,EAAAA,iBAAiB,EAAU,CAAI,EACjD,IAAM,EAAe,MAAMC,EAAAA,iBAAiB,EAAU,CAAS,EAE3D,IAAmB,IAAA,IACrB,EAAa,KAAK,CAAE,OAAQ,QAAS,SAAU,CAAe,CAAC,EAEjE,IAAM,EAAkB,EAAa,QAAQ,CAAE,SAAQ,cAAe,IAAW,WAAa,EAAW,CAAQ,CAAC,EAClH,EAAQ,MAAM,EAAY,EAAe,EAAiB,EAAU,CAAS,EAC7E,EAAa,GAAU,EAAgB,MAAM,CAAE,cAAe,EAAM,WAAW,GAAG,EAAS,EAAE,CAAC,CAChG,CACA,MAAO,CACL,QACA,uBACA,sBAAuB,EAAqB,EAAM,CAAQ,EAC1D,GAAG,EAAgB,EAAM,CAAS,EAClC,KAAM,EACN,WACF,CACF,CAEA,SAAS,EAAqB,EAAkB,EAAqC,CACnF,OAAO,IAAI,IAAI,EAAK,MAAM,KAAK,CAAE,OAAM,aAAc,CAACR,EAAAA,WAAW,EAAM,CAAI,EAAG,IAAI,IAAI,EAAQ,eAAe,CAAC,CAAC,CAAC,CAClH,CAGA,SAAS,EACP,EACA,EACyC,CACzC,MAAO,CACL,OAAQ,EAAK,OAAO,OAAQ,GAAU,EAAU,CAAK,CAAC,EACtD,SAAU,CAAC,GAAG,EAAK,SAAU,GAAG,EAAK,OAAO,OAAQ,GAAU,CAAC,EAAU,CAAK,CAAC,CAAC,CAClF,CACF,CAGA,eAAe,EACb,EACA,EACA,EACA,EACwB,CACxB,IAAM,EAAiB,IAAI,IAAI,EAAc,IAAK,GAAa,CAAC,EAAS,KAAM,CAAQ,CAAC,CAAC,EAkBzF,OAAO,MAAA,EAjBaS,EAAAA,gBAAAA,CAAgB,EAAcC,EAAAA,QAAG,qBAAqB,EAAI,EAAG,KAAO,IAAY,CAClG,IAAM,EAAW,EAAe,IAAI,EAAQ,QAAQ,EAEpD,GAAI,IAAa,IAAA,GACf,MAAO,CAAC,EAEV,GAAM,CAAE,WAAY,EACd,EAAsB,CAAE,UAAW,EAAG,UAAW,EAAQ,MAAM,KAAM,EACrE,EACJ,EAAQ,WAAa,IAAA,GACjB,CAAC,CAAS,EACV,MAAMC,EAAAA,cAAc,EAAU,EAAW,EAAQ,SAAU,EAAQ,QAAQ,EAIjF,OAHI,EAAQ,WAAa,UAAY,EAAM,MAAM,CAAE,eAAgB,IAAc,CAAC,GAChF,EAAsB,EAAO,MAAA,EAAMnB,EAAAA,SAAAA,CAASC,EAAAA,QAAK,KAAK,EAAU,EAAQ,QAAQ,EAAG,MAAM,CAAC,EAErF,CAAC,CAAE,GAAG,EAAU,OAAM,CAAC,CAChC,CAAC,EAAA,CACY,KAAK,CACpB,CAOA,SAAS,EAAsB,EAAmB,EAAuB,CACvE,IAAM,EAAQ,EAAQ,MAAM,aAAa,EACzC,IAAK,IAAM,KAAQ,EACb,OAAK,UAAY,GACrB,KAAO,EAAK,UAAY,GAAK,iBAAiB,KAAK,EAAM,EAAK,UAAY,IAAM,EAAE,GAChF,IAAK,SAGX,CAGA,SAAS,EAAmB,EAAkB,EAAmE,CAC/G,GAAI,EAAK,WACP,MAAU,MAAM,EAAK,UAAU,EAGjC,OADA,EAAA,wBAAwB,EAAM,CAAO,EAC9B,EAAK,oBACd,CAEA,SAAS,EAAgB,EAA6B,EAAmB,EAA2B,CAClG,GAAM,CAAE,cAAe,EACjB,EAAU,GAAG,EAAM,MAAM,OAAO,GAAG,EAAW,OAAS,IAAA,GAAY,GAAK,WAAW,SAAS,EAAO,qBAAqB,oBAAoB,EAAa,EAAY,CAAK,IAC5K,EAAM,OAAO,OAAS,EAExB,EAAA,YACE,6BAA6B,EAAM,OAAO,OAAO,6CAA6C,EAAe,CAAU,EAAE,0BAA0B,EAAQ,KAC7J,EACS,EAAW,SAAW,EAC/B,EAAA,YAAY,4BAA4B,EAAQ,IAAI,EAEpD,EAAA,YAAY,yBAAyB,EAAe,CAAU,EAAE,IAAI,EAAQ,KAAK,EAEnF,IAAK,IAAM,KAAa,EACtB,EAAA,YAAY,GAAG,EAAgB,CAAS,EAAE,GAAG,EAG/C,IAAM,EAAkB,IAAI,IAAI,EAAW,SAAS,CAAE,cAAe,EAAS,KAAK,CAAE,YAAa,CAAM,CAAC,CAAC,EACpG,EAAqBmB,EAAAA,WAAW,OAAQ,GAAc,EAAgB,IAAIC,EAAAA,aAAa,CAAS,CAAC,CAAC,EACxG,GAAI,EAAmB,OAAS,EAAG,CACjC,EAAA,YAAY;;CAAiB,EAC7B,IAAK,GAAM,CAAE,QAAO,UAAU,EAC5B,EAAA,YAAY,KAAK,EAAM,IAAI,EAAK,GAAG,CAEvC,CAEA,IAAK,IAAM,KAAW,EAAM,SAC1B,EAAA,YAAY,YAAY,EAAQ,GAAG,EAErC,IAAK,IAAM,KAAS,EAAM,OACxB,EAAA,YAAY,UAAU,EAAM,GAAG,CAEnC,CAEA,SAAS,EAAa,EAA6B,EAA2B,CAC5E,OAAO,EAAM,YAAc,IAAA,GAAY,GAAK,UAAU,EAAW,KAAK,eAAe,EAAM,UAAU,MAAM,EAAG,EAAE,GAClH,CAEA,SAAS,EAAe,EAA0C,CAChE,IAAM,EAAS,EAAgB,CAAU,EACzC,MAAO,GAAG,EAAO,oBAAoB,WAAW,EAAO,sBAAsB,aAAa,EAAO,uBAAuB,cAAc,EAAO,mBAAmB,UAAU,EAAO,0BAA0B,oBAC7M,CAEA,SAAS,EAAgB,EAOvB,CACA,IAAM,EAAS,GAAoC,EAAW,OAAQ,GAAc,EAAU,OAAS,CAAI,CAAC,CAAC,OACvG,EAAsB,EAAW,QAAQ,CAAE,WAAY,IAAU,OAAO,CAAC,CAAC,OAChF,MAAO,CACL,eAAgB,EAAW,OAC3B,sBACA,sBAAuB,EAAW,OAAS,EAC3C,uBAAwB,EAAM,UAAU,EACxC,mBAAoB,EAAM,MAAM,EAChC,0BAA2B,EAAM,aAAa,CAChD,CACF,CAGA,SAAS,EAAgB,EAA8B,CACrD,IAAM,EAAgB,IAAI,IAAID,EAAAA,WAAW,IAAK,GAAc,CAACC,EAAAA,aAAa,CAAS,EAAG,EAAU,KAAK,CAAC,CAAC,EACjG,EAAW,EAAU,SACxB,KAAK,CAAE,SAAQ,QAAO,QAAO,WAAY,CAExC,IAAM,EAAe,KAAK,KAAK,EAAQ,EAAE,EAAI,GACvC,EAAc,IAAU,EAAU,MAAQ,GAAK,GAAG,EAAM,GAGxD,EAAa,EAAU,OAAS,cAAgB,KAAK,IAAI,KAAK,KAAK,CAAK,EAAI,EAAG,CAAC,EAAI,EACpF,EAAS,IAAW,8BAAgC,EAAsB,CAAS,EAAI,GAC7F,MAAO,GAAG,EAAc,IAAI,CAAM,EAAE,GAAG,EAAa,IAAI,EAAY,MAAM,IAAa,EAAO,EAChG,CAAC,CAAC,CACD,KAAK,IAAI,EACZ,MAAO,GAAG,EAAU,MAAM,IAAI,EAAkB,EAAW,CAAQ,GACrE,CAEA,SAAS,EAAsB,CAAE,gBAAgB,CAAC,GAAwB,CACxE,IAAM,EAAS,EAAc,KAC1B,CAAE,YAAW,UAAS,OAAM,yBAC3B,IAAI,IAAY,IAAY,EAAY,GAAK,IAAI,MAAY,EAAO,IAAI,EAAU,CAAI,IAAM,GAAG,GAAG,GACtG,EACA,OAAO,EAAO,OAAS,EAAI,mBAAmB,EAAO,KAAK,IAAI,IAAM,EACtE,CAEA,SAAS,EAAkB,EAAsB,EAA0B,CACzE,OAAQ,EAAU,KAAlB,CACE,IAAK,OACH,MAAO,GAAG,EAAU,KAAK,IAAI,IAE/B,IAAK,WACH,MAAO,GAAG,EAAe,CAAS,EAAE,GAAG,EAAU,EAAU,MAAQ,EAAE,EAAE,IAAI,IAE7E,IAAK,cAAe,CAClB,IAAM,GAAY,EAAU,UAAY,CAAC,EAAA,CAAG,IAAK,GAAY,EAAe,CAAO,CAAC,EAC9E,EAAS,CAAC,GAAG,EAAS,MAAM,EAAG,CAAiB,EAAG,GAAI,EAAS,OAAS,EAAoB,CAAC,KAAK,EAAI,CAAC,CAAE,EAChH,MAAO,GAAG,EAAe,CAAS,EAAE,IAAI,IAAW,EAAO,OAAS,EAAI,aAAa,EAAO,KAAK,IAAI,IAAM,IAC5G,CACF,CACF,CAGA,SAAS,EAAU,EAAsB,CACvC,OAAO,EAAK,WAAW,iBAAkB,GAAG,CAC9C,CAEA,SAAS,EAAe,CAAE,OAAM,YAAW,WAAkC,CAC3E,MAAO,GAAG,EAAK,GAAG,EAAU,GAAG,GACjC,CAEA,SAAS,EAAgB,EAA6B,EAAmB,EAA2B,CAClG,EAAA,YACE,KAAK,UACH,CACE,OAAQ,CAAC,EAAU,EAAO,UAAU,GAAK,EAAM,OAAO,SAAW,EACjE,KAAM,EAAW,KACjB,UAAW,EAAM,UACjB,QAAS,CACP,GAAG,EAAgB,EAAO,UAAU,EACpC,iBAAkB,EAAM,MAAM,OAC9B,qBAAsB,EAAO,oBAC/B,EACA,WAAY,EAAO,WACnB,OAAQ,EAAM,OACd,SAAU,EAAM,QAClB,EACA,IAAA,GACA,CACF,EAAI;CACN,CACF"}
+{"version":3,"file":"checkCommand.cjs","names":["resolveTarget","loadConfig","configSearchDirectory","resolveOptions","checkThresholds","resolveLimits","getErrorMessage","readFile","path","realpath","stat","getLanguage","resolveRepoRoot","scanTarget","loadExclusion","formatPath","isIgnored","listRepositoryFiles","keepPaths","loadRepositoryExclusion","scanListedFiles","isWithinDirectory","resolveMergeBase","listChangedFiles","mapConcurrently","os","listLineHunks","thresholds","metricNameOf"],"sources":["../src/checkCommand.ts"],"sourcesContent":["import { readFile, realpath, stat } from 'node:fs/promises';\nimport os from 'node:os';\nimport path from 'node:path';\nimport { getErrorMessage, mapConcurrently } from '@willbooster/shared-lib';\nimport { loadConfig, resolveOptions, type CliOptions, type ResolvedOptions } from './cliConfig.js';\nimport { keepPaths, loadExclusion, loadRepositoryExclusion } from './exclusion.js';\nimport {\n  isIgnored,\n  listChangedFiles,\n  listLineHunks,\n  listRepositoryFiles,\n  resolveMergeBase,\n  resolveRepoRoot,\n  type ChangedFile,\n  type LineHunk,\n} from './git.js';\nimport {\n  addCrossFileDuplication,\n  configSearchDirectory,\n  formatPath,\n  getLanguage,\n  isWithinDirectory,\n  resolveTarget,\n  scanListedFiles,\n  scanTarget,\n  writeStderr,\n  writeStdout,\n  type ScanResult,\n} from './scan.js';\nimport {\n  checkThresholds,\n  type BlockLocation,\n  type CheckedFile,\n  type CheckedDuplication,\n  type CheckResult,\n  type Violation,\n} from './thresholdCheck.js';\nimport { metricNameOf, resolveLimits, thresholds, type LimitsByLevel } from './thresholds.js';\n\n/** Raw options of the `check` subcommand; every field is undefined unless the flag was passed. */\nexport interface CheckCliOptions extends CliOptions {\n  base?: string;\n}\n\n/** The measured files a check covers and the duplication they take part in. */\nexport interface CheckScope {\n  files: CheckedFile[];\n  duplication: CheckedDuplication;\n  /** Measurement failures of files the check covers. */\n  errors: string[];\n  warnings: string[];\n  /** The directory the files' paths are relative to. */\n  root: string;\n  mergeBase?: string;\n}\n\n/** Caps how many partner locations a duplicated block lists so the report stays scannable. */\nconst maxListedPartners = 3;\n\n/**\n * Reports every exceeded threshold of the target or, with `base`, of what the working tree changed\n * under it since the merge-base with that ref. Exit codes: 0 no error-level violations, 1 error-level\n * violations, 2 files the check covers could not be measured.\n */\nexport async function runCheckCommand(\n  target: string,\n  cliOptions: CheckCliOptions,\n  cliLimits: LimitsByLevel\n): Promise<void> {\n  try {\n    const resolvedTarget = resolveTarget(target);\n    const config = await loadConfig(cliOptions.config, await configSearchDirectory(resolvedTarget));\n    const options = resolveOptions(cliOptions, config);\n    const scope = await scanCheckScope(resolvedTarget, cliOptions.base, options);\n    const result = checkThresholds(scope.files, scope.duplication, resolveLimits(cliLimits, config.config.thresholds));\n    await endFileViolationsAtLastLine(result.violations, scope.root);\n\n    if (options.json) {\n      printJsonReport(cliOptions, scope, result);\n    } else {\n      printTextReport(cliOptions, scope, result);\n    }\n    if (scope.errors.length > 0) {\n      process.exitCode = 2;\n    } else if (hasErrors(result.violations)) {\n      process.exitCode = 1;\n    }\n  } catch (error) {\n    writeStderr(`Error: ${getErrorMessage(error)}\\n`);\n    process.exitCode = 2;\n  }\n}\n\nfunction hasErrors(violations: readonly Violation[]): boolean {\n  return violations.some(({ level }) => level === 'error');\n}\n\n/**\n * A file's measured line count includes the empty line after a final line terminator, which is\n * not a line of the file, so file-level violations end at the last line that exists.\n */\nasync function endFileViolationsAtLastLine(violations: readonly Violation[], root: string): Promise<void> {\n  for (const violation of violations.filter(({ kind }) => kind === 'file')) {\n    const content = await readFile(path.join(root, violation.file), 'utf8');\n    if (/[\\n\\r]$/u.test(content)) {\n      violation.endLine = Math.max(violation.endLine - 1, 1);\n    }\n  }\n}\n\n/**\n * Inside a git repository, the scope is the repository's git-visible files under the target, and\n * duplication is detected against all of them; `base` only narrows the scope to what changed.\n * Outside a repository, where there is no change to narrow to, the target directory is walked.\n */\nexport async function scanCheckScope(\n  resolvedTarget: string,\n  base: string | undefined,\n  options: ResolvedOptions\n): Promise<CheckScope> {\n  const canonicalTarget = await realpath(resolvedTarget);\n  const targetStat = await stat(canonicalTarget);\n  const targetFile = targetStat.isFile() ? canonicalTarget : undefined;\n  if (targetFile && !getLanguage(targetFile, options, true)) {\n    throw new Error(`${path.basename(targetFile)}: unsupported file type`);\n  }\n  let repoRoot;\n  try {\n    repoRoot = await realpath(await resolveRepoRoot(targetFile ? path.dirname(canonicalTarget) : canonicalTarget));\n  } catch (error) {\n    if (base !== undefined) {\n      throw error;\n    }\n    return await scanDirectoryWalk(canonicalTarget, options);\n  }\n  return await scanRepository({ repoRoot, canonicalTarget, targetFile }, base, options);\n}\n\nasync function scanDirectoryWalk(canonicalTarget: string, options: ResolvedOptions): Promise<CheckScope> {\n  const searchDirectory = await configSearchDirectory(canonicalTarget);\n  const scan = await scanTarget(canonicalTarget, {\n    ...options,\n    loadExclusion: (absolutePaths) => loadExclusion(searchDirectory, options.exclude, absolutePaths),\n  });\n  return {\n    files: scan.files.map(({ file, metrics }) => ({ file: formatPath(file, scan.displayRoot), metrics })),\n    duplication: measureDuplication(scan, options),\n    errors: scan.errors,\n    warnings: scan.warnings,\n    root: scan.displayRoot,\n  };\n}\n\ninterface RepositoryTarget {\n  repoRoot: string;\n  canonicalTarget: string;\n  /** Set when the target is a file, which is measured even when git ignores it or it is excluded. */\n  targetFile?: string;\n}\n\n/**\n * Measures every git-visible file (tracked or untracked non-ignored) of the repository, so that\n * code copied from outside the target is detected as duplication, and keeps the files under the\n * target: all of them or, with `base`, the changed ones with their line hunks.\n */\nasync function scanRepository(\n  { repoRoot, canonicalTarget, targetFile }: RepositoryTarget,\n  base: string | undefined,\n  options: ResolvedOptions\n): Promise<CheckScope> {\n  // Only git-visible files are measured, so a directory git ignores would pass as a check of no\n  // files. A directory git lists but whose sources it ignores is not told apart from a repository\n  // or package whose only sources are ignored build output, which must pass.\n  if (targetFile === undefined && canonicalTarget !== repoRoot && (await isIgnored(repoRoot, canonicalTarget))) {\n    throw new Error(\n      `${formatPath(canonicalTarget, repoRoot)}: git ignores this directory, so no file in it is checked; target a file to check it regardless`\n    );\n  }\n  const repositoryFiles = await listRepositoryFiles(repoRoot);\n  const explicitFiles = new Set(targetFile === undefined ? [] : [targetFile]);\n  const unlistedTarget =\n    targetFile !== undefined && !repositoryFiles.has(formatPath(targetFile, repoRoot))\n      ? formatPath(targetFile, repoRoot)\n      : undefined;\n  if (unlistedTarget !== undefined) {\n    repositoryFiles.add(unlistedTarget);\n  }\n  const exclusion = keepPaths(await loadRepositoryExclusion(repoRoot, repositoryFiles, options.exclude), explicitFiles);\n  // Scan errors start with the file or directory they concern.\n  const targetPath = formatPath(canonicalTarget, repoRoot);\n  const scan = await scanListedFiles(\n    repoRoot,\n    repositoryFiles,\n    { ...options, loadExclusion: () => Promise.resolve(exclusion) },\n    explicitFiles,\n    targetFile === undefined && canonicalTarget !== repoRoot ? targetPath : ''\n  );\n  const duplication = measureDuplication(scan, options);\n  const isInTarget = (relativePath: string): boolean =>\n    isWithinDirectory(path.join(repoRoot, relativePath), canonicalTarget);\n  const measuredFiles = scan.files.map(({ file, metrics }) => ({ file: formatPath(file, repoRoot), metrics }));\n  let files = measuredFiles.filter(({ file }) => isInTarget(file));\n  let isCovered = (error: string): boolean =>\n    canonicalTarget === repoRoot || error.startsWith(`${targetPath}/`) || error.startsWith(`${targetPath}:`);\n  let mergeBase;\n  if (base !== undefined) {\n    mergeBase = await resolveMergeBase(repoRoot, base);\n    const changedFiles = await listChangedFiles(repoRoot, mergeBase);\n    // A targeted file git does not list has no diff; all of it counts as added.\n    if (unlistedTarget !== undefined) {\n      changedFiles.push({ status: 'added', headPath: unlistedTarget });\n    }\n    const changesInTarget = changedFiles.filter(({ status, headPath }) => status !== 'deleted' && isInTarget(headPath));\n    files = await attachHunks(measuredFiles, changesInTarget, repoRoot, mergeBase);\n    isCovered = (error) => changesInTarget.some(({ headPath }) => error.startsWith(`${headPath}:`));\n  }\n  return {\n    files,\n    duplication,\n    ...partitionErrors(scan, isCovered),\n    root: repoRoot,\n    mergeBase,\n  };\n}\n\n/** Only a failure on a file the check covers leaves it incomplete; the others are warnings. */\nfunction partitionErrors(\n  scan: ScanResult,\n  isCovered: (error: string) => boolean\n): Pick<CheckScope, 'errors' | 'warnings'> {\n  return {\n    errors: scan.errors.filter((error) => isCovered(error)),\n    warnings: [...scan.warnings, ...scan.errors.filter((error) => !isCovered(error))],\n  };\n}\n\n/** The measured changed files with their line hunks against the merge-base. */\nasync function attachHunks(\n  measuredFiles: readonly CheckedFile[],\n  changedFiles: readonly ChangedFile[],\n  repoRoot: string,\n  mergeBase: string\n): Promise<CheckedFile[]> {\n  const measuredByPath = new Map(measuredFiles.map((measured) => [measured.file, measured]));\n  const files = await mapConcurrently(changedFiles, os.availableParallelism() * 2, async (changed) => {\n    const measured = measuredByPath.get(changed.headPath);\n    // A changed file the scan left out (unsupported, excluded, or failed) has nothing to check.\n    if (measured === undefined) {\n      return [];\n    }\n    const { metrics } = measured;\n    const wholeFile: LineHunk = { headStart: 1, headCount: metrics.lines.total };\n    const hunks =\n      changed.basePath === undefined\n        ? [wholeFile]\n        : await listLineHunks(repoRoot, mergeBase, changed.basePath, changed.headPath);\n    if (metrics.language === 'python' && hunks.some(({ headCount }) => headCount === 0)) {\n      anchorDeletionsToCode(hunks, await readFile(path.join(repoRoot, changed.headPath), 'utf8'));\n    }\n    return [{ ...measured, hunks }];\n  });\n  return files.flat();\n}\n\n/**\n * Moves each pure deletion up to the last code line before it. A Python function ends at its last\n * statement, so the blank and comment-only lines left above a deleted tail lie outside the span the\n * deletion shortened; anchored to that statement, the deletion counts as touching the function.\n */\nfunction anchorDeletionsToCode(hunks: LineHunk[], content: string): void {\n  const lines = content.split(/\\r\\n|\\n|\\r/u);\n  for (const hunk of hunks) {\n    if (hunk.headCount > 0) continue;\n    while (hunk.headStart > 0 && /^\\s*(?:#.*)?$/u.test(lines[hunk.headStart - 1] ?? '')) {\n      hunk.headStart -= 1;\n    }\n  }\n}\n\n/** A run-wide failure (a missing target or native addon) leaves nothing to check. */\nfunction measureDuplication(scan: ScanResult, options: ResolvedOptions): CheckedDuplication {\n  if (scan.fatalError) {\n    throw new Error(scan.fatalError);\n  }\n  addCrossFileDuplication(scan, options);\n  return {\n    crossFile: scan.crossFileDuplication,\n    minSimilarityPercent: options.duplication.minSimilarityPercent,\n  };\n}\n\nfunction printTextReport(cliOptions: CheckCliOptions, scope: CheckScope, result: CheckResult): void {\n  const { violations } = result;\n  const checked = `${scope.files.length} ${cliOptions.base === undefined ? '' : 'changed '}files, ${result.checkedFunctionCount} functions checked${describeBase(cliOptions, scope)}`;\n  if (scope.errors.length > 0) {\n    // Unmeasured files were not checked, so \"no violations\" would be vacuous; never claim a pass.\n    writeStdout(\n      `Check could not complete: ${scope.errors.length} measurement failures (details on stderr); ${describeCounts(violations)} in the measured files (${checked}).\\n`\n    );\n  } else if (violations.length === 0) {\n    writeStdout(`No threshold violations: ${checked}.\\n`);\n  } else {\n    writeStdout(`Threshold violations: ${describeCounts(violations)} (${checked}).\\n`);\n  }\n  for (const violation of violations) {\n    writeStdout(`${formatViolation(violation)}\\n`);\n  }\n\n  const violatedMetrics = new Set(violations.flatMap(({ exceeded }) => exceeded.map(({ metric }) => metric)));\n  const violatedThresholds = thresholds.filter((threshold) => violatedMetrics.has(metricNameOf(threshold)));\n  if (violatedThresholds.length > 0) {\n    writeStdout('\\nHow to fix:\\n');\n    for (const { label, hint } of violatedThresholds) {\n      writeStdout(`- ${label}: ${hint}\\n`);\n    }\n  }\n\n  for (const warning of scope.warnings) {\n    writeStderr(`Warning: ${warning}\\n`);\n  }\n  for (const error of scope.errors) {\n    writeStderr(`Error: ${error}\\n`);\n  }\n}\n\nfunction describeBase(cliOptions: CheckCliOptions, scope: CheckScope): string {\n  return scope.mergeBase === undefined ? '' : `; base ${cliOptions.base}, merge-base ${scope.mergeBase.slice(0, 12)}`;\n}\n\nfunction describeCounts(violations: readonly Violation[]): string {\n  const counts = countViolations(violations);\n  return `${counts.errorViolationCount} errors, ${counts.warningViolationCount} warnings (${counts.functionViolationCount} functions, ${counts.fileViolationCount} files, ${counts.duplicationViolationCount} duplicated blocks)`;\n}\n\nfunction countViolations(violations: readonly Violation[]): {\n  violationCount: number;\n  errorViolationCount: number;\n  warningViolationCount: number;\n  functionViolationCount: number;\n  fileViolationCount: number;\n  duplicationViolationCount: number;\n} {\n  const count = (kind: Violation['kind']): number => violations.filter((violation) => violation.kind === kind).length;\n  const errorViolationCount = violations.filter(({ level }) => level === 'error').length;\n  return {\n    violationCount: violations.length,\n    errorViolationCount,\n    warningViolationCount: violations.length - errorViolationCount,\n    functionViolationCount: count('function'),\n    fileViolationCount: count('file'),\n    duplicationViolationCount: count('duplication'),\n  };\n}\n\n/** One violation as a single line: its level, the location, then every limit it exceeds. */\nfunction formatViolation(violation: Violation): string {\n  const labelByMetric = new Map(thresholds.map((threshold) => [metricNameOf(threshold), threshold.label]));\n  const exceeded = violation.exceeded\n    .map(({ metric, value, level, limit }) => {\n      // Halstead values are fractional; rounding up keeps a violating value above the printed maximum.\n      const roundedValue = Math.ceil(value * 10) / 10;\n      const levelPrefix = level === violation.level ? '' : `${level} `;\n      // A duplicated block violates from its limit on, so the largest allowed code-line count is\n      // the last whole number below it; a limit of 0 allows none.\n      const maxAllowed = violation.kind === 'duplication' ? Math.max(Math.ceil(limit) - 1, 0) : limit;\n      const blocks = metric === 'functionCognitiveComplexity' ? describeLargestBlocks(violation) : '';\n      return `${labelByMetric.get(metric)} ${roundedValue} (${levelPrefix}max ${maxAllowed}${blocks})`;\n    })\n    .join(', ');\n  return `${violation.level}: ${describeViolation(violation, exceeded)}`;\n}\n\nfunction describeLargestBlocks({ largestBlocks = [] }: Violation): string {\n  const blocks = largestBlocks.map(\n    ({ startLine, endLine, name, cognitiveComplexity }) =>\n      `L${startLine}${endLine === startLine ? '' : `-${endLine}`}${name ? ` ${toOneLine(name)}` : ''} ${cognitiveComplexity}`\n  );\n  return blocks.length > 0 ? `; largest parts ${blocks.join(', ')}` : '';\n}\n\nfunction describeViolation(violation: Violation, exceeded: string): string {\n  switch (violation.kind) {\n    case 'file': {\n      return `${violation.file}: ${exceeded}`;\n    }\n    case 'function': {\n      return `${formatLocation(violation)} ${toOneLine(violation.name ?? '')}: ${exceeded}`;\n    }\n    case 'duplication': {\n      const partners = (violation.partners ?? []).map((partner) => formatLocation(partner));\n      const listed = [...partners.slice(0, maxListedPartners), ...(partners.length > maxListedPartners ? ['...'] : [])];\n      return `${formatLocation(violation)}: ${exceeded}${listed.length > 0 ? `, also at ${listed.join(', ')}` : ''}`;\n    }\n  }\n}\n\n/** A computed name can span lines in the source; the report keeps one line per function. */\nfunction toOneLine(name: string): string {\n  return name.replaceAll(/\\s*[\\n\\r]\\s*/gu, ' ');\n}\n\nfunction formatLocation({ file, startLine, endLine }: BlockLocation): string {\n  return `${file}:${startLine}-${endLine}`;\n}\n\nfunction printJsonReport(cliOptions: CheckCliOptions, scope: CheckScope, result: CheckResult): void {\n  writeStdout(\n    JSON.stringify(\n      {\n        passed: !hasErrors(result.violations) && scope.errors.length === 0,\n        base: cliOptions.base,\n        mergeBase: scope.mergeBase,\n        summary: {\n          ...countViolations(result.violations),\n          checkedFileCount: scope.files.length,\n          checkedFunctionCount: result.checkedFunctionCount,\n        },\n        violations: result.violations,\n        errors: scope.errors,\n        warnings: scope.warnings,\n      },\n      undefined,\n      2\n    ) + '\\n'\n  );\n}\n"],"mappings":"mYAgEA,eAAsB,EACpB,EACA,EACA,EACe,CACf,GAAI,CACF,IAAM,EAAiBA,EAAAA,cAAc,CAAM,EACrC,EAAS,MAAMC,EAAAA,WAAW,EAAW,OAAQ,MAAMC,EAAAA,sBAAsB,CAAc,CAAC,EACxF,EAAUC,EAAAA,eAAe,EAAY,CAAM,EAC3C,EAAQ,MAAM,EAAe,EAAgB,EAAW,KAAM,CAAO,EACrE,EAASC,EAAAA,gBAAgB,EAAM,MAAO,EAAM,YAAaC,EAAAA,cAAc,EAAW,EAAO,OAAO,UAAU,CAAC,EACjH,MAAM,EAA4B,EAAO,WAAY,EAAM,IAAI,EAE3D,EAAQ,KACV,EAAgB,EAAY,EAAO,CAAM,EAEzC,EAAgB,EAAY,EAAO,CAAM,EAEvC,EAAM,OAAO,OAAS,EACxB,QAAQ,SAAW,EACV,EAAU,EAAO,UAAU,IACpC,QAAQ,SAAW,EAEvB,OAAS,EAAO,CACd,EAAA,YAAY,WAAA,EAAUC,EAAAA,gBAAAA,CAAgB,CAAK,EAAE,GAAG,EAChD,QAAQ,SAAW,CACrB,CACF,CAEA,SAAS,EAAU,EAA2C,CAC5D,OAAO,EAAW,MAAM,CAAE,WAAY,IAAU,OAAO,CACzD,CAMA,eAAe,EAA4B,EAAkC,EAA6B,CACxG,IAAK,IAAM,KAAa,EAAW,QAAQ,CAAE,UAAW,IAAS,MAAM,EAAG,CACxE,IAAM,EAAU,MAAA,EAAMC,EAAAA,SAAAA,CAASC,EAAAA,QAAK,KAAK,EAAM,EAAU,IAAI,EAAG,MAAM,EAClE,WAAW,KAAK,CAAO,IACzB,EAAU,QAAU,KAAK,IAAI,EAAU,QAAU,EAAG,CAAC,EAEzD,CACF,CAOA,eAAsB,EACpB,EACA,EACA,EACqB,CACrB,IAAM,EAAkB,MAAA,EAAMC,EAAAA,SAAAA,CAAS,CAAc,EAE/C,GAAa,MAAA,EADMC,EAAAA,KAAAA,CAAK,CAAe,EAAA,CACf,OAAO,EAAI,EAAkB,IAAA,GAC3D,GAAI,GAAc,CAACC,EAAAA,YAAY,EAAY,EAAS,EAAI,EACtD,MAAU,MAAM,GAAGH,EAAAA,QAAK,SAAS,CAAU,EAAE,wBAAwB,EAEvE,IAAI,EACJ,GAAI,CACF,EAAW,MAAA,EAAMC,EAAAA,SAAAA,CAAS,MAAMG,EAAAA,gBAAgB,EAAaJ,EAAAA,QAAK,QAAQ,CAAe,EAAI,CAAe,CAAC,CAC/G,OAAS,EAAO,CACd,GAAI,IAAS,IAAA,GACX,MAAM,EAER,OAAO,MAAM,EAAkB,EAAiB,CAAO,CACzD,CACA,OAAO,MAAM,EAAe,CAAE,WAAU,kBAAiB,YAAW,EAAG,EAAM,CAAO,CACtF,CAEA,eAAe,EAAkB,EAAyB,EAA+C,CACvG,IAAM,EAAkB,MAAMN,EAAAA,sBAAsB,CAAe,EAC7D,EAAO,MAAMW,EAAAA,WAAW,EAAiB,CAC7C,GAAG,EACH,cAAgB,GAAkBC,EAAAA,cAAc,EAAiB,EAAQ,QAAS,CAAa,CACjG,CAAC,EACD,MAAO,CACL,MAAO,EAAK,MAAM,KAAK,CAAE,OAAM,cAAe,CAAE,KAAMC,EAAAA,WAAW,EAAM,EAAK,WAAW,EAAG,SAAQ,EAAE,EACpG,YAAa,EAAmB,EAAM,CAAO,EAC7C,OAAQ,EAAK,OACb,SAAU,EAAK,SACf,KAAM,EAAK,WACb,CACF,CAcA,eAAe,EACb,CAAE,WAAU,kBAAiB,cAC7B,EACA,EACqB,CAIrB,GAAI,IAAe,IAAA,IAAa,IAAoB,GAAa,MAAMC,EAAAA,UAAU,EAAU,CAAe,EACxG,MAAU,MACR,GAAGD,EAAAA,WAAW,EAAiB,CAAQ,EAAE,gGAC3C,EAEF,IAAM,EAAkB,MAAME,EAAAA,oBAAoB,CAAQ,EACpD,EAAgB,IAAI,IAAI,IAAe,IAAA,GAAY,CAAC,EAAI,CAAC,CAAU,CAAC,EACpE,EACJ,IAAe,IAAA,IAAa,CAAC,EAAgB,IAAIF,EAAAA,WAAW,EAAY,CAAQ,CAAC,EAC7EA,EAAAA,WAAW,EAAY,CAAQ,EAC/B,IAAA,GACF,IAAmB,IAAA,IACrB,EAAgB,IAAI,CAAc,EAEpC,IAAM,EAAYG,EAAAA,UAAU,MAAMC,EAAAA,wBAAwB,EAAU,EAAiB,EAAQ,OAAO,EAAG,CAAa,EAE9G,EAAaJ,EAAAA,WAAW,EAAiB,CAAQ,EACjD,EAAO,MAAMK,EAAAA,gBACjB,EACA,EACA,CAAE,GAAG,EAAS,kBAAqB,QAAQ,QAAQ,CAAS,CAAE,EAC9D,EACA,IAAe,IAAA,IAAa,IAAoB,EAAW,EAAa,EAC1E,EACM,EAAc,EAAmB,EAAM,CAAO,EAC9C,EAAc,GAClBC,EAAAA,kBAAkBb,EAAAA,QAAK,KAAK,EAAU,CAAY,EAAG,CAAe,EAChE,EAAgB,EAAK,MAAM,KAAK,CAAE,OAAM,cAAe,CAAE,KAAMO,EAAAA,WAAW,EAAM,CAAQ,EAAG,SAAQ,EAAE,EACvG,EAAQ,EAAc,QAAQ,CAAE,UAAW,EAAW,CAAI,CAAC,EAC3D,EAAa,GACf,IAAoB,GAAY,EAAM,WAAW,GAAG,EAAW,EAAE,GAAK,EAAM,WAAW,GAAG,EAAW,EAAE,EACrG,EACJ,GAAI,IAAS,IAAA,GAAW,CACtB,EAAY,MAAMO,EAAAA,iBAAiB,EAAU,CAAI,EACjD,IAAM,EAAe,MAAMC,EAAAA,iBAAiB,EAAU,CAAS,EAE3D,IAAmB,IAAA,IACrB,EAAa,KAAK,CAAE,OAAQ,QAAS,SAAU,CAAe,CAAC,EAEjE,IAAM,EAAkB,EAAa,QAAQ,CAAE,SAAQ,cAAe,IAAW,WAAa,EAAW,CAAQ,CAAC,EAClH,EAAQ,MAAM,EAAY,EAAe,EAAiB,EAAU,CAAS,EAC7E,EAAa,GAAU,EAAgB,MAAM,CAAE,cAAe,EAAM,WAAW,GAAG,EAAS,EAAE,CAAC,CAChG,CACA,MAAO,CACL,QACA,cACA,GAAG,EAAgB,EAAM,CAAS,EAClC,KAAM,EACN,WACF,CACF,CAGA,SAAS,EACP,EACA,EACyC,CACzC,MAAO,CACL,OAAQ,EAAK,OAAO,OAAQ,GAAU,EAAU,CAAK,CAAC,EACtD,SAAU,CAAC,GAAG,EAAK,SAAU,GAAG,EAAK,OAAO,OAAQ,GAAU,CAAC,EAAU,CAAK,CAAC,CAAC,CAClF,CACF,CAGA,eAAe,EACb,EACA,EACA,EACA,EACwB,CACxB,IAAM,EAAiB,IAAI,IAAI,EAAc,IAAK,GAAa,CAAC,EAAS,KAAM,CAAQ,CAAC,CAAC,EAkBzF,OAAO,MAAA,EAjBaC,EAAAA,gBAAAA,CAAgB,EAAcC,EAAAA,QAAG,qBAAqB,EAAI,EAAG,KAAO,IAAY,CAClG,IAAM,EAAW,EAAe,IAAI,EAAQ,QAAQ,EAEpD,GAAI,IAAa,IAAA,GACf,MAAO,CAAC,EAEV,GAAM,CAAE,WAAY,EACd,EAAsB,CAAE,UAAW,EAAG,UAAW,EAAQ,MAAM,KAAM,EACrE,EACJ,EAAQ,WAAa,IAAA,GACjB,CAAC,CAAS,EACV,MAAMC,EAAAA,cAAc,EAAU,EAAW,EAAQ,SAAU,EAAQ,QAAQ,EAIjF,OAHI,EAAQ,WAAa,UAAY,EAAM,MAAM,CAAE,eAAgB,IAAc,CAAC,GAChF,EAAsB,EAAO,MAAA,EAAMnB,EAAAA,SAAAA,CAASC,EAAAA,QAAK,KAAK,EAAU,EAAQ,QAAQ,EAAG,MAAM,CAAC,EAErF,CAAC,CAAE,GAAG,EAAU,OAAM,CAAC,CAChC,CAAC,EAAA,CACY,KAAK,CACpB,CAOA,SAAS,EAAsB,EAAmB,EAAuB,CACvE,IAAM,EAAQ,EAAQ,MAAM,aAAa,EACzC,IAAK,IAAM,KAAQ,EACb,OAAK,UAAY,GACrB,KAAO,EAAK,UAAY,GAAK,iBAAiB,KAAK,EAAM,EAAK,UAAY,IAAM,EAAE,GAChF,IAAK,SAGX,CAGA,SAAS,EAAmB,EAAkB,EAA8C,CAC1F,GAAI,EAAK,WACP,MAAU,MAAM,EAAK,UAAU,EAGjC,OADA,EAAA,wBAAwB,EAAM,CAAO,EAC9B,CACL,UAAW,EAAK,qBAChB,qBAAsB,EAAQ,YAAY,oBAC5C,CACF,CAEA,SAAS,EAAgB,EAA6B,EAAmB,EAA2B,CAClG,GAAM,CAAE,cAAe,EACjB,EAAU,GAAG,EAAM,MAAM,OAAO,GAAG,EAAW,OAAS,IAAA,GAAY,GAAK,WAAW,SAAS,EAAO,qBAAqB,oBAAoB,EAAa,EAAY,CAAK,IAC5K,EAAM,OAAO,OAAS,EAExB,EAAA,YACE,6BAA6B,EAAM,OAAO,OAAO,6CAA6C,EAAe,CAAU,EAAE,0BAA0B,EAAQ,KAC7J,EACS,EAAW,SAAW,EAC/B,EAAA,YAAY,4BAA4B,EAAQ,IAAI,EAEpD,EAAA,YAAY,yBAAyB,EAAe,CAAU,EAAE,IAAI,EAAQ,KAAK,EAEnF,IAAK,IAAM,KAAa,EACtB,EAAA,YAAY,GAAG,EAAgB,CAAS,EAAE,GAAG,EAG/C,IAAM,EAAkB,IAAI,IAAI,EAAW,SAAS,CAAE,cAAe,EAAS,KAAK,CAAE,YAAa,CAAM,CAAC,CAAC,EACpG,EAAqBmB,EAAAA,WAAW,OAAQ,GAAc,EAAgB,IAAIC,EAAAA,aAAa,CAAS,CAAC,CAAC,EACxG,GAAI,EAAmB,OAAS,EAAG,CACjC,EAAA,YAAY;;CAAiB,EAC7B,IAAK,GAAM,CAAE,QAAO,UAAU,EAC5B,EAAA,YAAY,KAAK,EAAM,IAAI,EAAK,GAAG,CAEvC,CAEA,IAAK,IAAM,KAAW,EAAM,SAC1B,EAAA,YAAY,YAAY,EAAQ,GAAG,EAErC,IAAK,IAAM,KAAS,EAAM,OACxB,EAAA,YAAY,UAAU,EAAM,GAAG,CAEnC,CAEA,SAAS,EAAa,EAA6B,EAA2B,CAC5E,OAAO,EAAM,YAAc,IAAA,GAAY,GAAK,UAAU,EAAW,KAAK,eAAe,EAAM,UAAU,MAAM,EAAG,EAAE,GAClH,CAEA,SAAS,EAAe,EAA0C,CAChE,IAAM,EAAS,EAAgB,CAAU,EACzC,MAAO,GAAG,EAAO,oBAAoB,WAAW,EAAO,sBAAsB,aAAa,EAAO,uBAAuB,cAAc,EAAO,mBAAmB,UAAU,EAAO,0BAA0B,oBAC7M,CAEA,SAAS,EAAgB,EAOvB,CACA,IAAM,EAAS,GAAoC,EAAW,OAAQ,GAAc,EAAU,OAAS,CAAI,CAAC,CAAC,OACvG,EAAsB,EAAW,QAAQ,CAAE,WAAY,IAAU,OAAO,CAAC,CAAC,OAChF,MAAO,CACL,eAAgB,EAAW,OAC3B,sBACA,sBAAuB,EAAW,OAAS,EAC3C,uBAAwB,EAAM,UAAU,EACxC,mBAAoB,EAAM,MAAM,EAChC,0BAA2B,EAAM,aAAa,CAChD,CACF,CAGA,SAAS,EAAgB,EAA8B,CACrD,IAAM,EAAgB,IAAI,IAAID,EAAAA,WAAW,IAAK,GAAc,CAACC,EAAAA,aAAa,CAAS,EAAG,EAAU,KAAK,CAAC,CAAC,EACjG,EAAW,EAAU,SACxB,KAAK,CAAE,SAAQ,QAAO,QAAO,WAAY,CAExC,IAAM,EAAe,KAAK,KAAK,EAAQ,EAAE,EAAI,GACvC,EAAc,IAAU,EAAU,MAAQ,GAAK,GAAG,EAAM,GAGxD,EAAa,EAAU,OAAS,cAAgB,KAAK,IAAI,KAAK,KAAK,CAAK,EAAI,EAAG,CAAC,EAAI,EACpF,EAAS,IAAW,8BAAgC,EAAsB,CAAS,EAAI,GAC7F,MAAO,GAAG,EAAc,IAAI,CAAM,EAAE,GAAG,EAAa,IAAI,EAAY,MAAM,IAAa,EAAO,EAChG,CAAC,CAAC,CACD,KAAK,IAAI,EACZ,MAAO,GAAG,EAAU,MAAM,IAAI,EAAkB,EAAW,CAAQ,GACrE,CAEA,SAAS,EAAsB,CAAE,gBAAgB,CAAC,GAAwB,CACxE,IAAM,EAAS,EAAc,KAC1B,CAAE,YAAW,UAAS,OAAM,yBAC3B,IAAI,IAAY,IAAY,EAAY,GAAK,IAAI,MAAY,EAAO,IAAI,EAAU,CAAI,IAAM,GAAG,GAAG,GACtG,EACA,OAAO,EAAO,OAAS,EAAI,mBAAmB,EAAO,KAAK,IAAI,IAAM,EACtE,CAEA,SAAS,EAAkB,EAAsB,EAA0B,CACzE,OAAQ,EAAU,KAAlB,CACE,IAAK,OACH,MAAO,GAAG,EAAU,KAAK,IAAI,IAE/B,IAAK,WACH,MAAO,GAAG,EAAe,CAAS,EAAE,GAAG,EAAU,EAAU,MAAQ,EAAE,EAAE,IAAI,IAE7E,IAAK,cAAe,CAClB,IAAM,GAAY,EAAU,UAAY,CAAC,EAAA,CAAG,IAAK,GAAY,EAAe,CAAO,CAAC,EAC9E,EAAS,CAAC,GAAG,EAAS,MAAM,EAAG,CAAiB,EAAG,GAAI,EAAS,OAAS,EAAoB,CAAC,KAAK,EAAI,CAAC,CAAE,EAChH,MAAO,GAAG,EAAe,CAAS,EAAE,IAAI,IAAW,EAAO,OAAS,EAAI,aAAa,EAAO,KAAK,IAAI,IAAM,IAC5G,CACF,CACF,CAGA,SAAS,EAAU,EAAsB,CACvC,OAAO,EAAK,WAAW,iBAAkB,GAAG,CAC9C,CAEA,SAAS,EAAe,CAAE,OAAM,YAAW,WAAkC,CAC3E,MAAO,GAAG,EAAK,GAAG,EAAU,GAAG,GACjC,CAEA,SAAS,EAAgB,EAA6B,EAAmB,EAA2B,CAClG,EAAA,YACE,KAAK,UACH,CACE,OAAQ,CAAC,EAAU,EAAO,UAAU,GAAK,EAAM,OAAO,SAAW,EACjE,KAAM,EAAW,KACjB,UAAW,EAAM,UACjB,QAAS,CACP,GAAG,EAAgB,EAAO,UAAU,EACpC,iBAAkB,EAAM,MAAM,OAC9B,qBAAsB,EAAO,oBAC/B,EACA,WAAY,EAAO,WACnB,OAAQ,EAAM,OACd,SAAU,EAAM,QAClB,EACA,IAAA,GACA,CACF,EAAI;CACN,CACF"}

package/dist/checkCommand.d.ts

@@ -1,6 +1,5 @@
 import { type CliOptions, type ResolvedOptions } from './cliConfig.js';
-import type { CrossFileDuplicationMetrics } from './crossFileDuplication.js';
-import { type CheckedFile, type CodeLineNumbersByFile } from './thresholdCheck.js';
+import { type CheckedFile, type CheckedDuplication } from './thresholdCheck.js';
 import { type LimitsByLevel } from './thresholds.js';
 /** Raw options of the `check` subcommand; every field is undefined unless the flag was passed. */
 export interface CheckCliOptions extends CliOptions {
@@ -9,9 +8,7 @@ export interface CheckCliOptions extends CliOptions {
 /** The measured files a check covers and the duplication they take part in. */
 export interface CheckScope {
     files: CheckedFile[];
-    crossFileDuplication?: CrossFileDuplicationMetrics;
-    /** Of every measured file, so that a copy outside the files the check covers is sized like one inside. */
-    codeLineNumbersByFile: CodeLineNumbersByFile;
+    duplication: CheckedDuplication;
     /** Measurement failures of files the check covers. */
     errors: string[];
     warnings: string[];

package/dist/checkCommand.js

@@ -1,5 +1,5 @@
-import{metricNameOf as e,resolveLimits as t,thresholds as n}from"./thresholds.js";import{loadConfig as r,resolveOptions as i}from"./cliConfig.js";import{isIgnored as a,listChangedFiles as o,listLineHunks as s,listRepositoryFiles as c,resolveMergeBase as l,resolveRepoRoot as u}from"./git.js";import{keepPaths as d,loadExclusion as f,loadRepositoryExclusion as p}from"./exclusion.js";import{addCrossFileDuplication as m,configSearchDirectory as h,formatPath as g,getLanguage as _,isWithinDirectory as v,resolveTarget as y,scanListedFiles as b,scanTarget as x,writeStderr as S,writeStdout as C}from"./scan.js";import{checkThresholds as w}from"./thresholdCheck.js";import{getErrorMessage as T,mapConcurrently as E}from"@willbooster/shared-lib";import{readFile as D,realpath as O,stat as k}from"node:fs/promises";import A from"node:path";import j from"node:os";async function M(e,n,a){try{let o=y(e),s=await r(n.config,await h(o)),c=i(n,s),l=await F(o,n.base,c),u=w(l.files,l.crossFileDuplication,t(a,s.config.thresholds),l.codeLineNumbersByFile);await P(u.violations,l.root),c.json?Q(n,l,u):U(n,l,u),l.errors.length>0?process.exitCode=2:N(u.violations)&&(process.exitCode=1)}catch(e){S(`Error: ${T(e)}\n`),process.exitCode=2}}function N(e){return e.some(({level:e})=>e===`error`)}async function P(e,t){for(let n of e.filter(({kind:e})=>e===`file`)){let e=await D(A.join(t,n.file),`utf8`);/[\n\r]$/u.test(e)&&(n.endLine=Math.max(n.endLine-1,1))}}async function F(e,t,n){let r=await O(e),i=(await k(r)).isFile()?r:void 0;if(i&&!_(i,n,!0))throw Error(`${A.basename(i)}: unsupported file type`);let a;try{a=await O(await u(i?A.dirname(r):r))}catch(e){if(t!==void 0)throw e;return await I(r,n)}return await L({repoRoot:a,canonicalTarget:r,targetFile:i},t,n)}async function I(e,t){let n=await h(e),r=await x(e,{...t,loadExclusion:e=>f(n,t.exclude,e)}),i=H(r,t);return{files:r.files.map(({file:e,metrics:t})=>({file:g(e,r.displayRoot),metrics:t})),codeLineNumbersByFile:R(r,r.displayRoot),crossFileDuplication:i,errors:r.errors,warnings:r.warnings,root:r.displayRoot}}async function L({repoRoot:e,canonicalTarget:t,targetFile:n},r,i){if(n===void 0&&t!==e&&await a(e,t))throw Error(`${g(t,e)}: git ignores this directory, so no file in it is checked; target a file to check it regardless`);let s=await c(e),u=new Set(n===void 0?[]:[n]),f=n!==void 0&&!s.has(g(n,e))?g(n,e):void 0;f!==void 0&&s.add(f);let m=d(await p(e,s,i.exclude),u),h=g(t,e),_=await b(e,s,{...i,loadExclusion:()=>Promise.resolve(m)},u,n===void 0&&t!==e?h:``),y=H(_,i),x=n=>v(A.join(e,n),t),S=_.files.map(({file:t,metrics:n})=>({file:g(t,e),metrics:n})),C=S.filter(({file:e})=>x(e)),w=n=>t===e||n.startsWith(`${h}/`)||n.startsWith(`${h}:`),T;if(r!==void 0){T=await l(e,r);let t=await o(e,T);f!==void 0&&t.push({status:`added`,headPath:f});let n=t.filter(({status:e,headPath:t})=>e!==`deleted`&&x(t));C=await B(S,n,e,T),w=e=>n.some(({headPath:t})=>e.startsWith(`${t}:`))}return{files:C,crossFileDuplication:y,codeLineNumbersByFile:R(_,e),...z(_,w),root:e,mergeBase:T}}function R(e,t){return new Map(e.files.map(({file:e,metrics:n})=>[g(e,t),new Set(n.codeLineNumbers)]))}function z(e,t){return{errors:e.errors.filter(e=>t(e)),warnings:[...e.warnings,...e.errors.filter(e=>!t(e))]}}async function B(e,t,n,r){let i=new Map(e.map(e=>[e.file,e]));return(await E(t,j.availableParallelism()*2,async e=>{let t=i.get(e.headPath);if(t===void 0)return[];let{metrics:a}=t,o={headStart:1,headCount:a.lines.total},c=e.basePath===void 0?[o]:await s(n,r,e.basePath,e.headPath);return a.language===`python`&&c.some(({headCount:e})=>e===0)&&V(c,await D(A.join(n,e.headPath),`utf8`)),[{...t,hunks:c}]})).flat()}function V(e,t){let n=t.split(/\r\n|\n|\r/u);for(let t of e)if(!(t.headCount>0))for(;t.headStart>0&&/^\s*(?:#.*)?$/u.test(n[t.headStart-1]??``);)--t.headStart}function H(e,t){if(e.fatalError)throw Error(e.fatalError);return m(e,t),e.crossFileDuplication}function U(t,r,i){let{violations:a}=i,o=`${r.files.length} ${t.base===void 0?``:`changed `}files, ${i.checkedFunctionCount} functions checked${W(t,r)}`;r.errors.length>0?C(`Check could not complete: ${r.errors.length} measurement failures (details on stderr); ${G(a)} in the measured files (${o}).\n`):a.length===0?C(`No threshold violations: ${o}.\n`):C(`Threshold violations: ${G(a)} (${o}).\n`);for(let e of a)C(`${q(e)}\n`);let s=new Set(a.flatMap(({exceeded:e})=>e.map(({metric:e})=>e))),c=n.filter(t=>s.has(e(t)));if(c.length>0){C(`
+import{metricNameOf as e,resolveLimits as t,thresholds as n}from"./thresholds.js";import{loadConfig as r,resolveOptions as i}from"./cliConfig.js";import{isIgnored as a,listChangedFiles as o,listLineHunks as s,listRepositoryFiles as c,resolveMergeBase as l,resolveRepoRoot as u}from"./git.js";import{keepPaths as d,loadExclusion as f,loadRepositoryExclusion as p}from"./exclusion.js";import{addCrossFileDuplication as m,configSearchDirectory as h,formatPath as g,getLanguage as _,isWithinDirectory as v,resolveTarget as y,scanListedFiles as b,scanTarget as x,writeStderr as S,writeStdout as C}from"./scan.js";import{checkThresholds as w}from"./thresholdCheck.js";import{getErrorMessage as T,mapConcurrently as E}from"@willbooster/shared-lib";import{readFile as D,realpath as O,stat as k}from"node:fs/promises";import A from"node:path";import j from"node:os";async function M(e,n,a){try{let o=y(e),s=await r(n.config,await h(o)),c=i(n,s),l=await F(o,n.base,c),u=w(l.files,l.duplication,t(a,s.config.thresholds));await P(u.violations,l.root),c.json?Z(n,l,u):H(n,l,u),l.errors.length>0?process.exitCode=2:N(u.violations)&&(process.exitCode=1)}catch(e){S(`Error: ${T(e)}\n`),process.exitCode=2}}function N(e){return e.some(({level:e})=>e===`error`)}async function P(e,t){for(let n of e.filter(({kind:e})=>e===`file`)){let e=await D(A.join(t,n.file),`utf8`);/[\n\r]$/u.test(e)&&(n.endLine=Math.max(n.endLine-1,1))}}async function F(e,t,n){let r=await O(e),i=(await k(r)).isFile()?r:void 0;if(i&&!_(i,n,!0))throw Error(`${A.basename(i)}: unsupported file type`);let a;try{a=await O(await u(i?A.dirname(r):r))}catch(e){if(t!==void 0)throw e;return await I(r,n)}return await L({repoRoot:a,canonicalTarget:r,targetFile:i},t,n)}async function I(e,t){let n=await h(e),r=await x(e,{...t,loadExclusion:e=>f(n,t.exclude,e)});return{files:r.files.map(({file:e,metrics:t})=>({file:g(e,r.displayRoot),metrics:t})),duplication:V(r,t),errors:r.errors,warnings:r.warnings,root:r.displayRoot}}async function L({repoRoot:e,canonicalTarget:t,targetFile:n},r,i){if(n===void 0&&t!==e&&await a(e,t))throw Error(`${g(t,e)}: git ignores this directory, so no file in it is checked; target a file to check it regardless`);let s=await c(e),u=new Set(n===void 0?[]:[n]),f=n!==void 0&&!s.has(g(n,e))?g(n,e):void 0;f!==void 0&&s.add(f);let m=d(await p(e,s,i.exclude),u),h=g(t,e),_=await b(e,s,{...i,loadExclusion:()=>Promise.resolve(m)},u,n===void 0&&t!==e?h:``),y=V(_,i),x=n=>v(A.join(e,n),t),S=_.files.map(({file:t,metrics:n})=>({file:g(t,e),metrics:n})),C=S.filter(({file:e})=>x(e)),w=n=>t===e||n.startsWith(`${h}/`)||n.startsWith(`${h}:`),T;if(r!==void 0){T=await l(e,r);let t=await o(e,T);f!==void 0&&t.push({status:`added`,headPath:f});let n=t.filter(({status:e,headPath:t})=>e!==`deleted`&&x(t));C=await z(S,n,e,T),w=e=>n.some(({headPath:t})=>e.startsWith(`${t}:`))}return{files:C,duplication:y,...R(_,w),root:e,mergeBase:T}}function R(e,t){return{errors:e.errors.filter(e=>t(e)),warnings:[...e.warnings,...e.errors.filter(e=>!t(e))]}}async function z(e,t,n,r){let i=new Map(e.map(e=>[e.file,e]));return(await E(t,j.availableParallelism()*2,async e=>{let t=i.get(e.headPath);if(t===void 0)return[];let{metrics:a}=t,o={headStart:1,headCount:a.lines.total},c=e.basePath===void 0?[o]:await s(n,r,e.basePath,e.headPath);return a.language===`python`&&c.some(({headCount:e})=>e===0)&&B(c,await D(A.join(n,e.headPath),`utf8`)),[{...t,hunks:c}]})).flat()}function B(e,t){let n=t.split(/\r\n|\n|\r/u);for(let t of e)if(!(t.headCount>0))for(;t.headStart>0&&/^\s*(?:#.*)?$/u.test(n[t.headStart-1]??``);)--t.headStart}function V(e,t){if(e.fatalError)throw Error(e.fatalError);return m(e,t),{crossFile:e.crossFileDuplication,minSimilarityPercent:t.duplication.minSimilarityPercent}}function H(t,r,i){let{violations:a}=i,o=`${r.files.length} ${t.base===void 0?``:`changed `}files, ${i.checkedFunctionCount} functions checked${U(t,r)}`;r.errors.length>0?C(`Check could not complete: ${r.errors.length} measurement failures (details on stderr); ${W(a)} in the measured files (${o}).\n`):a.length===0?C(`No threshold violations: ${o}.\n`):C(`Threshold violations: ${W(a)} (${o}).\n`);for(let e of a)C(`${K(e)}\n`);let s=new Set(a.flatMap(({exceeded:e})=>e.map(({metric:e})=>e))),c=n.filter(t=>s.has(e(t)));if(c.length>0){C(`
 How to fix:
-`);for(let{label:e,hint:t}of c)C(`- ${e}: ${t}\n`)}for(let e of r.warnings)S(`Warning: ${e}\n`);for(let e of r.errors)S(`Error: ${e}\n`)}function W(e,t){return t.mergeBase===void 0?``:`; base ${e.base}, merge-base ${t.mergeBase.slice(0,12)}`}function G(e){let t=K(e);return`${t.errorViolationCount} errors, ${t.warningViolationCount} warnings (${t.functionViolationCount} functions, ${t.fileViolationCount} files, ${t.duplicationViolationCount} duplicated blocks)`}function K(e){let t=t=>e.filter(e=>e.kind===t).length,n=e.filter(({level:e})=>e===`error`).length;return{violationCount:e.length,errorViolationCount:n,warningViolationCount:e.length-n,functionViolationCount:t(`function`),fileViolationCount:t(`file`),duplicationViolationCount:t(`duplication`)}}function q(t){let r=new Map(n.map(t=>[e(t),t.label])),i=t.exceeded.map(({metric:e,value:n,level:i,limit:a})=>{let o=Math.ceil(n*10)/10,s=i===t.level?``:`${i} `,c=t.kind===`duplication`?Math.max(Math.ceil(a)-1,0):a,l=e===`functionCognitiveComplexity`?J(t):``;return`${r.get(e)} ${o} (${s}max ${c}${l})`}).join(`, `);return`${t.level}: ${Y(t,i)}`}function J({largestBlocks:e=[]}){let t=e.map(({startLine:e,endLine:t,name:n,cognitiveComplexity:r})=>`L${e}${t===e?``:`-${t}`}${n?` ${X(n)}`:``} ${r}`);return t.length>0?`; largest parts ${t.join(`, `)}`:``}function Y(e,t){switch(e.kind){case`file`:return`${e.file}: ${t}`;case`function`:return`${Z(e)} ${X(e.name??``)}: ${t}`;case`duplication`:{let n=(e.partners??[]).map(e=>Z(e)),r=[...n.slice(0,3),...n.length>3?[`...`]:[]];return`${Z(e)}: ${t}${r.length>0?`, also at ${r.join(`, `)}`:``}`}}}function X(e){return e.replaceAll(/\s*[\n\r]\s*/gu,` `)}function Z({file:e,startLine:t,endLine:n}){return`${e}:${t}-${n}`}function Q(e,t,n){C(JSON.stringify({passed:!N(n.violations)&&t.errors.length===0,base:e.base,mergeBase:t.mergeBase,summary:{...K(n.violations),checkedFileCount:t.files.length,checkedFunctionCount:n.checkedFunctionCount},violations:n.violations,errors:t.errors,warnings:t.warnings},void 0,2)+`
+`);for(let{label:e,hint:t}of c)C(`- ${e}: ${t}\n`)}for(let e of r.warnings)S(`Warning: ${e}\n`);for(let e of r.errors)S(`Error: ${e}\n`)}function U(e,t){return t.mergeBase===void 0?``:`; base ${e.base}, merge-base ${t.mergeBase.slice(0,12)}`}function W(e){let t=G(e);return`${t.errorViolationCount} errors, ${t.warningViolationCount} warnings (${t.functionViolationCount} functions, ${t.fileViolationCount} files, ${t.duplicationViolationCount} duplicated blocks)`}function G(e){let t=t=>e.filter(e=>e.kind===t).length,n=e.filter(({level:e})=>e===`error`).length;return{violationCount:e.length,errorViolationCount:n,warningViolationCount:e.length-n,functionViolationCount:t(`function`),fileViolationCount:t(`file`),duplicationViolationCount:t(`duplication`)}}function K(t){let r=new Map(n.map(t=>[e(t),t.label])),i=t.exceeded.map(({metric:e,value:n,level:i,limit:a})=>{let o=Math.ceil(n*10)/10,s=i===t.level?``:`${i} `,c=t.kind===`duplication`?Math.max(Math.ceil(a)-1,0):a,l=e===`functionCognitiveComplexity`?q(t):``;return`${r.get(e)} ${o} (${s}max ${c}${l})`}).join(`, `);return`${t.level}: ${J(t,i)}`}function q({largestBlocks:e=[]}){let t=e.map(({startLine:e,endLine:t,name:n,cognitiveComplexity:r})=>`L${e}${t===e?``:`-${t}`}${n?` ${Y(n)}`:``} ${r}`);return t.length>0?`; largest parts ${t.join(`, `)}`:``}function J(e,t){switch(e.kind){case`file`:return`${e.file}: ${t}`;case`function`:return`${X(e)} ${Y(e.name??``)}: ${t}`;case`duplication`:{let n=(e.partners??[]).map(e=>X(e)),r=[...n.slice(0,3),...n.length>3?[`...`]:[]];return`${X(e)}: ${t}${r.length>0?`, also at ${r.join(`, `)}`:``}`}}}function Y(e){return e.replaceAll(/\s*[\n\r]\s*/gu,` `)}function X({file:e,startLine:t,endLine:n}){return`${e}:${t}-${n}`}function Z(e,t,n){C(JSON.stringify({passed:!N(n.violations)&&t.errors.length===0,base:e.base,mergeBase:t.mergeBase,summary:{...G(n.violations),checkedFileCount:t.files.length,checkedFunctionCount:n.checkedFunctionCount},violations:n.violations,errors:t.errors,warnings:t.warnings},void 0,2)+`
 `)}export{M as runCheckCommand,F as scanCheckScope};
 //# sourceMappingURL=checkCommand.js.map

package/dist/checkCommand.js.map

@@ -1 +1 @@
-{"version":3,"file":"checkCommand.js","names":[],"sources":["../src/checkCommand.ts"],"sourcesContent":["import { readFile, realpath, stat } from 'node:fs/promises';\nimport os from 'node:os';\nimport path from 'node:path';\nimport { getErrorMessage, mapConcurrently } from '@willbooster/shared-lib';\nimport { loadConfig, resolveOptions, type CliOptions, type ResolvedOptions } from './cliConfig.js';\nimport type { CrossFileDuplicationMetrics } from './crossFileDuplication.js';\nimport { keepPaths, loadExclusion, loadRepositoryExclusion } from './exclusion.js';\nimport {\n  isIgnored,\n  listChangedFiles,\n  listLineHunks,\n  listRepositoryFiles,\n  resolveMergeBase,\n  resolveRepoRoot,\n  type ChangedFile,\n  type LineHunk,\n} from './git.js';\nimport {\n  addCrossFileDuplication,\n  configSearchDirectory,\n  formatPath,\n  getLanguage,\n  isWithinDirectory,\n  resolveTarget,\n  scanListedFiles,\n  scanTarget,\n  writeStderr,\n  writeStdout,\n  type ScanResult,\n} from './scan.js';\nimport {\n  checkThresholds,\n  type BlockLocation,\n  type CheckedFile,\n  type CodeLineNumbersByFile,\n  type CheckResult,\n  type Violation,\n} from './thresholdCheck.js';\nimport { metricNameOf, resolveLimits, thresholds, type LimitsByLevel } from './thresholds.js';\n\n/** Raw options of the `check` subcommand; every field is undefined unless the flag was passed. */\nexport interface CheckCliOptions extends CliOptions {\n  base?: string;\n}\n\n/** The measured files a check covers and the duplication they take part in. */\nexport interface CheckScope {\n  files: CheckedFile[];\n  crossFileDuplication?: CrossFileDuplicationMetrics;\n  /** Of every measured file, so that a copy outside the files the check covers is sized like one inside. */\n  codeLineNumbersByFile: CodeLineNumbersByFile;\n  /** Measurement failures of files the check covers. */\n  errors: string[];\n  warnings: string[];\n  /** The directory the files' paths are relative to. */\n  root: string;\n  mergeBase?: string;\n}\n\n/** Caps how many partner locations a duplicated block lists so the report stays scannable. */\nconst maxListedPartners = 3;\n\n/**\n * Reports every exceeded threshold of the target or, with `base`, of what the working tree changed\n * under it since the merge-base with that ref. Exit codes: 0 no error-level violations, 1 error-level\n * violations, 2 files the check covers could not be measured.\n */\nexport async function runCheckCommand(\n  target: string,\n  cliOptions: CheckCliOptions,\n  cliLimits: LimitsByLevel\n): Promise<void> {\n  try {\n    const resolvedTarget = resolveTarget(target);\n    const config = await loadConfig(cliOptions.config, await configSearchDirectory(resolvedTarget));\n    const options = resolveOptions(cliOptions, config);\n    const scope = await scanCheckScope(resolvedTarget, cliOptions.base, options);\n    const result = checkThresholds(\n      scope.files,\n      scope.crossFileDuplication,\n      resolveLimits(cliLimits, config.config.thresholds),\n      scope.codeLineNumbersByFile\n    );\n    await endFileViolationsAtLastLine(result.violations, scope.root);\n\n    if (options.json) {\n      printJsonReport(cliOptions, scope, result);\n    } else {\n      printTextReport(cliOptions, scope, result);\n    }\n    if (scope.errors.length > 0) {\n      process.exitCode = 2;\n    } else if (hasErrors(result.violations)) {\n      process.exitCode = 1;\n    }\n  } catch (error) {\n    writeStderr(`Error: ${getErrorMessage(error)}\\n`);\n    process.exitCode = 2;\n  }\n}\n\nfunction hasErrors(violations: readonly Violation[]): boolean {\n  return violations.some(({ level }) => level === 'error');\n}\n\n/**\n * A file's measured line count includes the empty line after a final line terminator, which is\n * not a line of the file, so file-level violations end at the last line that exists.\n */\nasync function endFileViolationsAtLastLine(violations: readonly Violation[], root: string): Promise<void> {\n  for (const violation of violations.filter(({ kind }) => kind === 'file')) {\n    const content = await readFile(path.join(root, violation.file), 'utf8');\n    if (/[\\n\\r]$/u.test(content)) {\n      violation.endLine = Math.max(violation.endLine - 1, 1);\n    }\n  }\n}\n\n/**\n * Inside a git repository, the scope is the repository's git-visible files under the target, and\n * duplication is detected against all of them; `base` only narrows the scope to what changed.\n * Outside a repository, where there is no change to narrow to, the target directory is walked.\n */\nexport async function scanCheckScope(\n  resolvedTarget: string,\n  base: string | undefined,\n  options: ResolvedOptions\n): Promise<CheckScope> {\n  const canonicalTarget = await realpath(resolvedTarget);\n  const targetStat = await stat(canonicalTarget);\n  const targetFile = targetStat.isFile() ? canonicalTarget : undefined;\n  if (targetFile && !getLanguage(targetFile, options, true)) {\n    throw new Error(`${path.basename(targetFile)}: unsupported file type`);\n  }\n  let repoRoot;\n  try {\n    repoRoot = await realpath(await resolveRepoRoot(targetFile ? path.dirname(canonicalTarget) : canonicalTarget));\n  } catch (error) {\n    if (base !== undefined) {\n      throw error;\n    }\n    return await scanDirectoryWalk(canonicalTarget, options);\n  }\n  return await scanRepository({ repoRoot, canonicalTarget, targetFile }, base, options);\n}\n\nasync function scanDirectoryWalk(canonicalTarget: string, options: ResolvedOptions): Promise<CheckScope> {\n  const searchDirectory = await configSearchDirectory(canonicalTarget);\n  const scan = await scanTarget(canonicalTarget, {\n    ...options,\n    loadExclusion: (absolutePaths) => loadExclusion(searchDirectory, options.exclude, absolutePaths),\n  });\n  const crossFileDuplication = measureDuplication(scan, options);\n  return {\n    files: scan.files.map(({ file, metrics }) => ({ file: formatPath(file, scan.displayRoot), metrics })),\n    codeLineNumbersByFile: indexCodeLineNumbers(scan, scan.displayRoot),\n    crossFileDuplication,\n    errors: scan.errors,\n    warnings: scan.warnings,\n    root: scan.displayRoot,\n  };\n}\n\ninterface RepositoryTarget {\n  repoRoot: string;\n  canonicalTarget: string;\n  /** Set when the target is a file, which is measured even when git ignores it or it is excluded. */\n  targetFile?: string;\n}\n\n/**\n * Measures every git-visible file (tracked or untracked non-ignored) of the repository, so that\n * code copied from outside the target is detected as duplication, and keeps the files under the\n * target: all of them or, with `base`, the changed ones with their line hunks.\n */\nasync function scanRepository(\n  { repoRoot, canonicalTarget, targetFile }: RepositoryTarget,\n  base: string | undefined,\n  options: ResolvedOptions\n): Promise<CheckScope> {\n  // Only git-visible files are measured, so a directory git ignores would pass as a check of no\n  // files. A directory git lists but whose sources it ignores is not told apart from a repository\n  // or package whose only sources are ignored build output, which must pass.\n  if (targetFile === undefined && canonicalTarget !== repoRoot && (await isIgnored(repoRoot, canonicalTarget))) {\n    throw new Error(\n      `${formatPath(canonicalTarget, repoRoot)}: git ignores this directory, so no file in it is checked; target a file to check it regardless`\n    );\n  }\n  const repositoryFiles = await listRepositoryFiles(repoRoot);\n  const explicitFiles = new Set(targetFile === undefined ? [] : [targetFile]);\n  const unlistedTarget =\n    targetFile !== undefined && !repositoryFiles.has(formatPath(targetFile, repoRoot))\n      ? formatPath(targetFile, repoRoot)\n      : undefined;\n  if (unlistedTarget !== undefined) {\n    repositoryFiles.add(unlistedTarget);\n  }\n  const exclusion = keepPaths(await loadRepositoryExclusion(repoRoot, repositoryFiles, options.exclude), explicitFiles);\n  // Scan errors start with the file or directory they concern.\n  const targetPath = formatPath(canonicalTarget, repoRoot);\n  const scan = await scanListedFiles(\n    repoRoot,\n    repositoryFiles,\n    { ...options, loadExclusion: () => Promise.resolve(exclusion) },\n    explicitFiles,\n    targetFile === undefined && canonicalTarget !== repoRoot ? targetPath : ''\n  );\n  const crossFileDuplication = measureDuplication(scan, options);\n  const isInTarget = (relativePath: string): boolean =>\n    isWithinDirectory(path.join(repoRoot, relativePath), canonicalTarget);\n  const measuredFiles = scan.files.map(({ file, metrics }) => ({ file: formatPath(file, repoRoot), metrics }));\n  let files = measuredFiles.filter(({ file }) => isInTarget(file));\n  let isCovered = (error: string): boolean =>\n    canonicalTarget === repoRoot || error.startsWith(`${targetPath}/`) || error.startsWith(`${targetPath}:`);\n  let mergeBase;\n  if (base !== undefined) {\n    mergeBase = await resolveMergeBase(repoRoot, base);\n    const changedFiles = await listChangedFiles(repoRoot, mergeBase);\n    // A targeted file git does not list has no diff; all of it counts as added.\n    if (unlistedTarget !== undefined) {\n      changedFiles.push({ status: 'added', headPath: unlistedTarget });\n    }\n    const changesInTarget = changedFiles.filter(({ status, headPath }) => status !== 'deleted' && isInTarget(headPath));\n    files = await attachHunks(measuredFiles, changesInTarget, repoRoot, mergeBase);\n    isCovered = (error) => changesInTarget.some(({ headPath }) => error.startsWith(`${headPath}:`));\n  }\n  return {\n    files,\n    crossFileDuplication,\n    codeLineNumbersByFile: indexCodeLineNumbers(scan, repoRoot),\n    ...partitionErrors(scan, isCovered),\n    root: repoRoot,\n    mergeBase,\n  };\n}\n\nfunction indexCodeLineNumbers(scan: ScanResult, root: string): CodeLineNumbersByFile {\n  return new Map(scan.files.map(({ file, metrics }) => [formatPath(file, root), new Set(metrics.codeLineNumbers)]));\n}\n\n/** Only a failure on a file the check covers leaves it incomplete; the others are warnings. */\nfunction partitionErrors(\n  scan: ScanResult,\n  isCovered: (error: string) => boolean\n): Pick<CheckScope, 'errors' | 'warnings'> {\n  return {\n    errors: scan.errors.filter((error) => isCovered(error)),\n    warnings: [...scan.warnings, ...scan.errors.filter((error) => !isCovered(error))],\n  };\n}\n\n/** The measured changed files with their line hunks against the merge-base. */\nasync function attachHunks(\n  measuredFiles: readonly CheckedFile[],\n  changedFiles: readonly ChangedFile[],\n  repoRoot: string,\n  mergeBase: string\n): Promise<CheckedFile[]> {\n  const measuredByPath = new Map(measuredFiles.map((measured) => [measured.file, measured]));\n  const files = await mapConcurrently(changedFiles, os.availableParallelism() * 2, async (changed) => {\n    const measured = measuredByPath.get(changed.headPath);\n    // A changed file the scan left out (unsupported, excluded, or failed) has nothing to check.\n    if (measured === undefined) {\n      return [];\n    }\n    const { metrics } = measured;\n    const wholeFile: LineHunk = { headStart: 1, headCount: metrics.lines.total };\n    const hunks =\n      changed.basePath === undefined\n        ? [wholeFile]\n        : await listLineHunks(repoRoot, mergeBase, changed.basePath, changed.headPath);\n    if (metrics.language === 'python' && hunks.some(({ headCount }) => headCount === 0)) {\n      anchorDeletionsToCode(hunks, await readFile(path.join(repoRoot, changed.headPath), 'utf8'));\n    }\n    return [{ ...measured, hunks }];\n  });\n  return files.flat();\n}\n\n/**\n * Moves each pure deletion up to the last code line before it. A Python function ends at its last\n * statement, so the blank and comment-only lines left above a deleted tail lie outside the span the\n * deletion shortened; anchored to that statement, the deletion counts as touching the function.\n */\nfunction anchorDeletionsToCode(hunks: LineHunk[], content: string): void {\n  const lines = content.split(/\\r\\n|\\n|\\r/u);\n  for (const hunk of hunks) {\n    if (hunk.headCount > 0) continue;\n    while (hunk.headStart > 0 && /^\\s*(?:#.*)?$/u.test(lines[hunk.headStart - 1] ?? '')) {\n      hunk.headStart -= 1;\n    }\n  }\n}\n\n/** A run-wide failure (a missing target or native addon) leaves nothing to check. */\nfunction measureDuplication(scan: ScanResult, options: ResolvedOptions): CrossFileDuplicationMetrics | undefined {\n  if (scan.fatalError) {\n    throw new Error(scan.fatalError);\n  }\n  addCrossFileDuplication(scan, options);\n  return scan.crossFileDuplication;\n}\n\nfunction printTextReport(cliOptions: CheckCliOptions, scope: CheckScope, result: CheckResult): void {\n  const { violations } = result;\n  const checked = `${scope.files.length} ${cliOptions.base === undefined ? '' : 'changed '}files, ${result.checkedFunctionCount} functions checked${describeBase(cliOptions, scope)}`;\n  if (scope.errors.length > 0) {\n    // Unmeasured files were not checked, so \"no violations\" would be vacuous; never claim a pass.\n    writeStdout(\n      `Check could not complete: ${scope.errors.length} measurement failures (details on stderr); ${describeCounts(violations)} in the measured files (${checked}).\\n`\n    );\n  } else if (violations.length === 0) {\n    writeStdout(`No threshold violations: ${checked}.\\n`);\n  } else {\n    writeStdout(`Threshold violations: ${describeCounts(violations)} (${checked}).\\n`);\n  }\n  for (const violation of violations) {\n    writeStdout(`${formatViolation(violation)}\\n`);\n  }\n\n  const violatedMetrics = new Set(violations.flatMap(({ exceeded }) => exceeded.map(({ metric }) => metric)));\n  const violatedThresholds = thresholds.filter((threshold) => violatedMetrics.has(metricNameOf(threshold)));\n  if (violatedThresholds.length > 0) {\n    writeStdout('\\nHow to fix:\\n');\n    for (const { label, hint } of violatedThresholds) {\n      writeStdout(`- ${label}: ${hint}\\n`);\n    }\n  }\n\n  for (const warning of scope.warnings) {\n    writeStderr(`Warning: ${warning}\\n`);\n  }\n  for (const error of scope.errors) {\n    writeStderr(`Error: ${error}\\n`);\n  }\n}\n\nfunction describeBase(cliOptions: CheckCliOptions, scope: CheckScope): string {\n  return scope.mergeBase === undefined ? '' : `; base ${cliOptions.base}, merge-base ${scope.mergeBase.slice(0, 12)}`;\n}\n\nfunction describeCounts(violations: readonly Violation[]): string {\n  const counts = countViolations(violations);\n  return `${counts.errorViolationCount} errors, ${counts.warningViolationCount} warnings (${counts.functionViolationCount} functions, ${counts.fileViolationCount} files, ${counts.duplicationViolationCount} duplicated blocks)`;\n}\n\nfunction countViolations(violations: readonly Violation[]): {\n  violationCount: number;\n  errorViolationCount: number;\n  warningViolationCount: number;\n  functionViolationCount: number;\n  fileViolationCount: number;\n  duplicationViolationCount: number;\n} {\n  const count = (kind: Violation['kind']): number => violations.filter((violation) => violation.kind === kind).length;\n  const errorViolationCount = violations.filter(({ level }) => level === 'error').length;\n  return {\n    violationCount: violations.length,\n    errorViolationCount,\n    warningViolationCount: violations.length - errorViolationCount,\n    functionViolationCount: count('function'),\n    fileViolationCount: count('file'),\n    duplicationViolationCount: count('duplication'),\n  };\n}\n\n/** One violation as a single line: its level, the location, then every limit it exceeds. */\nfunction formatViolation(violation: Violation): string {\n  const labelByMetric = new Map(thresholds.map((threshold) => [metricNameOf(threshold), threshold.label]));\n  const exceeded = violation.exceeded\n    .map(({ metric, value, level, limit }) => {\n      // Halstead values are fractional; rounding up keeps a violating value above the printed maximum.\n      const roundedValue = Math.ceil(value * 10) / 10;\n      const levelPrefix = level === violation.level ? '' : `${level} `;\n      // A duplicated block violates from its limit on, so the largest allowed code-line count is\n      // the last whole number below it; a limit of 0 allows none.\n      const maxAllowed = violation.kind === 'duplication' ? Math.max(Math.ceil(limit) - 1, 0) : limit;\n      const blocks = metric === 'functionCognitiveComplexity' ? describeLargestBlocks(violation) : '';\n      return `${labelByMetric.get(metric)} ${roundedValue} (${levelPrefix}max ${maxAllowed}${blocks})`;\n    })\n    .join(', ');\n  return `${violation.level}: ${describeViolation(violation, exceeded)}`;\n}\n\nfunction describeLargestBlocks({ largestBlocks = [] }: Violation): string {\n  const blocks = largestBlocks.map(\n    ({ startLine, endLine, name, cognitiveComplexity }) =>\n      `L${startLine}${endLine === startLine ? '' : `-${endLine}`}${name ? ` ${toOneLine(name)}` : ''} ${cognitiveComplexity}`\n  );\n  return blocks.length > 0 ? `; largest parts ${blocks.join(', ')}` : '';\n}\n\nfunction describeViolation(violation: Violation, exceeded: string): string {\n  switch (violation.kind) {\n    case 'file': {\n      return `${violation.file}: ${exceeded}`;\n    }\n    case 'function': {\n      return `${formatLocation(violation)} ${toOneLine(violation.name ?? '')}: ${exceeded}`;\n    }\n    case 'duplication': {\n      const partners = (violation.partners ?? []).map((partner) => formatLocation(partner));\n      const listed = [...partners.slice(0, maxListedPartners), ...(partners.length > maxListedPartners ? ['...'] : [])];\n      return `${formatLocation(violation)}: ${exceeded}${listed.length > 0 ? `, also at ${listed.join(', ')}` : ''}`;\n    }\n  }\n}\n\n/** A computed name can span lines in the source; the report keeps one line per function. */\nfunction toOneLine(name: string): string {\n  return name.replaceAll(/\\s*[\\n\\r]\\s*/gu, ' ');\n}\n\nfunction formatLocation({ file, startLine, endLine }: BlockLocation): string {\n  return `${file}:${startLine}-${endLine}`;\n}\n\nfunction printJsonReport(cliOptions: CheckCliOptions, scope: CheckScope, result: CheckResult): void {\n  writeStdout(\n    JSON.stringify(\n      {\n        passed: !hasErrors(result.violations) && scope.errors.length === 0,\n        base: cliOptions.base,\n        mergeBase: scope.mergeBase,\n        summary: {\n          ...countViolations(result.violations),\n          checkedFileCount: scope.files.length,\n          checkedFunctionCount: result.checkedFunctionCount,\n        },\n        violations: result.violations,\n        errors: scope.errors,\n        warnings: scope.warnings,\n      },\n      undefined,\n      2\n    ) + '\\n'\n  );\n}\n"],"mappings":"y1BAmEA,eAAsB,EACpB,EACA,EACA,EACe,CACf,GAAI,CACF,IAAM,EAAiB,EAAc,CAAM,EACrC,EAAS,MAAM,EAAW,EAAW,OAAQ,MAAM,EAAsB,CAAc,CAAC,EACxF,EAAU,EAAe,EAAY,CAAM,EAC3C,EAAQ,MAAM,EAAe,EAAgB,EAAW,KAAM,CAAO,EACrE,EAAS,EACb,EAAM,MACN,EAAM,qBACN,EAAc,EAAW,EAAO,OAAO,UAAU,EACjD,EAAM,qBACR,EACA,MAAM,EAA4B,EAAO,WAAY,EAAM,IAAI,EAE3D,EAAQ,KACV,EAAgB,EAAY,EAAO,CAAM,EAEzC,EAAgB,EAAY,EAAO,CAAM,EAEvC,EAAM,OAAO,OAAS,EACxB,QAAQ,SAAW,EACV,EAAU,EAAO,UAAU,IACpC,QAAQ,SAAW,EAEvB,OAAS,EAAO,CACd,EAAY,UAAU,EAAgB,CAAK,EAAE,GAAG,EAChD,QAAQ,SAAW,CACrB,CACF,CAEA,SAAS,EAAU,EAA2C,CAC5D,OAAO,EAAW,MAAM,CAAE,WAAY,IAAU,OAAO,CACzD,CAMA,eAAe,EAA4B,EAAkC,EAA6B,CACxG,IAAK,IAAM,KAAa,EAAW,QAAQ,CAAE,UAAW,IAAS,MAAM,EAAG,CACxE,IAAM,EAAU,MAAM,EAAS,EAAK,KAAK,EAAM,EAAU,IAAI,EAAG,MAAM,EAClE,WAAW,KAAK,CAAO,IACzB,EAAU,QAAU,KAAK,IAAI,EAAU,QAAU,EAAG,CAAC,EAEzD,CACF,CAOA,eAAsB,EACpB,EACA,EACA,EACqB,CACrB,IAAM,EAAkB,MAAM,EAAS,CAAc,EAE/C,GAAa,MADM,EAAK,CAAe,EAAA,CACf,OAAO,EAAI,EAAkB,IAAA,GAC3D,GAAI,GAAc,CAAC,EAAY,EAAY,EAAS,EAAI,EACtD,MAAU,MAAM,GAAG,EAAK,SAAS,CAAU,EAAE,wBAAwB,EAEvE,IAAI,EACJ,GAAI,CACF,EAAW,MAAM,EAAS,MAAM,EAAgB,EAAa,EAAK,QAAQ,CAAe,EAAI,CAAe,CAAC,CAC/G,OAAS,EAAO,CACd,GAAI,IAAS,IAAA,GACX,MAAM,EAER,OAAO,MAAM,EAAkB,EAAiB,CAAO,CACzD,CACA,OAAO,MAAM,EAAe,CAAE,WAAU,kBAAiB,YAAW,EAAG,EAAM,CAAO,CACtF,CAEA,eAAe,EAAkB,EAAyB,EAA+C,CACvG,IAAM,EAAkB,MAAM,EAAsB,CAAe,EAC7D,EAAO,MAAM,EAAW,EAAiB,CAC7C,GAAG,EACH,cAAgB,GAAkB,EAAc,EAAiB,EAAQ,QAAS,CAAa,CACjG,CAAC,EACK,EAAuB,EAAmB,EAAM,CAAO,EAC7D,MAAO,CACL,MAAO,EAAK,MAAM,KAAK,CAAE,OAAM,cAAe,CAAE,KAAM,EAAW,EAAM,EAAK,WAAW,EAAG,SAAQ,EAAE,EACpG,sBAAuB,EAAqB,EAAM,EAAK,WAAW,EAClE,uBACA,OAAQ,EAAK,OACb,SAAU,EAAK,SACf,KAAM,EAAK,WACb,CACF,CAcA,eAAe,EACb,CAAE,WAAU,kBAAiB,cAC7B,EACA,EACqB,CAIrB,GAAI,IAAe,IAAA,IAAa,IAAoB,GAAa,MAAM,EAAU,EAAU,CAAe,EACxG,MAAU,MACR,GAAG,EAAW,EAAiB,CAAQ,EAAE,gGAC3C,EAEF,IAAM,EAAkB,MAAM,EAAoB,CAAQ,EACpD,EAAgB,IAAI,IAAI,IAAe,IAAA,GAAY,CAAC,EAAI,CAAC,CAAU,CAAC,EACpE,EACJ,IAAe,IAAA,IAAa,CAAC,EAAgB,IAAI,EAAW,EAAY,CAAQ,CAAC,EAC7E,EAAW,EAAY,CAAQ,EAC/B,IAAA,GACF,IAAmB,IAAA,IACrB,EAAgB,IAAI,CAAc,EAEpC,IAAM,EAAY,EAAU,MAAM,EAAwB,EAAU,EAAiB,EAAQ,OAAO,EAAG,CAAa,EAE9G,EAAa,EAAW,EAAiB,CAAQ,EACjD,EAAO,MAAM,EACjB,EACA,EACA,CAAE,GAAG,EAAS,kBAAqB,QAAQ,QAAQ,CAAS,CAAE,EAC9D,EACA,IAAe,IAAA,IAAa,IAAoB,EAAW,EAAa,EAC1E,EACM,EAAuB,EAAmB,EAAM,CAAO,EACvD,EAAc,GAClB,EAAkB,EAAK,KAAK,EAAU,CAAY,EAAG,CAAe,EAChE,EAAgB,EAAK,MAAM,KAAK,CAAE,OAAM,cAAe,CAAE,KAAM,EAAW,EAAM,CAAQ,EAAG,SAAQ,EAAE,EACvG,EAAQ,EAAc,QAAQ,CAAE,UAAW,EAAW,CAAI,CAAC,EAC3D,EAAa,GACf,IAAoB,GAAY,EAAM,WAAW,GAAG,EAAW,EAAE,GAAK,EAAM,WAAW,GAAG,EAAW,EAAE,EACrG,EACJ,GAAI,IAAS,IAAA,GAAW,CACtB,EAAY,MAAM,EAAiB,EAAU,CAAI,EACjD,IAAM,EAAe,MAAM,EAAiB,EAAU,CAAS,EAE3D,IAAmB,IAAA,IACrB,EAAa,KAAK,CAAE,OAAQ,QAAS,SAAU,CAAe,CAAC,EAEjE,IAAM,EAAkB,EAAa,QAAQ,CAAE,SAAQ,cAAe,IAAW,WAAa,EAAW,CAAQ,CAAC,EAClH,EAAQ,MAAM,EAAY,EAAe,EAAiB,EAAU,CAAS,EAC7E,EAAa,GAAU,EAAgB,MAAM,CAAE,cAAe,EAAM,WAAW,GAAG,EAAS,EAAE,CAAC,CAChG,CACA,MAAO,CACL,QACA,uBACA,sBAAuB,EAAqB,EAAM,CAAQ,EAC1D,GAAG,EAAgB,EAAM,CAAS,EAClC,KAAM,EACN,WACF,CACF,CAEA,SAAS,EAAqB,EAAkB,EAAqC,CACnF,OAAO,IAAI,IAAI,EAAK,MAAM,KAAK,CAAE,OAAM,aAAc,CAAC,EAAW,EAAM,CAAI,EAAG,IAAI,IAAI,EAAQ,eAAe,CAAC,CAAC,CAAC,CAClH,CAGA,SAAS,EACP,EACA,EACyC,CACzC,MAAO,CACL,OAAQ,EAAK,OAAO,OAAQ,GAAU,EAAU,CAAK,CAAC,EACtD,SAAU,CAAC,GAAG,EAAK,SAAU,GAAG,EAAK,OAAO,OAAQ,GAAU,CAAC,EAAU,CAAK,CAAC,CAAC,CAClF,CACF,CAGA,eAAe,EACb,EACA,EACA,EACA,EACwB,CACxB,IAAM,EAAiB,IAAI,IAAI,EAAc,IAAK,GAAa,CAAC,EAAS,KAAM,CAAQ,CAAC,CAAC,EAkBzF,OAAO,MAjBa,EAAgB,EAAc,EAAG,qBAAqB,EAAI,EAAG,KAAO,IAAY,CAClG,IAAM,EAAW,EAAe,IAAI,EAAQ,QAAQ,EAEpD,GAAI,IAAa,IAAA,GACf,MAAO,CAAC,EAEV,GAAM,CAAE,WAAY,EACd,EAAsB,CAAE,UAAW,EAAG,UAAW,EAAQ,MAAM,KAAM,EACrE,EACJ,EAAQ,WAAa,IAAA,GACjB,CAAC,CAAS,EACV,MAAM,EAAc,EAAU,EAAW,EAAQ,SAAU,EAAQ,QAAQ,EAIjF,OAHI,EAAQ,WAAa,UAAY,EAAM,MAAM,CAAE,eAAgB,IAAc,CAAC,GAChF,EAAsB,EAAO,MAAM,EAAS,EAAK,KAAK,EAAU,EAAQ,QAAQ,EAAG,MAAM,CAAC,EAErF,CAAC,CAAE,GAAG,EAAU,OAAM,CAAC,CAChC,CAAC,EAAA,CACY,KAAK,CACpB,CAOA,SAAS,EAAsB,EAAmB,EAAuB,CACvE,IAAM,EAAQ,EAAQ,MAAM,aAAa,EACzC,IAAK,IAAM,KAAQ,EACb,OAAK,UAAY,GACrB,KAAO,EAAK,UAAY,GAAK,iBAAiB,KAAK,EAAM,EAAK,UAAY,IAAM,EAAE,GAChF,IAAK,SAGX,CAGA,SAAS,EAAmB,EAAkB,EAAmE,CAC/G,GAAI,EAAK,WACP,MAAU,MAAM,EAAK,UAAU,EAGjC,OADA,EAAwB,EAAM,CAAO,EAC9B,EAAK,oBACd,CAEA,SAAS,EAAgB,EAA6B,EAAmB,EAA2B,CAClG,GAAM,CAAE,cAAe,EACjB,EAAU,GAAG,EAAM,MAAM,OAAO,GAAG,EAAW,OAAS,IAAA,GAAY,GAAK,WAAW,SAAS,EAAO,qBAAqB,oBAAoB,EAAa,EAAY,CAAK,IAC5K,EAAM,OAAO,OAAS,EAExB,EACE,6BAA6B,EAAM,OAAO,OAAO,6CAA6C,EAAe,CAAU,EAAE,0BAA0B,EAAQ,KAC7J,EACS,EAAW,SAAW,EAC/B,EAAY,4BAA4B,EAAQ,IAAI,EAEpD,EAAY,yBAAyB,EAAe,CAAU,EAAE,IAAI,EAAQ,KAAK,EAEnF,IAAK,IAAM,KAAa,EACtB,EAAY,GAAG,EAAgB,CAAS,EAAE,GAAG,EAG/C,IAAM,EAAkB,IAAI,IAAI,EAAW,SAAS,CAAE,cAAe,EAAS,KAAK,CAAE,YAAa,CAAM,CAAC,CAAC,EACpG,EAAqB,EAAW,OAAQ,GAAc,EAAgB,IAAI,EAAa,CAAS,CAAC,CAAC,EACxG,GAAI,EAAmB,OAAS,EAAG,CACjC,EAAY;;CAAiB,EAC7B,IAAK,GAAM,CAAE,QAAO,UAAU,EAC5B,EAAY,KAAK,EAAM,IAAI,EAAK,GAAG,CAEvC,CAEA,IAAK,IAAM,KAAW,EAAM,SAC1B,EAAY,YAAY,EAAQ,GAAG,EAErC,IAAK,IAAM,KAAS,EAAM,OACxB,EAAY,UAAU,EAAM,GAAG,CAEnC,CAEA,SAAS,EAAa,EAA6B,EAA2B,CAC5E,OAAO,EAAM,YAAc,IAAA,GAAY,GAAK,UAAU,EAAW,KAAK,eAAe,EAAM,UAAU,MAAM,EAAG,EAAE,GAClH,CAEA,SAAS,EAAe,EAA0C,CAChE,IAAM,EAAS,EAAgB,CAAU,EACzC,MAAO,GAAG,EAAO,oBAAoB,WAAW,EAAO,sBAAsB,aAAa,EAAO,uBAAuB,cAAc,EAAO,mBAAmB,UAAU,EAAO,0BAA0B,oBAC7M,CAEA,SAAS,EAAgB,EAOvB,CACA,IAAM,EAAS,GAAoC,EAAW,OAAQ,GAAc,EAAU,OAAS,CAAI,CAAC,CAAC,OACvG,EAAsB,EAAW,QAAQ,CAAE,WAAY,IAAU,OAAO,CAAC,CAAC,OAChF,MAAO,CACL,eAAgB,EAAW,OAC3B,sBACA,sBAAuB,EAAW,OAAS,EAC3C,uBAAwB,EAAM,UAAU,EACxC,mBAAoB,EAAM,MAAM,EAChC,0BAA2B,EAAM,aAAa,CAChD,CACF,CAGA,SAAS,EAAgB,EAA8B,CACrD,IAAM,EAAgB,IAAI,IAAI,EAAW,IAAK,GAAc,CAAC,EAAa,CAAS,EAAG,EAAU,KAAK,CAAC,CAAC,EACjG,EAAW,EAAU,SACxB,KAAK,CAAE,SAAQ,QAAO,QAAO,WAAY,CAExC,IAAM,EAAe,KAAK,KAAK,EAAQ,EAAE,EAAI,GACvC,EAAc,IAAU,EAAU,MAAQ,GAAK,GAAG,EAAM,GAGxD,EAAa,EAAU,OAAS,cAAgB,KAAK,IAAI,KAAK,KAAK,CAAK,EAAI,EAAG,CAAC,EAAI,EACpF,EAAS,IAAW,8BAAgC,EAAsB,CAAS,EAAI,GAC7F,MAAO,GAAG,EAAc,IAAI,CAAM,EAAE,GAAG,EAAa,IAAI,EAAY,MAAM,IAAa,EAAO,EAChG,CAAC,CAAC,CACD,KAAK,IAAI,EACZ,MAAO,GAAG,EAAU,MAAM,IAAI,EAAkB,EAAW,CAAQ,GACrE,CAEA,SAAS,EAAsB,CAAE,gBAAgB,CAAC,GAAwB,CACxE,IAAM,EAAS,EAAc,KAC1B,CAAE,YAAW,UAAS,OAAM,yBAC3B,IAAI,IAAY,IAAY,EAAY,GAAK,IAAI,MAAY,EAAO,IAAI,EAAU,CAAI,IAAM,GAAG,GAAG,GACtG,EACA,OAAO,EAAO,OAAS,EAAI,mBAAmB,EAAO,KAAK,IAAI,IAAM,EACtE,CAEA,SAAS,EAAkB,EAAsB,EAA0B,CACzE,OAAQ,EAAU,KAAlB,CACE,IAAK,OACH,MAAO,GAAG,EAAU,KAAK,IAAI,IAE/B,IAAK,WACH,MAAO,GAAG,EAAe,CAAS,EAAE,GAAG,EAAU,EAAU,MAAQ,EAAE,EAAE,IAAI,IAE7E,IAAK,cAAe,CAClB,IAAM,GAAY,EAAU,UAAY,CAAC,EAAA,CAAG,IAAK,GAAY,EAAe,CAAO,CAAC,EAC9E,EAAS,CAAC,GAAG,EAAS,MAAM,EAAG,CAAiB,EAAG,GAAI,EAAS,OAAS,EAAoB,CAAC,KAAK,EAAI,CAAC,CAAE,EAChH,MAAO,GAAG,EAAe,CAAS,EAAE,IAAI,IAAW,EAAO,OAAS,EAAI,aAAa,EAAO,KAAK,IAAI,IAAM,IAC5G,CACF,CACF,CAGA,SAAS,EAAU,EAAsB,CACvC,OAAO,EAAK,WAAW,iBAAkB,GAAG,CAC9C,CAEA,SAAS,EAAe,CAAE,OAAM,YAAW,WAAkC,CAC3E,MAAO,GAAG,EAAK,GAAG,EAAU,GAAG,GACjC,CAEA,SAAS,EAAgB,EAA6B,EAAmB,EAA2B,CAClG,EACE,KAAK,UACH,CACE,OAAQ,CAAC,EAAU,EAAO,UAAU,GAAK,EAAM,OAAO,SAAW,EACjE,KAAM,EAAW,KACjB,UAAW,EAAM,UACjB,QAAS,CACP,GAAG,EAAgB,EAAO,UAAU,EACpC,iBAAkB,EAAM,MAAM,OAC9B,qBAAsB,EAAO,oBAC/B,EACA,WAAY,EAAO,WACnB,OAAQ,EAAM,OACd,SAAU,EAAM,QAClB,EACA,IAAA,GACA,CACF,EAAI;CACN,CACF"}
+{"version":3,"file":"checkCommand.js","names":[],"sources":["../src/checkCommand.ts"],"sourcesContent":["import { readFile, realpath, stat } from 'node:fs/promises';\nimport os from 'node:os';\nimport path from 'node:path';\nimport { getErrorMessage, mapConcurrently } from '@willbooster/shared-lib';\nimport { loadConfig, resolveOptions, type CliOptions, type ResolvedOptions } from './cliConfig.js';\nimport { keepPaths, loadExclusion, loadRepositoryExclusion } from './exclusion.js';\nimport {\n  isIgnored,\n  listChangedFiles,\n  listLineHunks,\n  listRepositoryFiles,\n  resolveMergeBase,\n  resolveRepoRoot,\n  type ChangedFile,\n  type LineHunk,\n} from './git.js';\nimport {\n  addCrossFileDuplication,\n  configSearchDirectory,\n  formatPath,\n  getLanguage,\n  isWithinDirectory,\n  resolveTarget,\n  scanListedFiles,\n  scanTarget,\n  writeStderr,\n  writeStdout,\n  type ScanResult,\n} from './scan.js';\nimport {\n  checkThresholds,\n  type BlockLocation,\n  type CheckedFile,\n  type CheckedDuplication,\n  type CheckResult,\n  type Violation,\n} from './thresholdCheck.js';\nimport { metricNameOf, resolveLimits, thresholds, type LimitsByLevel } from './thresholds.js';\n\n/** Raw options of the `check` subcommand; every field is undefined unless the flag was passed. */\nexport interface CheckCliOptions extends CliOptions {\n  base?: string;\n}\n\n/** The measured files a check covers and the duplication they take part in. */\nexport interface CheckScope {\n  files: CheckedFile[];\n  duplication: CheckedDuplication;\n  /** Measurement failures of files the check covers. */\n  errors: string[];\n  warnings: string[];\n  /** The directory the files' paths are relative to. */\n  root: string;\n  mergeBase?: string;\n}\n\n/** Caps how many partner locations a duplicated block lists so the report stays scannable. */\nconst maxListedPartners = 3;\n\n/**\n * Reports every exceeded threshold of the target or, with `base`, of what the working tree changed\n * under it since the merge-base with that ref. Exit codes: 0 no error-level violations, 1 error-level\n * violations, 2 files the check covers could not be measured.\n */\nexport async function runCheckCommand(\n  target: string,\n  cliOptions: CheckCliOptions,\n  cliLimits: LimitsByLevel\n): Promise<void> {\n  try {\n    const resolvedTarget = resolveTarget(target);\n    const config = await loadConfig(cliOptions.config, await configSearchDirectory(resolvedTarget));\n    const options = resolveOptions(cliOptions, config);\n    const scope = await scanCheckScope(resolvedTarget, cliOptions.base, options);\n    const result = checkThresholds(scope.files, scope.duplication, resolveLimits(cliLimits, config.config.thresholds));\n    await endFileViolationsAtLastLine(result.violations, scope.root);\n\n    if (options.json) {\n      printJsonReport(cliOptions, scope, result);\n    } else {\n      printTextReport(cliOptions, scope, result);\n    }\n    if (scope.errors.length > 0) {\n      process.exitCode = 2;\n    } else if (hasErrors(result.violations)) {\n      process.exitCode = 1;\n    }\n  } catch (error) {\n    writeStderr(`Error: ${getErrorMessage(error)}\\n`);\n    process.exitCode = 2;\n  }\n}\n\nfunction hasErrors(violations: readonly Violation[]): boolean {\n  return violations.some(({ level }) => level === 'error');\n}\n\n/**\n * A file's measured line count includes the empty line after a final line terminator, which is\n * not a line of the file, so file-level violations end at the last line that exists.\n */\nasync function endFileViolationsAtLastLine(violations: readonly Violation[], root: string): Promise<void> {\n  for (const violation of violations.filter(({ kind }) => kind === 'file')) {\n    const content = await readFile(path.join(root, violation.file), 'utf8');\n    if (/[\\n\\r]$/u.test(content)) {\n      violation.endLine = Math.max(violation.endLine - 1, 1);\n    }\n  }\n}\n\n/**\n * Inside a git repository, the scope is the repository's git-visible files under the target, and\n * duplication is detected against all of them; `base` only narrows the scope to what changed.\n * Outside a repository, where there is no change to narrow to, the target directory is walked.\n */\nexport async function scanCheckScope(\n  resolvedTarget: string,\n  base: string | undefined,\n  options: ResolvedOptions\n): Promise<CheckScope> {\n  const canonicalTarget = await realpath(resolvedTarget);\n  const targetStat = await stat(canonicalTarget);\n  const targetFile = targetStat.isFile() ? canonicalTarget : undefined;\n  if (targetFile && !getLanguage(targetFile, options, true)) {\n    throw new Error(`${path.basename(targetFile)}: unsupported file type`);\n  }\n  let repoRoot;\n  try {\n    repoRoot = await realpath(await resolveRepoRoot(targetFile ? path.dirname(canonicalTarget) : canonicalTarget));\n  } catch (error) {\n    if (base !== undefined) {\n      throw error;\n    }\n    return await scanDirectoryWalk(canonicalTarget, options);\n  }\n  return await scanRepository({ repoRoot, canonicalTarget, targetFile }, base, options);\n}\n\nasync function scanDirectoryWalk(canonicalTarget: string, options: ResolvedOptions): Promise<CheckScope> {\n  const searchDirectory = await configSearchDirectory(canonicalTarget);\n  const scan = await scanTarget(canonicalTarget, {\n    ...options,\n    loadExclusion: (absolutePaths) => loadExclusion(searchDirectory, options.exclude, absolutePaths),\n  });\n  return {\n    files: scan.files.map(({ file, metrics }) => ({ file: formatPath(file, scan.displayRoot), metrics })),\n    duplication: measureDuplication(scan, options),\n    errors: scan.errors,\n    warnings: scan.warnings,\n    root: scan.displayRoot,\n  };\n}\n\ninterface RepositoryTarget {\n  repoRoot: string;\n  canonicalTarget: string;\n  /** Set when the target is a file, which is measured even when git ignores it or it is excluded. */\n  targetFile?: string;\n}\n\n/**\n * Measures every git-visible file (tracked or untracked non-ignored) of the repository, so that\n * code copied from outside the target is detected as duplication, and keeps the files under the\n * target: all of them or, with `base`, the changed ones with their line hunks.\n */\nasync function scanRepository(\n  { repoRoot, canonicalTarget, targetFile }: RepositoryTarget,\n  base: string | undefined,\n  options: ResolvedOptions\n): Promise<CheckScope> {\n  // Only git-visible files are measured, so a directory git ignores would pass as a check of no\n  // files. A directory git lists but whose sources it ignores is not told apart from a repository\n  // or package whose only sources are ignored build output, which must pass.\n  if (targetFile === undefined && canonicalTarget !== repoRoot && (await isIgnored(repoRoot, canonicalTarget))) {\n    throw new Error(\n      `${formatPath(canonicalTarget, repoRoot)}: git ignores this directory, so no file in it is checked; target a file to check it regardless`\n    );\n  }\n  const repositoryFiles = await listRepositoryFiles(repoRoot);\n  const explicitFiles = new Set(targetFile === undefined ? [] : [targetFile]);\n  const unlistedTarget =\n    targetFile !== undefined && !repositoryFiles.has(formatPath(targetFile, repoRoot))\n      ? formatPath(targetFile, repoRoot)\n      : undefined;\n  if (unlistedTarget !== undefined) {\n    repositoryFiles.add(unlistedTarget);\n  }\n  const exclusion = keepPaths(await loadRepositoryExclusion(repoRoot, repositoryFiles, options.exclude), explicitFiles);\n  // Scan errors start with the file or directory they concern.\n  const targetPath = formatPath(canonicalTarget, repoRoot);\n  const scan = await scanListedFiles(\n    repoRoot,\n    repositoryFiles,\n    { ...options, loadExclusion: () => Promise.resolve(exclusion) },\n    explicitFiles,\n    targetFile === undefined && canonicalTarget !== repoRoot ? targetPath : ''\n  );\n  const duplication = measureDuplication(scan, options);\n  const isInTarget = (relativePath: string): boolean =>\n    isWithinDirectory(path.join(repoRoot, relativePath), canonicalTarget);\n  const measuredFiles = scan.files.map(({ file, metrics }) => ({ file: formatPath(file, repoRoot), metrics }));\n  let files = measuredFiles.filter(({ file }) => isInTarget(file));\n  let isCovered = (error: string): boolean =>\n    canonicalTarget === repoRoot || error.startsWith(`${targetPath}/`) || error.startsWith(`${targetPath}:`);\n  let mergeBase;\n  if (base !== undefined) {\n    mergeBase = await resolveMergeBase(repoRoot, base);\n    const changedFiles = await listChangedFiles(repoRoot, mergeBase);\n    // A targeted file git does not list has no diff; all of it counts as added.\n    if (unlistedTarget !== undefined) {\n      changedFiles.push({ status: 'added', headPath: unlistedTarget });\n    }\n    const changesInTarget = changedFiles.filter(({ status, headPath }) => status !== 'deleted' && isInTarget(headPath));\n    files = await attachHunks(measuredFiles, changesInTarget, repoRoot, mergeBase);\n    isCovered = (error) => changesInTarget.some(({ headPath }) => error.startsWith(`${headPath}:`));\n  }\n  return {\n    files,\n    duplication,\n    ...partitionErrors(scan, isCovered),\n    root: repoRoot,\n    mergeBase,\n  };\n}\n\n/** Only a failure on a file the check covers leaves it incomplete; the others are warnings. */\nfunction partitionErrors(\n  scan: ScanResult,\n  isCovered: (error: string) => boolean\n): Pick<CheckScope, 'errors' | 'warnings'> {\n  return {\n    errors: scan.errors.filter((error) => isCovered(error)),\n    warnings: [...scan.warnings, ...scan.errors.filter((error) => !isCovered(error))],\n  };\n}\n\n/** The measured changed files with their line hunks against the merge-base. */\nasync function attachHunks(\n  measuredFiles: readonly CheckedFile[],\n  changedFiles: readonly ChangedFile[],\n  repoRoot: string,\n  mergeBase: string\n): Promise<CheckedFile[]> {\n  const measuredByPath = new Map(measuredFiles.map((measured) => [measured.file, measured]));\n  const files = await mapConcurrently(changedFiles, os.availableParallelism() * 2, async (changed) => {\n    const measured = measuredByPath.get(changed.headPath);\n    // A changed file the scan left out (unsupported, excluded, or failed) has nothing to check.\n    if (measured === undefined) {\n      return [];\n    }\n    const { metrics } = measured;\n    const wholeFile: LineHunk = { headStart: 1, headCount: metrics.lines.total };\n    const hunks =\n      changed.basePath === undefined\n        ? [wholeFile]\n        : await listLineHunks(repoRoot, mergeBase, changed.basePath, changed.headPath);\n    if (metrics.language === 'python' && hunks.some(({ headCount }) => headCount === 0)) {\n      anchorDeletionsToCode(hunks, await readFile(path.join(repoRoot, changed.headPath), 'utf8'));\n    }\n    return [{ ...measured, hunks }];\n  });\n  return files.flat();\n}\n\n/**\n * Moves each pure deletion up to the last code line before it. A Python function ends at its last\n * statement, so the blank and comment-only lines left above a deleted tail lie outside the span the\n * deletion shortened; anchored to that statement, the deletion counts as touching the function.\n */\nfunction anchorDeletionsToCode(hunks: LineHunk[], content: string): void {\n  const lines = content.split(/\\r\\n|\\n|\\r/u);\n  for (const hunk of hunks) {\n    if (hunk.headCount > 0) continue;\n    while (hunk.headStart > 0 && /^\\s*(?:#.*)?$/u.test(lines[hunk.headStart - 1] ?? '')) {\n      hunk.headStart -= 1;\n    }\n  }\n}\n\n/** A run-wide failure (a missing target or native addon) leaves nothing to check. */\nfunction measureDuplication(scan: ScanResult, options: ResolvedOptions): CheckedDuplication {\n  if (scan.fatalError) {\n    throw new Error(scan.fatalError);\n  }\n  addCrossFileDuplication(scan, options);\n  return {\n    crossFile: scan.crossFileDuplication,\n    minSimilarityPercent: options.duplication.minSimilarityPercent,\n  };\n}\n\nfunction printTextReport(cliOptions: CheckCliOptions, scope: CheckScope, result: CheckResult): void {\n  const { violations } = result;\n  const checked = `${scope.files.length} ${cliOptions.base === undefined ? '' : 'changed '}files, ${result.checkedFunctionCount} functions checked${describeBase(cliOptions, scope)}`;\n  if (scope.errors.length > 0) {\n    // Unmeasured files were not checked, so \"no violations\" would be vacuous; never claim a pass.\n    writeStdout(\n      `Check could not complete: ${scope.errors.length} measurement failures (details on stderr); ${describeCounts(violations)} in the measured files (${checked}).\\n`\n    );\n  } else if (violations.length === 0) {\n    writeStdout(`No threshold violations: ${checked}.\\n`);\n  } else {\n    writeStdout(`Threshold violations: ${describeCounts(violations)} (${checked}).\\n`);\n  }\n  for (const violation of violations) {\n    writeStdout(`${formatViolation(violation)}\\n`);\n  }\n\n  const violatedMetrics = new Set(violations.flatMap(({ exceeded }) => exceeded.map(({ metric }) => metric)));\n  const violatedThresholds = thresholds.filter((threshold) => violatedMetrics.has(metricNameOf(threshold)));\n  if (violatedThresholds.length > 0) {\n    writeStdout('\\nHow to fix:\\n');\n    for (const { label, hint } of violatedThresholds) {\n      writeStdout(`- ${label}: ${hint}\\n`);\n    }\n  }\n\n  for (const warning of scope.warnings) {\n    writeStderr(`Warning: ${warning}\\n`);\n  }\n  for (const error of scope.errors) {\n    writeStderr(`Error: ${error}\\n`);\n  }\n}\n\nfunction describeBase(cliOptions: CheckCliOptions, scope: CheckScope): string {\n  return scope.mergeBase === undefined ? '' : `; base ${cliOptions.base}, merge-base ${scope.mergeBase.slice(0, 12)}`;\n}\n\nfunction describeCounts(violations: readonly Violation[]): string {\n  const counts = countViolations(violations);\n  return `${counts.errorViolationCount} errors, ${counts.warningViolationCount} warnings (${counts.functionViolationCount} functions, ${counts.fileViolationCount} files, ${counts.duplicationViolationCount} duplicated blocks)`;\n}\n\nfunction countViolations(violations: readonly Violation[]): {\n  violationCount: number;\n  errorViolationCount: number;\n  warningViolationCount: number;\n  functionViolationCount: number;\n  fileViolationCount: number;\n  duplicationViolationCount: number;\n} {\n  const count = (kind: Violation['kind']): number => violations.filter((violation) => violation.kind === kind).length;\n  const errorViolationCount = violations.filter(({ level }) => level === 'error').length;\n  return {\n    violationCount: violations.length,\n    errorViolationCount,\n    warningViolationCount: violations.length - errorViolationCount,\n    functionViolationCount: count('function'),\n    fileViolationCount: count('file'),\n    duplicationViolationCount: count('duplication'),\n  };\n}\n\n/** One violation as a single line: its level, the location, then every limit it exceeds. */\nfunction formatViolation(violation: Violation): string {\n  const labelByMetric = new Map(thresholds.map((threshold) => [metricNameOf(threshold), threshold.label]));\n  const exceeded = violation.exceeded\n    .map(({ metric, value, level, limit }) => {\n      // Halstead values are fractional; rounding up keeps a violating value above the printed maximum.\n      const roundedValue = Math.ceil(value * 10) / 10;\n      const levelPrefix = level === violation.level ? '' : `${level} `;\n      // A duplicated block violates from its limit on, so the largest allowed code-line count is\n      // the last whole number below it; a limit of 0 allows none.\n      const maxAllowed = violation.kind === 'duplication' ? Math.max(Math.ceil(limit) - 1, 0) : limit;\n      const blocks = metric === 'functionCognitiveComplexity' ? describeLargestBlocks(violation) : '';\n      return `${labelByMetric.get(metric)} ${roundedValue} (${levelPrefix}max ${maxAllowed}${blocks})`;\n    })\n    .join(', ');\n  return `${violation.level}: ${describeViolation(violation, exceeded)}`;\n}\n\nfunction describeLargestBlocks({ largestBlocks = [] }: Violation): string {\n  const blocks = largestBlocks.map(\n    ({ startLine, endLine, name, cognitiveComplexity }) =>\n      `L${startLine}${endLine === startLine ? '' : `-${endLine}`}${name ? ` ${toOneLine(name)}` : ''} ${cognitiveComplexity}`\n  );\n  return blocks.length > 0 ? `; largest parts ${blocks.join(', ')}` : '';\n}\n\nfunction describeViolation(violation: Violation, exceeded: string): string {\n  switch (violation.kind) {\n    case 'file': {\n      return `${violation.file}: ${exceeded}`;\n    }\n    case 'function': {\n      return `${formatLocation(violation)} ${toOneLine(violation.name ?? '')}: ${exceeded}`;\n    }\n    case 'duplication': {\n      const partners = (violation.partners ?? []).map((partner) => formatLocation(partner));\n      const listed = [...partners.slice(0, maxListedPartners), ...(partners.length > maxListedPartners ? ['...'] : [])];\n      return `${formatLocation(violation)}: ${exceeded}${listed.length > 0 ? `, also at ${listed.join(', ')}` : ''}`;\n    }\n  }\n}\n\n/** A computed name can span lines in the source; the report keeps one line per function. */\nfunction toOneLine(name: string): string {\n  return name.replaceAll(/\\s*[\\n\\r]\\s*/gu, ' ');\n}\n\nfunction formatLocation({ file, startLine, endLine }: BlockLocation): string {\n  return `${file}:${startLine}-${endLine}`;\n}\n\nfunction printJsonReport(cliOptions: CheckCliOptions, scope: CheckScope, result: CheckResult): void {\n  writeStdout(\n    JSON.stringify(\n      {\n        passed: !hasErrors(result.violations) && scope.errors.length === 0,\n        base: cliOptions.base,\n        mergeBase: scope.mergeBase,\n        summary: {\n          ...countViolations(result.violations),\n          checkedFileCount: scope.files.length,\n          checkedFunctionCount: result.checkedFunctionCount,\n        },\n        violations: result.violations,\n        errors: scope.errors,\n        warnings: scope.warnings,\n      },\n      undefined,\n      2\n    ) + '\\n'\n  );\n}\n"],"mappings":"y1BAgEA,eAAsB,EACpB,EACA,EACA,EACe,CACf,GAAI,CACF,IAAM,EAAiB,EAAc,CAAM,EACrC,EAAS,MAAM,EAAW,EAAW,OAAQ,MAAM,EAAsB,CAAc,CAAC,EACxF,EAAU,EAAe,EAAY,CAAM,EAC3C,EAAQ,MAAM,EAAe,EAAgB,EAAW,KAAM,CAAO,EACrE,EAAS,EAAgB,EAAM,MAAO,EAAM,YAAa,EAAc,EAAW,EAAO,OAAO,UAAU,CAAC,EACjH,MAAM,EAA4B,EAAO,WAAY,EAAM,IAAI,EAE3D,EAAQ,KACV,EAAgB,EAAY,EAAO,CAAM,EAEzC,EAAgB,EAAY,EAAO,CAAM,EAEvC,EAAM,OAAO,OAAS,EACxB,QAAQ,SAAW,EACV,EAAU,EAAO,UAAU,IACpC,QAAQ,SAAW,EAEvB,OAAS,EAAO,CACd,EAAY,UAAU,EAAgB,CAAK,EAAE,GAAG,EAChD,QAAQ,SAAW,CACrB,CACF,CAEA,SAAS,EAAU,EAA2C,CAC5D,OAAO,EAAW,MAAM,CAAE,WAAY,IAAU,OAAO,CACzD,CAMA,eAAe,EAA4B,EAAkC,EAA6B,CACxG,IAAK,IAAM,KAAa,EAAW,QAAQ,CAAE,UAAW,IAAS,MAAM,EAAG,CACxE,IAAM,EAAU,MAAM,EAAS,EAAK,KAAK,EAAM,EAAU,IAAI,EAAG,MAAM,EAClE,WAAW,KAAK,CAAO,IACzB,EAAU,QAAU,KAAK,IAAI,EAAU,QAAU,EAAG,CAAC,EAEzD,CACF,CAOA,eAAsB,EACpB,EACA,EACA,EACqB,CACrB,IAAM,EAAkB,MAAM,EAAS,CAAc,EAE/C,GAAa,MADM,EAAK,CAAe,EAAA,CACf,OAAO,EAAI,EAAkB,IAAA,GAC3D,GAAI,GAAc,CAAC,EAAY,EAAY,EAAS,EAAI,EACtD,MAAU,MAAM,GAAG,EAAK,SAAS,CAAU,EAAE,wBAAwB,EAEvE,IAAI,EACJ,GAAI,CACF,EAAW,MAAM,EAAS,MAAM,EAAgB,EAAa,EAAK,QAAQ,CAAe,EAAI,CAAe,CAAC,CAC/G,OAAS,EAAO,CACd,GAAI,IAAS,IAAA,GACX,MAAM,EAER,OAAO,MAAM,EAAkB,EAAiB,CAAO,CACzD,CACA,OAAO,MAAM,EAAe,CAAE,WAAU,kBAAiB,YAAW,EAAG,EAAM,CAAO,CACtF,CAEA,eAAe,EAAkB,EAAyB,EAA+C,CACvG,IAAM,EAAkB,MAAM,EAAsB,CAAe,EAC7D,EAAO,MAAM,EAAW,EAAiB,CAC7C,GAAG,EACH,cAAgB,GAAkB,EAAc,EAAiB,EAAQ,QAAS,CAAa,CACjG,CAAC,EACD,MAAO,CACL,MAAO,EAAK,MAAM,KAAK,CAAE,OAAM,cAAe,CAAE,KAAM,EAAW,EAAM,EAAK,WAAW,EAAG,SAAQ,EAAE,EACpG,YAAa,EAAmB,EAAM,CAAO,EAC7C,OAAQ,EAAK,OACb,SAAU,EAAK,SACf,KAAM,EAAK,WACb,CACF,CAcA,eAAe,EACb,CAAE,WAAU,kBAAiB,cAC7B,EACA,EACqB,CAIrB,GAAI,IAAe,IAAA,IAAa,IAAoB,GAAa,MAAM,EAAU,EAAU,CAAe,EACxG,MAAU,MACR,GAAG,EAAW,EAAiB,CAAQ,EAAE,gGAC3C,EAEF,IAAM,EAAkB,MAAM,EAAoB,CAAQ,EACpD,EAAgB,IAAI,IAAI,IAAe,IAAA,GAAY,CAAC,EAAI,CAAC,CAAU,CAAC,EACpE,EACJ,IAAe,IAAA,IAAa,CAAC,EAAgB,IAAI,EAAW,EAAY,CAAQ,CAAC,EAC7E,EAAW,EAAY,CAAQ,EAC/B,IAAA,GACF,IAAmB,IAAA,IACrB,EAAgB,IAAI,CAAc,EAEpC,IAAM,EAAY,EAAU,MAAM,EAAwB,EAAU,EAAiB,EAAQ,OAAO,EAAG,CAAa,EAE9G,EAAa,EAAW,EAAiB,CAAQ,EACjD,EAAO,MAAM,EACjB,EACA,EACA,CAAE,GAAG,EAAS,kBAAqB,QAAQ,QAAQ,CAAS,CAAE,EAC9D,EACA,IAAe,IAAA,IAAa,IAAoB,EAAW,EAAa,EAC1E,EACM,EAAc,EAAmB,EAAM,CAAO,EAC9C,EAAc,GAClB,EAAkB,EAAK,KAAK,EAAU,CAAY,EAAG,CAAe,EAChE,EAAgB,EAAK,MAAM,KAAK,CAAE,OAAM,cAAe,CAAE,KAAM,EAAW,EAAM,CAAQ,EAAG,SAAQ,EAAE,EACvG,EAAQ,EAAc,QAAQ,CAAE,UAAW,EAAW,CAAI,CAAC,EAC3D,EAAa,GACf,IAAoB,GAAY,EAAM,WAAW,GAAG,EAAW,EAAE,GAAK,EAAM,WAAW,GAAG,EAAW,EAAE,EACrG,EACJ,GAAI,IAAS,IAAA,GAAW,CACtB,EAAY,MAAM,EAAiB,EAAU,CAAI,EACjD,IAAM,EAAe,MAAM,EAAiB,EAAU,CAAS,EAE3D,IAAmB,IAAA,IACrB,EAAa,KAAK,CAAE,OAAQ,QAAS,SAAU,CAAe,CAAC,EAEjE,IAAM,EAAkB,EAAa,QAAQ,CAAE,SAAQ,cAAe,IAAW,WAAa,EAAW,CAAQ,CAAC,EAClH,EAAQ,MAAM,EAAY,EAAe,EAAiB,EAAU,CAAS,EAC7E,EAAa,GAAU,EAAgB,MAAM,CAAE,cAAe,EAAM,WAAW,GAAG,EAAS,EAAE,CAAC,CAChG,CACA,MAAO,CACL,QACA,cACA,GAAG,EAAgB,EAAM,CAAS,EAClC,KAAM,EACN,WACF,CACF,CAGA,SAAS,EACP,EACA,EACyC,CACzC,MAAO,CACL,OAAQ,EAAK,OAAO,OAAQ,GAAU,EAAU,CAAK,CAAC,EACtD,SAAU,CAAC,GAAG,EAAK,SAAU,GAAG,EAAK,OAAO,OAAQ,GAAU,CAAC,EAAU,CAAK,CAAC,CAAC,CAClF,CACF,CAGA,eAAe,EACb,EACA,EACA,EACA,EACwB,CACxB,IAAM,EAAiB,IAAI,IAAI,EAAc,IAAK,GAAa,CAAC,EAAS,KAAM,CAAQ,CAAC,CAAC,EAkBzF,OAAO,MAjBa,EAAgB,EAAc,EAAG,qBAAqB,EAAI,EAAG,KAAO,IAAY,CAClG,IAAM,EAAW,EAAe,IAAI,EAAQ,QAAQ,EAEpD,GAAI,IAAa,IAAA,GACf,MAAO,CAAC,EAEV,GAAM,CAAE,WAAY,EACd,EAAsB,CAAE,UAAW,EAAG,UAAW,EAAQ,MAAM,KAAM,EACrE,EACJ,EAAQ,WAAa,IAAA,GACjB,CAAC,CAAS,EACV,MAAM,EAAc,EAAU,EAAW,EAAQ,SAAU,EAAQ,QAAQ,EAIjF,OAHI,EAAQ,WAAa,UAAY,EAAM,MAAM,CAAE,eAAgB,IAAc,CAAC,GAChF,EAAsB,EAAO,MAAM,EAAS,EAAK,KAAK,EAAU,EAAQ,QAAQ,EAAG,MAAM,CAAC,EAErF,CAAC,CAAE,GAAG,EAAU,OAAM,CAAC,CAChC,CAAC,EAAA,CACY,KAAK,CACpB,CAOA,SAAS,EAAsB,EAAmB,EAAuB,CACvE,IAAM,EAAQ,EAAQ,MAAM,aAAa,EACzC,IAAK,IAAM,KAAQ,EACb,OAAK,UAAY,GACrB,KAAO,EAAK,UAAY,GAAK,iBAAiB,KAAK,EAAM,EAAK,UAAY,IAAM,EAAE,GAChF,IAAK,SAGX,CAGA,SAAS,EAAmB,EAAkB,EAA8C,CAC1F,GAAI,EAAK,WACP,MAAU,MAAM,EAAK,UAAU,EAGjC,OADA,EAAwB,EAAM,CAAO,EAC9B,CACL,UAAW,EAAK,qBAChB,qBAAsB,EAAQ,YAAY,oBAC5C,CACF,CAEA,SAAS,EAAgB,EAA6B,EAAmB,EAA2B,CAClG,GAAM,CAAE,cAAe,EACjB,EAAU,GAAG,EAAM,MAAM,OAAO,GAAG,EAAW,OAAS,IAAA,GAAY,GAAK,WAAW,SAAS,EAAO,qBAAqB,oBAAoB,EAAa,EAAY,CAAK,IAC5K,EAAM,OAAO,OAAS,EAExB,EACE,6BAA6B,EAAM,OAAO,OAAO,6CAA6C,EAAe,CAAU,EAAE,0BAA0B,EAAQ,KAC7J,EACS,EAAW,SAAW,EAC/B,EAAY,4BAA4B,EAAQ,IAAI,EAEpD,EAAY,yBAAyB,EAAe,CAAU,EAAE,IAAI,EAAQ,KAAK,EAEnF,IAAK,IAAM,KAAa,EACtB,EAAY,GAAG,EAAgB,CAAS,EAAE,GAAG,EAG/C,IAAM,EAAkB,IAAI,IAAI,EAAW,SAAS,CAAE,cAAe,EAAS,KAAK,CAAE,YAAa,CAAM,CAAC,CAAC,EACpG,EAAqB,EAAW,OAAQ,GAAc,EAAgB,IAAI,EAAa,CAAS,CAAC,CAAC,EACxG,GAAI,EAAmB,OAAS,EAAG,CACjC,EAAY;;CAAiB,EAC7B,IAAK,GAAM,CAAE,QAAO,UAAU,EAC5B,EAAY,KAAK,EAAM,IAAI,EAAK,GAAG,CAEvC,CAEA,IAAK,IAAM,KAAW,EAAM,SAC1B,EAAY,YAAY,EAAQ,GAAG,EAErC,IAAK,IAAM,KAAS,EAAM,OACxB,EAAY,UAAU,EAAM,GAAG,CAEnC,CAEA,SAAS,EAAa,EAA6B,EAA2B,CAC5E,OAAO,EAAM,YAAc,IAAA,GAAY,GAAK,UAAU,EAAW,KAAK,eAAe,EAAM,UAAU,MAAM,EAAG,EAAE,GAClH,CAEA,SAAS,EAAe,EAA0C,CAChE,IAAM,EAAS,EAAgB,CAAU,EACzC,MAAO,GAAG,EAAO,oBAAoB,WAAW,EAAO,sBAAsB,aAAa,EAAO,uBAAuB,cAAc,EAAO,mBAAmB,UAAU,EAAO,0BAA0B,oBAC7M,CAEA,SAAS,EAAgB,EAOvB,CACA,IAAM,EAAS,GAAoC,EAAW,OAAQ,GAAc,EAAU,OAAS,CAAI,CAAC,CAAC,OACvG,EAAsB,EAAW,QAAQ,CAAE,WAAY,IAAU,OAAO,CAAC,CAAC,OAChF,MAAO,CACL,eAAgB,EAAW,OAC3B,sBACA,sBAAuB,EAAW,OAAS,EAC3C,uBAAwB,EAAM,UAAU,EACxC,mBAAoB,EAAM,MAAM,EAChC,0BAA2B,EAAM,aAAa,CAChD,CACF,CAGA,SAAS,EAAgB,EAA8B,CACrD,IAAM,EAAgB,IAAI,IAAI,EAAW,IAAK,GAAc,CAAC,EAAa,CAAS,EAAG,EAAU,KAAK,CAAC,CAAC,EACjG,EAAW,EAAU,SACxB,KAAK,CAAE,SAAQ,QAAO,QAAO,WAAY,CAExC,IAAM,EAAe,KAAK,KAAK,EAAQ,EAAE,EAAI,GACvC,EAAc,IAAU,EAAU,MAAQ,GAAK,GAAG,EAAM,GAGxD,EAAa,EAAU,OAAS,cAAgB,KAAK,IAAI,KAAK,KAAK,CAAK,EAAI,EAAG,CAAC,EAAI,EACpF,EAAS,IAAW,8BAAgC,EAAsB,CAAS,EAAI,GAC7F,MAAO,GAAG,EAAc,IAAI,CAAM,EAAE,GAAG,EAAa,IAAI,EAAY,MAAM,IAAa,EAAO,EAChG,CAAC,CAAC,CACD,KAAK,IAAI,EACZ,MAAO,GAAG,EAAU,MAAM,IAAI,EAAkB,EAAW,CAAQ,GACrE,CAEA,SAAS,EAAsB,CAAE,gBAAgB,CAAC,GAAwB,CACxE,IAAM,EAAS,EAAc,KAC1B,CAAE,YAAW,UAAS,OAAM,yBAC3B,IAAI,IAAY,IAAY,EAAY,GAAK,IAAI,MAAY,EAAO,IAAI,EAAU,CAAI,IAAM,GAAG,GAAG,GACtG,EACA,OAAO,EAAO,OAAS,EAAI,mBAAmB,EAAO,KAAK,IAAI,IAAM,EACtE,CAEA,SAAS,EAAkB,EAAsB,EAA0B,CACzE,OAAQ,EAAU,KAAlB,CACE,IAAK,OACH,MAAO,GAAG,EAAU,KAAK,IAAI,IAE/B,IAAK,WACH,MAAO,GAAG,EAAe,CAAS,EAAE,GAAG,EAAU,EAAU,MAAQ,EAAE,EAAE,IAAI,IAE7E,IAAK,cAAe,CAClB,IAAM,GAAY,EAAU,UAAY,CAAC,EAAA,CAAG,IAAK,GAAY,EAAe,CAAO,CAAC,EAC9E,EAAS,CAAC,GAAG,EAAS,MAAM,EAAG,CAAiB,EAAG,GAAI,EAAS,OAAS,EAAoB,CAAC,KAAK,EAAI,CAAC,CAAE,EAChH,MAAO,GAAG,EAAe,CAAS,EAAE,IAAI,IAAW,EAAO,OAAS,EAAI,aAAa,EAAO,KAAK,IAAI,IAAM,IAC5G,CACF,CACF,CAGA,SAAS,EAAU,EAAsB,CACvC,OAAO,EAAK,WAAW,iBAAkB,GAAG,CAC9C,CAEA,SAAS,EAAe,CAAE,OAAM,YAAW,WAAkC,CAC3E,MAAO,GAAG,EAAK,GAAG,EAAU,GAAG,GACjC,CAEA,SAAS,EAAgB,EAA6B,EAAmB,EAA2B,CAClG,EACE,KAAK,UACH,CACE,OAAQ,CAAC,EAAU,EAAO,UAAU,GAAK,EAAM,OAAO,SAAW,EACjE,KAAM,EAAW,KACjB,UAAW,EAAM,UACjB,QAAS,CACP,GAAG,EAAgB,EAAO,UAAU,EACpC,iBAAkB,EAAM,MAAM,OAC9B,qBAAsB,EAAO,oBAC/B,EACA,WAAY,EAAO,WACnB,OAAQ,EAAM,OACd,SAAU,EAAM,QAClB,EACA,IAAA,GACA,CACF,EAAI;CACN,CACF"}

package/dist/crossFileDuplication.cjs

@@ -1,2 +1,2 @@
-"use strict";const e=require("./duplication.cjs"),t=require("./crossFileNearMiss.cjs"),n=require("./duplicateSelection.cjs");function r(t,r){let{minTokens:o,maxGapTokens:d,minSimilarityPercent:f}=e.resolveDuplicationOptions(r),p=t.flatMap(({file:e,candidates:t},n)=>t.map(t=>({...t,regionBucket:n,file:e})));for(let e of a(t,o))p.push(e);let m=n.selectMaximalGroups(p,s,(e,t)=>e.regionBucket-t.regionBucket||e.startIndex-t.startIndex),h=c(t,d),g=l([...m.values()],h,d);for(let e of i(t,g,h,o,f))g.push(e);return u(g,t,h)}function i(e,n,r,i,a){let o=e.map(()=>[]);for(let{fileIndex:e,startTokenIndex:t,endTokenIndex:i}of n.flat()){let n=r[e]??0;o[e]?.push({startTokenIndex:t-n,endTokenIndex:i-n})}return t.collectCrossFileNearMissGroups(e,o,i,a).map(t=>t.map(t=>{let n=r[t.fileIndex]??0;return{...t,file:e[t.fileIndex]?.file??``,segments:t.segments.map(e=>({startTokenIndex:e.startTokenIndex+n,endTokenIndex:e.endTokenIndex+n})),startTokenIndex:t.startTokenIndex+n,endTokenIndex:t.endTokenIndex+n}}))}function a(t,n){let r=[],i=[];for(let[n,{tokens:a,containerStatements:o}]of t.entries())a&&o&&(r.push(n),i.push({tokens:a,literalCountPrefix:e.buildLiteralCountPrefix(a),containers:o}));return i.length<2?[]:e.collectSequenceWindowCandidates(i,n,!0).flatMap(({candidate:e,contextIndex:n})=>{let i=r[n],a=i===void 0?void 0:t[i];return i===void 0||a===void 0?[]:[{...e,regionBucket:i,file:a.file}]})}function o(e){return new Set(e.map(e=>e.file)).size>=2}function s(e){return e.length>=2&&new Set(e.map(e=>e.regionBucket)).size>=2}function c(e,t){let n=[],r=0;for(let{tokens:i,candidates:a}of e){n.push(r);let e=i?.length??0;if(!i)for(let t of a)e=Math.max(e,t.endTokenIndex);r+=e+t+1}return n}function l(t,n,r){let i=t.map(e=>e.map(e=>{let t=e.startTokenIndex+(n[e.regionBucket]??0),r=e.endTokenIndex+(n[e.regionBucket]??0);return{file:e.file,fileIndex:e.regionBucket,spanCountedElsewhere:e.nestedInLargerGroup,nestedInLargerGroup:e.nestedInLargerGroup,segments:[{startTokenIndex:t,endTokenIndex:r}],tokenCount:e.tokenCount,startTokenIndex:t,endTokenIndex:r,startIndex:e.startIndex,endIndex:e.endIndex,startLine:e.startLine,endLine:e.endLine}}).toSorted((e,t)=>e.startTokenIndex-t.startTokenIndex));return e.mergeAdjacentGroups(i,r,o)}function u(t,n,r){let i=[],a=/* @__PURE__ */ new Map,o=new Map(n.map((e,t)=>[e.file,{tokens:e.tokens,codeLineNumbers:e.codeLineNumbers,offset:r[t]??0}])),s=/* @__PURE__ */ new Map,c=0;for(let n of t){c+=e.countRedundantFragments(n);for(let e of n)d(e,o,s);let t=n.map(({file:e,startLine:t,endLine:n})=>({file:e,startLine:t,endLine:n})).toSorted((e,t)=>e.file.localeCompare(t.file)||e.startLine-t.startLine),r=[...new Set(t.map(({file:e})=>e))];for(let e of r)a.set(e,(a.get(e)??0)+1);i.push({files:r,occurrences:t,tokenCount:Math.min(...n.map(({tokenCount:e})=>e))})}return i.sort((e,t)=>t.tokenCount-e.tokenCount||(e.occurrences[0]?.file??``).localeCompare(t.occurrences[0]?.file??``)||(e.occurrences[0]?.startLine??0)-(t.occurrences[0]?.startLine??0)),{duplicateBlockCount:c,duplicateBlockGroupCountByFile:Object.fromEntries(a),duplicateLineNumbersByFile:Object.fromEntries([...s].map(([e,t])=>[e,[...t].toSorted((e,t)=>e-t)])),groups:i}}function d(t,n,r){let i=n.get(t.file);if(!i?.tokens)return;let a=r.get(t.file);a||(a=/* @__PURE__ */ new Set,r.set(t.file,a));for(let n of t.segments)e.collectSegmentLines({startTokenIndex:n.startTokenIndex-i.offset,endTokenIndex:n.endTokenIndex-i.offset},i.tokens,i.codeLineNumbers,a)}exports.measureCrossFileDuplication=r;
+"use strict";const e=require("./duplication.cjs"),t=require("./crossFileNearMiss.cjs"),n=require("./duplicateSelection.cjs");function r(t,r){let{minTokens:o,maxGapTokens:d,minSimilarityPercent:f}=e.resolveDuplicationOptions(r),p=t.flatMap(({file:e,candidates:t},n)=>t.map(t=>({...t,regionBucket:n,file:e})));for(let e of a(t,o))p.push(e);let m=n.selectMaximalGroups(p,s,(e,t)=>e.regionBucket-t.regionBucket||e.startIndex-t.startIndex),h=c(t,d),g=l([...m.values()],h,d);for(let e of i(t,g,h,o,f))g.push(e);return u(g,t,h)}function i(e,n,r,i,a){let o=e.map(()=>[]);for(let{fileIndex:e,startTokenIndex:t,endTokenIndex:i}of n.flat()){let n=r[e]??0;o[e]?.push({startTokenIndex:t-n,endTokenIndex:i-n})}return t.collectCrossFileNearMissGroups(e,o,i,a).map(t=>t.map(t=>{let n=r[t.fileIndex]??0;return{...t,file:e[t.fileIndex]?.file??``,segments:t.segments.map(e=>({startTokenIndex:e.startTokenIndex+n,endTokenIndex:e.endTokenIndex+n})),startTokenIndex:t.startTokenIndex+n,endTokenIndex:t.endTokenIndex+n}}))}function a(t,n){let r=[],i=[];for(let[n,{tokens:a,containerStatements:o}]of t.entries())a&&o&&(r.push(n),i.push({tokens:a,literalCountPrefix:e.buildLiteralCountPrefix(a),containers:o}));return i.length<2?[]:e.collectSequenceWindowCandidates(i,n,!0).flatMap(({candidate:e,contextIndex:n})=>{let i=r[n],a=i===void 0?void 0:t[i];return i===void 0||a===void 0?[]:[{...e,regionBucket:i,file:a.file}]})}function o(e){return new Set(e.map(e=>e.file)).size>=2}function s(e){return e.length>=2&&new Set(e.map(e=>e.regionBucket)).size>=2}function c(e,t){let n=[],r=0;for(let{tokens:i,candidates:a}of e){n.push(r);let e=i?.length??0;if(!i)for(let t of a)e=Math.max(e,t.endTokenIndex);r+=e+t+1}return n}function l(t,n,r){let i=t.map(e=>e.map(e=>{let t=e.startTokenIndex+(n[e.regionBucket]??0),r=e.endTokenIndex+(n[e.regionBucket]??0);return{file:e.file,fileIndex:e.regionBucket,spanCountedElsewhere:e.nestedInLargerGroup,nestedInLargerGroup:e.nestedInLargerGroup,segments:[{startTokenIndex:t,endTokenIndex:r}],tokenCount:e.tokenCount,startTokenIndex:t,endTokenIndex:r,startIndex:e.startIndex,endIndex:e.endIndex,startLine:e.startLine,endLine:e.endLine}}).toSorted((e,t)=>e.startTokenIndex-t.startTokenIndex));return e.mergeAdjacentGroups(i,r,o)}function u(t,n,r){let i=[],a=/* @__PURE__ */ new Map,o=new Map(n.map((e,t)=>[e.file,{tokens:e.tokens,codeLineNumbers:e.codeLineNumbers,offset:r[t]??0}])),s=/* @__PURE__ */ new Map,c=0;for(let n of t){c+=e.countRedundantFragments(n);let t=n.map(e=>{let{file:t,startLine:n,endLine:r,tokenCount:i}=e,a=d(e,o,s);return{file:t,startLine:n,endLine:r,...a&&{lineNumbers:a},tokenCount:i}}).toSorted((e,t)=>e.file.localeCompare(t.file)||e.startLine-t.startLine),r=[...new Set(t.map(({file:e})=>e))];for(let e of r)a.set(e,(a.get(e)??0)+1);i.push({files:r,occurrences:t,tokenCount:Math.min(...n.map(({tokenCount:e})=>e))})}return i.sort((e,t)=>t.tokenCount-e.tokenCount||(e.occurrences[0]?.file??``).localeCompare(t.occurrences[0]?.file??``)||(e.occurrences[0]?.startLine??0)-(t.occurrences[0]?.startLine??0)),{duplicateBlockCount:c,duplicateBlockGroupCountByFile:Object.fromEntries(a),duplicateLineNumbersByFile:Object.fromEntries([...s].map(([e,t])=>[e,[...t].toSorted((e,t)=>e-t)])),groups:i}}function d(t,n,r){let i=n.get(t.file);if(!i?.tokens)return;let a=r.get(t.file);a||(a=/* @__PURE__ */ new Set,r.set(t.file,a));let o=/* @__PURE__ */ new Set;for(let n of t.segments)e.collectSegmentLines({startTokenIndex:n.startTokenIndex-i.offset,endTokenIndex:n.endTokenIndex-i.offset},i.tokens,i.codeLineNumbers,o);for(let e of o)a.add(e);return[...o].toSorted((e,t)=>e-t)}exports.measureCrossFileDuplication=r;
 //# sourceMappingURL=crossFileDuplication.cjs.map

package/dist/crossFileDuplication.cjs.map

@@ -1 +1 @@
-{"version":3,"file":"crossFileDuplication.cjs","names":["resolveDuplicationOptions","selectMaximalGroups","collectCrossFileNearMissGroups","buildLiteralCountPrefix","collectSequenceWindowCandidates","mergeAdjacentGroups","countRedundantFragments"],"sources":["../src/crossFileDuplication.ts"],"sourcesContent":["import { collectCrossFileNearMissGroups } from './crossFileNearMiss.js';\nimport { selectMaximalGroups, type SelectableRegion } from './duplicateSelection.js';\nimport {\n  buildLiteralCountPrefix,\n  collectSegmentLines,\n  collectSequenceWindowCandidates,\n  countRedundantFragments,\n  mergeAdjacentGroups,\n  resolveDuplicationOptions,\n  type CountedOccurrence,\n  type CrossFileDuplicateCandidate,\n  type CrossFileDuplicationFileData,\n  type SequenceWindowContext,\n} from './duplication.js';\nimport type { DuplicationOptions } from './types.js';\n\nexport interface CrossFileDuplicationSourceFile extends Partial<CrossFileDuplicationFileData> {\n  file: string;\n  candidates: CrossFileDuplicateCandidate[];\n}\n\nexport interface CrossFileDuplicateOccurrence {\n  endLine: number;\n  file: string;\n  startLine: number;\n}\n\nexport interface CrossFileDuplicateBlockGroup {\n  files: string[];\n  occurrences: CrossFileDuplicateOccurrence[];\n  /**\n   * Token count of the smallest occurrence. For exact and gapped groups it is the matched token\n   * count every occurrence shares (gaps are not counted); for near-miss (Type-3) groups it is the\n   * token count of the smallest whole block or set of matched cores, edited tokens included.\n   */\n  tokenCount: number;\n}\n\nexport interface CrossFileDuplicationMetrics {\n  /** Number of redundant copies across all groups, counted per matched fragment like within-file. */\n  duplicateBlockCount: number;\n  /** Groups the file participates in, keyed by the file name passed in. */\n  duplicateBlockGroupCountByFile: Record<string, number>;\n  /**\n   * Per file, the 1-based code lines covered by the tokens of its cross-file occurrences, sorted\n   * ascending: the matched tokens of exact and gapped occurrences, and every token of a near-miss\n   * block or its matched cores, edited ones included (like within-file near-miss coverage). The unmatched gap of a merged\n   * clone and comment/blank lines inside an occurrence's bounding range are excluded (blank rows\n   * inside multi-row tokens only when the file supplied codeLineNumbers). A file that supplied\n   * only candidates (no `tokens`) has no entry — without its token stream the covered lines are\n   * unknowable, and an approximate bounding range would break this field's exactness.\n   */\n  duplicateLineNumbersByFile: Record<string, number[]>;\n  groups: CrossFileDuplicateBlockGroup[];\n}\n\ninterface SelectableCandidate extends CrossFileDuplicateCandidate, SelectableRegion {\n  regionBucket: number;\n  file: string;\n}\n\n/** A cross-file occurrence: a within-file occurrence in the project-wide token index space. */\ninterface CrossFileOccurrence extends CountedOccurrence {\n  file: string;\n  fileIndex: number;\n}\n\n/**\n * Detects code regions duplicated across files. Per-file candidates (whole block subtrees and full\n * container runs, fingerprinted with the same normalization as within-file duplication) are joined\n * by a project-level window index over per-statement fingerprint sequences (CPD-style), so a\n * copy-pasted partial statement run embedded in different surrounding code is matched even though\n * no single file can know it repeats elsewhere. Candidates are grouped by fingerprint, and only\n * maximal, non-overlapping regions whose group spans at least two files are counted. Groups that\n * shrink to a single file during selection are shed — a within-file repeat is already reported by\n * that file's own duplication metrics. A copy nested inside a larger group's region (two files share\n * a whole function, a third file only a block of it) is reported with its group, so the third\n * file's copy still shows what it duplicates. Groups separated by a small token gap within each file then\n * merge into gapped (Type-3) clone groups under `maxGapTokens`, exactly like within-file merging.\n * Finally, blocks of files that supplied `nearMissBlocks` are compared across files for near-miss\n * (Type-3) clones under `minSimilarityPercent` (see crossFileNearMiss.ts).\n */\nexport function measureCrossFileDuplication(\n  files: CrossFileDuplicationSourceFile[],\n  options?: DuplicationOptions\n): CrossFileDuplicationMetrics {\n  const { minTokens, maxGapTokens, minSimilarityPercent } = resolveDuplicationOptions(options);\n  const candidates: SelectableCandidate[] = files.flatMap(({ file, candidates }, fileIndex) =>\n    candidates.map((candidate) => ({ ...candidate, regionBucket: fileIndex, file }))\n  );\n  // Pushed one by one: spreading the project-scale window-candidate array as call arguments\n  // overflows V8's argument limit (~124k) and crashes on Node, though Bun/JSC tolerates it.\n  for (const candidate of collectWindowCandidates(files, minTokens)) {\n    candidates.push(candidate);\n  }\n  const counted = selectMaximalGroups(\n    candidates,\n    spansMultipleFiles,\n    // File index and position break coverage ties deterministically.\n    (left, right) => left.regionBucket - right.regionBucket || left.startIndex - right.startIndex\n  );\n  const tokenOffsets = computeTokenOffsets(files, maxGapTokens);\n  const groups = mergeGapAdjacentGroups([...counted.values()], tokenOffsets, maxGapTokens);\n  for (const group of collectNearMissGroups(files, groups, tokenOffsets, minTokens, minSimilarityPercent)) {\n    groups.push(group);\n  }\n  return summarize(groups, files, tokenOffsets);\n}\n\n/** Near-miss groups among the blocks the exact groups leave unreported, in the project token space. */\nfunction collectNearMissGroups(\n  files: CrossFileDuplicationSourceFile[],\n  exactGroups: CrossFileOccurrence[][],\n  tokenOffsets: number[],\n  minTokens: number,\n  minSimilarityPercent: number\n): CrossFileOccurrence[][] {\n  const reportedSpansByFile: { startTokenIndex: number; endTokenIndex: number }[][] = files.map(() => []);\n  for (const { fileIndex, startTokenIndex, endTokenIndex } of exactGroups.flat()) {\n    const offset = tokenOffsets[fileIndex] ?? 0;\n    reportedSpansByFile[fileIndex]?.push({\n      startTokenIndex: startTokenIndex - offset,\n      endTokenIndex: endTokenIndex - offset,\n    });\n  }\n  return collectCrossFileNearMissGroups(files, reportedSpansByFile, minTokens, minSimilarityPercent).map((group) =>\n    group.map((occurrence) => {\n      const offset = tokenOffsets[occurrence.fileIndex] ?? 0;\n      return {\n        ...occurrence,\n        file: files[occurrence.fileIndex]?.file ?? '',\n        segments: occurrence.segments.map((segment) => ({\n          startTokenIndex: segment.startTokenIndex + offset,\n          endTokenIndex: segment.endTokenIndex + offset,\n        })),\n        startTokenIndex: occurrence.startTokenIndex + offset,\n        endTokenIndex: occurrence.endTokenIndex + offset,\n      };\n    })\n  );\n}\n\n/** Repeated sub-windows of sibling statements matched across the whole project's files. */\nfunction collectWindowCandidates(files: CrossFileDuplicationSourceFile[], minTokens: number): SelectableCandidate[] {\n  const fileIndexByContext: number[] = [];\n  const contexts: SequenceWindowContext[] = [];\n  for (const [fileIndex, { tokens, containerStatements }] of files.entries()) {\n    if (tokens && containerStatements) {\n      fileIndexByContext.push(fileIndex);\n      contexts.push({ tokens, literalCountPrefix: buildLiteralCountPrefix(tokens), containers: containerStatements });\n    }\n  }\n  if (contexts.length < 2) {\n    return [];\n  }\n  return collectSequenceWindowCandidates(contexts, minTokens, true).flatMap(({ candidate, contextIndex }) => {\n    const fileIndex = fileIndexByContext[contextIndex];\n    const file = fileIndex === undefined ? undefined : files[fileIndex];\n    return fileIndex === undefined || file === undefined\n      ? []\n      : [{ ...candidate, regionBucket: fileIndex, file: file.file }];\n  });\n}\n\n/** A merged group is reported only while it still covers more than one file. */\nfunction spansMultipleFilesAfterMerge(group: CrossFileOccurrence[]): boolean {\n  return new Set(group.map((occurrence) => occurrence.file)).size >= 2;\n}\n\nfunction spansMultipleFiles(group: SelectableCandidate[]): boolean {\n  return group.length >= 2 && new Set(group.map((candidate) => candidate.regionBucket)).size >= 2;\n}\n\n/**\n * Per-file token offsets that map every file into one project-wide token index space: each file's\n * tokens are offset by more than `maxGapTokens` past the previous file's, so occurrences in\n * different files are never gap-adjacent and merged pairs always stay within one file.\n */\nfunction computeTokenOffsets(files: CrossFileDuplicationSourceFile[], maxGapTokens: number): number[] {\n  const tokenOffsets: number[] = [];\n  let offset = 0;\n  for (const { tokens, candidates } of files) {\n    tokenOffsets.push(offset);\n    // Accumulated in a loop: spreading a project-scale candidate array as call arguments would\n    // overflow V8's argument limit (~124k) and crash on Node.\n    let tokenCount = tokens?.length ?? 0;\n    if (!tokens) {\n      for (const candidate of candidates) {\n        tokenCount = Math.max(tokenCount, candidate.endTokenIndex);\n      }\n    }\n    offset += tokenCount + maxGapTokens + 1;\n  }\n  return tokenOffsets;\n}\n\n/** Reuses the within-file gapped (Type-3) merging in the project-wide token index space. */\nfunction mergeGapAdjacentGroups(\n  groups: SelectableCandidate[][],\n  tokenOffsets: number[],\n  maxGapTokens: number\n): CrossFileOccurrence[][] {\n  const occurrenceGroups = groups.map((group) =>\n    group\n      .map((candidate): CrossFileOccurrence => {\n        const start = candidate.startTokenIndex + (tokenOffsets[candidate.regionBucket] ?? 0);\n        const end = candidate.endTokenIndex + (tokenOffsets[candidate.regionBucket] ?? 0);\n        return {\n          file: candidate.file,\n          fileIndex: candidate.regionBucket,\n          spanCountedElsewhere: candidate.nestedInLargerGroup,\n          nestedInLargerGroup: candidate.nestedInLargerGroup,\n          segments: [{ startTokenIndex: start, endTokenIndex: end }],\n          tokenCount: candidate.tokenCount,\n          startTokenIndex: start,\n          endTokenIndex: end,\n          startIndex: candidate.startIndex,\n          endIndex: candidate.endIndex,\n          startLine: candidate.startLine,\n          endLine: candidate.endLine,\n        };\n      })\n      .toSorted((left, right) => left.startTokenIndex - right.startTokenIndex)\n  );\n  return mergeAdjacentGroups(occurrenceGroups, maxGapTokens, spansMultipleFilesAfterMerge);\n}\n\nfunction summarize(\n  groups: CrossFileOccurrence[][],\n  files: CrossFileDuplicationSourceFile[],\n  tokenOffsets: number[]\n): CrossFileDuplicationMetrics {\n  const reported: CrossFileDuplicateBlockGroup[] = [];\n  // Accumulated in Maps: file names are arbitrary strings, and a plain object would read\n  // inherited properties for names like \"constructor\".\n  const groupCountByFile = new Map<string, number>();\n  const fileDataByName = new Map(\n    files.map((file, index) => [\n      file.file,\n      { tokens: file.tokens, codeLineNumbers: file.codeLineNumbers, offset: tokenOffsets[index] ?? 0 },\n    ])\n  );\n  const lineNumbersByFile = new Map<string, Set<number>>();\n  let duplicateBlockCount = 0;\n  for (const group of groups) {\n    // Mirrors within-file counting: each redundant occurrence contributes one count per matched\n    // fragment, gapped merging consolidates the grouping without halving the count, and spans a\n    // partial merge shares between a retained group and the merged group count once.\n    duplicateBlockCount += countRedundantFragments(group);\n    for (const occurrence of group) {\n      collectOccurrenceLines(occurrence, fileDataByName, lineNumbersByFile);\n    }\n    const occurrences = group\n      .map(({ file, startLine, endLine }) => ({ file, startLine, endLine }))\n      .toSorted((left, right) => left.file.localeCompare(right.file) || left.startLine - right.startLine);\n    const files = [...new Set(occurrences.map(({ file }) => file))];\n    for (const file of files) {\n      groupCountByFile.set(file, (groupCountByFile.get(file) ?? 0) + 1);\n    }\n    reported.push({ files, occurrences, tokenCount: Math.min(...group.map(({ tokenCount }) => tokenCount)) });\n  }\n  reported.sort(\n    (left, right) =>\n      right.tokenCount - left.tokenCount ||\n      (left.occurrences[0]?.file ?? '').localeCompare(right.occurrences[0]?.file ?? '') ||\n      (left.occurrences[0]?.startLine ?? 0) - (right.occurrences[0]?.startLine ?? 0)\n  );\n  return {\n    duplicateBlockCount,\n    duplicateBlockGroupCountByFile: Object.fromEntries(groupCountByFile),\n    duplicateLineNumbersByFile: Object.fromEntries(\n      [...lineNumbersByFile].map(([file, lines]) => [file, [...lines].toSorted((left, right) => left - right)])\n    ),\n    groups: reported,\n  };\n}\n\n/**\n * Adds the code lines an occurrence's segment tokens cover (matched tokens of an exact or gapped\n * occurrence, the whole block or matched cores of a near-miss one) to its file's line set, mapping the\n * project-wide token segments back into the file's own token stream. A file that supplied only\n * candidates (no token stream) is skipped rather than approximated from the bounding line range,\n * which would include gap and comment/blank lines and break the field's exactness contract.\n */\nfunction collectOccurrenceLines(\n  occurrence: CrossFileOccurrence,\n  fileDataByName: Map<\n    string,\n    { tokens?: CrossFileDuplicationSourceFile['tokens']; codeLineNumbers?: Set<number>; offset: number }\n  >,\n  lineNumbersByFile: Map<string, Set<number>>\n): void {\n  const fileData = fileDataByName.get(occurrence.file);\n  if (!fileData?.tokens) {\n    return;\n  }\n  let lines = lineNumbersByFile.get(occurrence.file);\n  if (!lines) {\n    lines = new Set();\n    lineNumbersByFile.set(occurrence.file, lines);\n  }\n  for (const segment of occurrence.segments) {\n    collectSegmentLines(\n      {\n        startTokenIndex: segment.startTokenIndex - fileData.offset,\n        endTokenIndex: segment.endTokenIndex - fileData.offset,\n      },\n      fileData.tokens,\n      fileData.codeLineNumbers,\n      lines\n    );\n  }\n}\n"],"mappings":"6HAkFA,SAAgB,EACd,EACA,EAC6B,CAC7B,GAAM,CAAE,YAAW,eAAc,wBAAyBA,EAAAA,0BAA0B,CAAO,EACrF,EAAoC,EAAM,SAAS,CAAE,OAAM,cAAc,IAC7E,EAAW,IAAK,IAAe,CAAE,GAAG,EAAW,aAAc,EAAW,MAAK,EAAE,CACjF,EAGA,IAAK,IAAM,KAAa,EAAwB,EAAO,CAAS,EAC9D,EAAW,KAAK,CAAS,EAE3B,IAAM,EAAUC,EAAAA,oBACd,EACA,GAEC,EAAM,IAAU,EAAK,aAAe,EAAM,cAAgB,EAAK,WAAa,EAAM,UACrF,EACM,EAAe,EAAoB,EAAO,CAAY,EACtD,EAAS,EAAuB,CAAC,GAAG,EAAQ,OAAO,CAAC,EAAG,EAAc,CAAY,EACvF,IAAK,IAAM,KAAS,EAAsB,EAAO,EAAQ,EAAc,EAAW,CAAoB,EACpG,EAAO,KAAK,CAAK,EAEnB,OAAO,EAAU,EAAQ,EAAO,CAAY,CAC9C,CAGA,SAAS,EACP,EACA,EACA,EACA,EACA,EACyB,CACzB,IAAM,EAA8E,EAAM,QAAU,CAAC,CAAC,EACtG,IAAK,GAAM,CAAE,YAAW,kBAAiB,mBAAmB,EAAY,KAAK,EAAG,CAC9E,IAAM,EAAS,EAAa,IAAc,EAC1C,EAAoB,EAAU,EAAE,KAAK,CACnC,gBAAiB,EAAkB,EACnC,cAAe,EAAgB,CACjC,CAAC,CACH,CACA,OAAOC,EAAAA,+BAA+B,EAAO,EAAqB,EAAW,CAAoB,CAAC,CAAC,IAAK,GACtG,EAAM,IAAK,GAAe,CACxB,IAAM,EAAS,EAAa,EAAW,YAAc,EACrD,MAAO,CACL,GAAG,EACH,KAAM,EAAM,EAAW,UAAU,EAAE,MAAQ,GAC3C,SAAU,EAAW,SAAS,IAAK,IAAa,CAC9C,gBAAiB,EAAQ,gBAAkB,EAC3C,cAAe,EAAQ,cAAgB,CACzC,EAAE,EACF,gBAAiB,EAAW,gBAAkB,EAC9C,cAAe,EAAW,cAAgB,CAC5C,CACF,CAAC,CACH,CACF,CAGA,SAAS,EAAwB,EAAyC,EAA0C,CAClH,IAAM,EAA+B,CAAC,EAChC,EAAoC,CAAC,EAC3C,IAAK,GAAM,CAAC,EAAW,CAAE,SAAQ,0BAA0B,EAAM,QAAQ,EACnE,GAAU,IACZ,EAAmB,KAAK,CAAS,EACjC,EAAS,KAAK,CAAE,SAAQ,mBAAoBC,EAAAA,wBAAwB,CAAM,EAAG,WAAY,CAAoB,CAAC,GAMlH,OAHI,EAAS,OAAS,EACb,CAAC,EAEHC,EAAAA,gCAAgC,EAAU,EAAW,EAAI,CAAC,CAAC,SAAS,CAAE,YAAW,kBAAmB,CACzG,IAAM,EAAY,EAAmB,GAC/B,EAAO,IAAc,IAAA,GAAY,IAAA,GAAY,EAAM,GACzD,OAAO,IAAc,IAAA,IAAa,IAAS,IAAA,GACvC,CAAC,EACD,CAAC,CAAE,GAAG,EAAW,aAAc,EAAW,KAAM,EAAK,IAAK,CAAC,CACjE,CAAC,CACH,CAGA,SAAS,EAA6B,EAAuC,CAC3E,OAAO,IAAI,IAAI,EAAM,IAAK,GAAe,EAAW,IAAI,CAAC,CAAC,CAAC,MAAQ,CACrE,CAEA,SAAS,EAAmB,EAAuC,CACjE,OAAO,EAAM,QAAU,GAAK,IAAI,IAAI,EAAM,IAAK,GAAc,EAAU,YAAY,CAAC,CAAC,CAAC,MAAQ,CAChG,CAOA,SAAS,EAAoB,EAAyC,EAAgC,CACpG,IAAM,EAAyB,CAAC,EAC5B,EAAS,EACb,IAAK,GAAM,CAAE,SAAQ,gBAAgB,EAAO,CAC1C,EAAa,KAAK,CAAM,EAGxB,IAAI,EAAa,GAAQ,QAAU,EACnC,GAAI,CAAC,EACH,IAAK,IAAM,KAAa,EACtB,EAAa,KAAK,IAAI,EAAY,EAAU,aAAa,EAG7D,GAAU,EAAa,EAAe,CACxC,CACA,OAAO,CACT,CAGA,SAAS,EACP,EACA,EACA,EACyB,CACzB,IAAM,EAAmB,EAAO,IAAK,GACnC,EACG,IAAK,GAAmC,CACvC,IAAM,EAAQ,EAAU,iBAAmB,EAAa,EAAU,eAAiB,GAC7E,EAAM,EAAU,eAAiB,EAAa,EAAU,eAAiB,GAC/E,MAAO,CACL,KAAM,EAAU,KAChB,UAAW,EAAU,aACrB,qBAAsB,EAAU,oBAChC,oBAAqB,EAAU,oBAC/B,SAAU,CAAC,CAAE,gBAAiB,EAAO,cAAe,CAAI,CAAC,EACzD,WAAY,EAAU,WACtB,gBAAiB,EACjB,cAAe,EACf,WAAY,EAAU,WACtB,SAAU,EAAU,SACpB,UAAW,EAAU,UACrB,QAAS,EAAU,OACrB,CACF,CAAC,CAAC,CACD,UAAU,EAAM,IAAU,EAAK,gBAAkB,EAAM,eAAe,CAC3E,EACA,OAAOC,EAAAA,oBAAoB,EAAkB,EAAc,CAA4B,CACzF,CAEA,SAAS,EACP,EACA,EACA,EAC6B,CAC7B,IAAM,EAA2C,CAAC,EAG5C,kBAAmB,IAAI,IACvB,EAAiB,IAAI,IACzB,EAAM,KAAK,EAAM,IAAU,CACzB,EAAK,KACL,CAAE,OAAQ,EAAK,OAAQ,gBAAiB,EAAK,gBAAiB,OAAQ,EAAa,IAAU,CAAE,CACjG,CAAC,CACH,EACM,kBAAoB,IAAI,IAC1B,EAAsB,EAC1B,IAAK,IAAM,KAAS,EAAQ,CAI1B,GAAuBC,EAAAA,wBAAwB,CAAK,EACpD,IAAK,IAAM,KAAc,EACvB,EAAuB,EAAY,EAAgB,CAAiB,EAEtE,IAAM,EAAc,EACjB,KAAK,CAAE,OAAM,YAAW,cAAe,CAAE,OAAM,YAAW,SAAQ,EAAE,CAAC,CACrE,UAAU,EAAM,IAAU,EAAK,KAAK,cAAc,EAAM,IAAI,GAAK,EAAK,UAAY,EAAM,SAAS,EAC9F,EAAQ,CAAC,GAAG,IAAI,IAAI,EAAY,KAAK,CAAE,UAAW,CAAI,CAAC,CAAC,EAC9D,IAAK,IAAM,KAAQ,EACjB,EAAiB,IAAI,GAAO,EAAiB,IAAI,CAAI,GAAK,GAAK,CAAC,EAElE,EAAS,KAAK,CAAE,QAAO,cAAa,WAAY,KAAK,IAAI,GAAG,EAAM,KAAK,CAAE,gBAAiB,CAAU,CAAC,CAAE,CAAC,CAC1G,CAOA,OANA,EAAS,MACN,EAAM,IACL,EAAM,WAAa,EAAK,aACvB,EAAK,YAAY,EAAE,EAAE,MAAQ,GAAA,CAAI,cAAc,EAAM,YAAY,EAAE,EAAE,MAAQ,EAAE,IAC/E,EAAK,YAAY,EAAE,EAAE,WAAa,IAAM,EAAM,YAAY,EAAE,EAAE,WAAa,EAChF,EACO,CACL,sBACA,+BAAgC,OAAO,YAAY,CAAgB,EACnE,2BAA4B,OAAO,YACjC,CAAC,GAAG,CAAiB,CAAC,CAAC,KAAK,CAAC,EAAM,KAAW,CAAC,EAAM,CAAC,GAAG,CAAK,CAAC,CAAC,UAAU,EAAM,IAAU,EAAO,CAAK,CAAC,CAAC,CAC1G,EACA,OAAQ,CACV,CACF,CASA,SAAS,EACP,EACA,EAIA,EACM,CACN,IAAM,EAAW,EAAe,IAAI,EAAW,IAAI,EACnD,GAAI,CAAC,GAAU,OACb,OAEF,IAAI,EAAQ,EAAkB,IAAI,EAAW,IAAI,EAC5C,IACH,kBAAQ,IAAI,IACZ,EAAkB,IAAI,EAAW,KAAM,CAAK,GAE9C,IAAK,IAAM,KAAW,EAAW,SAC/B,EAAA,oBACE,CACE,gBAAiB,EAAQ,gBAAkB,EAAS,OACpD,cAAe,EAAQ,cAAgB,EAAS,MAClD,EACA,EAAS,OACT,EAAS,gBACT,CACF,CAEJ"}
+{"version":3,"file":"crossFileDuplication.cjs","names":["resolveDuplicationOptions","selectMaximalGroups","collectCrossFileNearMissGroups","buildLiteralCountPrefix","collectSequenceWindowCandidates","mergeAdjacentGroups","countRedundantFragments"],"sources":["../src/crossFileDuplication.ts"],"sourcesContent":["import { collectCrossFileNearMissGroups } from './crossFileNearMiss.js';\nimport { selectMaximalGroups, type SelectableRegion } from './duplicateSelection.js';\nimport {\n  buildLiteralCountPrefix,\n  collectSegmentLines,\n  collectSequenceWindowCandidates,\n  countRedundantFragments,\n  mergeAdjacentGroups,\n  resolveDuplicationOptions,\n  type CountedOccurrence,\n  type CrossFileDuplicateCandidate,\n  type CrossFileDuplicationFileData,\n  type SequenceWindowContext,\n} from './duplication.js';\nimport type { DuplicationOptions } from './types.js';\n\nexport interface CrossFileDuplicationSourceFile extends Partial<CrossFileDuplicationFileData> {\n  file: string;\n  candidates: CrossFileDuplicateCandidate[];\n}\n\nexport interface CrossFileDuplicateOccurrence {\n  endLine: number;\n  file: string;\n  /**\n   * The code lines carrying the occurrence's matched tokens (see\n   * DuplicateBlockOccurrence.lineNumbers). Absent for a file that supplied only candidates (no\n   * `tokens`), whose matched lines are unknowable.\n   */\n  lineNumbers?: number[];\n  startLine: number;\n  /**\n   * The tokens of its matched runs, or of a near-miss block or its matched cores with the edited\n   * tokens among them; the group's `tokenCount` is the smallest of its occurrences'.\n   */\n  tokenCount: number;\n}\n\nexport interface CrossFileDuplicateBlockGroup {\n  files: string[];\n  occurrences: CrossFileDuplicateOccurrence[];\n  /**\n   * Token count of the smallest occurrence. For exact and gapped groups it is the matched token\n   * count every occurrence shares (gaps are not counted); for near-miss (Type-3) groups it is the\n   * token count of the smallest whole block or set of matched cores, edited tokens included.\n   */\n  tokenCount: number;\n}\n\nexport interface CrossFileDuplicationMetrics {\n  /** Number of redundant copies across all groups, counted per matched fragment like within-file. */\n  duplicateBlockCount: number;\n  /** Groups the file participates in, keyed by the file name passed in. */\n  duplicateBlockGroupCountByFile: Record<string, number>;\n  /**\n   * Per file, the 1-based code lines covered by the tokens of its cross-file occurrences, sorted\n   * ascending: the matched tokens of exact and gapped occurrences, and every token of a near-miss\n   * block or its matched cores, edited ones included (like within-file near-miss coverage). The unmatched gap of a merged\n   * clone and comment/blank lines inside an occurrence's bounding range are excluded (blank rows\n   * inside multi-row tokens only when the file supplied codeLineNumbers). A file that supplied\n   * only candidates (no `tokens`) has no entry — without its token stream the covered lines are\n   * unknowable, and an approximate bounding range would break this field's exactness.\n   */\n  duplicateLineNumbersByFile: Record<string, number[]>;\n  groups: CrossFileDuplicateBlockGroup[];\n}\n\ninterface SelectableCandidate extends CrossFileDuplicateCandidate, SelectableRegion {\n  regionBucket: number;\n  file: string;\n}\n\n/** A cross-file occurrence: a within-file occurrence in the project-wide token index space. */\ninterface CrossFileOccurrence extends CountedOccurrence {\n  file: string;\n  fileIndex: number;\n}\n\n/**\n * Detects code regions duplicated across files. Per-file candidates (whole block subtrees and full\n * container runs, fingerprinted with the same normalization as within-file duplication) are joined\n * by a project-level window index over per-statement fingerprint sequences (CPD-style), so a\n * copy-pasted partial statement run embedded in different surrounding code is matched even though\n * no single file can know it repeats elsewhere. Candidates are grouped by fingerprint, and only\n * maximal, non-overlapping regions whose group spans at least two files are counted. Groups that\n * shrink to a single file during selection are shed — a within-file repeat is already reported by\n * that file's own duplication metrics. A copy nested inside a larger group's region (two files share\n * a whole function, a third file only a block of it) is reported with its group, so the third\n * file's copy still shows what it duplicates. Groups separated by a small token gap within each file then\n * merge into gapped (Type-3) clone groups under `maxGapTokens`, exactly like within-file merging.\n * Finally, blocks of files that supplied `nearMissBlocks` are compared across files for near-miss\n * (Type-3) clones under `minSimilarityPercent` (see crossFileNearMiss.ts).\n */\nexport function measureCrossFileDuplication(\n  files: CrossFileDuplicationSourceFile[],\n  options?: DuplicationOptions\n): CrossFileDuplicationMetrics {\n  const { minTokens, maxGapTokens, minSimilarityPercent } = resolveDuplicationOptions(options);\n  const candidates: SelectableCandidate[] = files.flatMap(({ file, candidates }, fileIndex) =>\n    candidates.map((candidate) => ({ ...candidate, regionBucket: fileIndex, file }))\n  );\n  // Pushed one by one: spreading the project-scale window-candidate array as call arguments\n  // overflows V8's argument limit (~124k) and crashes on Node, though Bun/JSC tolerates it.\n  for (const candidate of collectWindowCandidates(files, minTokens)) {\n    candidates.push(candidate);\n  }\n  const counted = selectMaximalGroups(\n    candidates,\n    spansMultipleFiles,\n    // File index and position break coverage ties deterministically.\n    (left, right) => left.regionBucket - right.regionBucket || left.startIndex - right.startIndex\n  );\n  const tokenOffsets = computeTokenOffsets(files, maxGapTokens);\n  const groups = mergeGapAdjacentGroups([...counted.values()], tokenOffsets, maxGapTokens);\n  for (const group of collectNearMissGroups(files, groups, tokenOffsets, minTokens, minSimilarityPercent)) {\n    groups.push(group);\n  }\n  return summarize(groups, files, tokenOffsets);\n}\n\n/** Near-miss groups among the blocks the exact groups leave unreported, in the project token space. */\nfunction collectNearMissGroups(\n  files: CrossFileDuplicationSourceFile[],\n  exactGroups: CrossFileOccurrence[][],\n  tokenOffsets: number[],\n  minTokens: number,\n  minSimilarityPercent: number\n): CrossFileOccurrence[][] {\n  const reportedSpansByFile: { startTokenIndex: number; endTokenIndex: number }[][] = files.map(() => []);\n  for (const { fileIndex, startTokenIndex, endTokenIndex } of exactGroups.flat()) {\n    const offset = tokenOffsets[fileIndex] ?? 0;\n    reportedSpansByFile[fileIndex]?.push({\n      startTokenIndex: startTokenIndex - offset,\n      endTokenIndex: endTokenIndex - offset,\n    });\n  }\n  return collectCrossFileNearMissGroups(files, reportedSpansByFile, minTokens, minSimilarityPercent).map((group) =>\n    group.map((occurrence) => {\n      const offset = tokenOffsets[occurrence.fileIndex] ?? 0;\n      return {\n        ...occurrence,\n        file: files[occurrence.fileIndex]?.file ?? '',\n        segments: occurrence.segments.map((segment) => ({\n          startTokenIndex: segment.startTokenIndex + offset,\n          endTokenIndex: segment.endTokenIndex + offset,\n        })),\n        startTokenIndex: occurrence.startTokenIndex + offset,\n        endTokenIndex: occurrence.endTokenIndex + offset,\n      };\n    })\n  );\n}\n\n/** Repeated sub-windows of sibling statements matched across the whole project's files. */\nfunction collectWindowCandidates(files: CrossFileDuplicationSourceFile[], minTokens: number): SelectableCandidate[] {\n  const fileIndexByContext: number[] = [];\n  const contexts: SequenceWindowContext[] = [];\n  for (const [fileIndex, { tokens, containerStatements }] of files.entries()) {\n    if (tokens && containerStatements) {\n      fileIndexByContext.push(fileIndex);\n      contexts.push({ tokens, literalCountPrefix: buildLiteralCountPrefix(tokens), containers: containerStatements });\n    }\n  }\n  if (contexts.length < 2) {\n    return [];\n  }\n  return collectSequenceWindowCandidates(contexts, minTokens, true).flatMap(({ candidate, contextIndex }) => {\n    const fileIndex = fileIndexByContext[contextIndex];\n    const file = fileIndex === undefined ? undefined : files[fileIndex];\n    return fileIndex === undefined || file === undefined\n      ? []\n      : [{ ...candidate, regionBucket: fileIndex, file: file.file }];\n  });\n}\n\n/** A merged group is reported only while it still covers more than one file. */\nfunction spansMultipleFilesAfterMerge(group: CrossFileOccurrence[]): boolean {\n  return new Set(group.map((occurrence) => occurrence.file)).size >= 2;\n}\n\nfunction spansMultipleFiles(group: SelectableCandidate[]): boolean {\n  return group.length >= 2 && new Set(group.map((candidate) => candidate.regionBucket)).size >= 2;\n}\n\n/**\n * Per-file token offsets that map every file into one project-wide token index space: each file's\n * tokens are offset by more than `maxGapTokens` past the previous file's, so occurrences in\n * different files are never gap-adjacent and merged pairs always stay within one file.\n */\nfunction computeTokenOffsets(files: CrossFileDuplicationSourceFile[], maxGapTokens: number): number[] {\n  const tokenOffsets: number[] = [];\n  let offset = 0;\n  for (const { tokens, candidates } of files) {\n    tokenOffsets.push(offset);\n    // Accumulated in a loop: spreading a project-scale candidate array as call arguments would\n    // overflow V8's argument limit (~124k) and crash on Node.\n    let tokenCount = tokens?.length ?? 0;\n    if (!tokens) {\n      for (const candidate of candidates) {\n        tokenCount = Math.max(tokenCount, candidate.endTokenIndex);\n      }\n    }\n    offset += tokenCount + maxGapTokens + 1;\n  }\n  return tokenOffsets;\n}\n\n/** Reuses the within-file gapped (Type-3) merging in the project-wide token index space. */\nfunction mergeGapAdjacentGroups(\n  groups: SelectableCandidate[][],\n  tokenOffsets: number[],\n  maxGapTokens: number\n): CrossFileOccurrence[][] {\n  const occurrenceGroups = groups.map((group) =>\n    group\n      .map((candidate): CrossFileOccurrence => {\n        const start = candidate.startTokenIndex + (tokenOffsets[candidate.regionBucket] ?? 0);\n        const end = candidate.endTokenIndex + (tokenOffsets[candidate.regionBucket] ?? 0);\n        return {\n          file: candidate.file,\n          fileIndex: candidate.regionBucket,\n          spanCountedElsewhere: candidate.nestedInLargerGroup,\n          nestedInLargerGroup: candidate.nestedInLargerGroup,\n          segments: [{ startTokenIndex: start, endTokenIndex: end }],\n          tokenCount: candidate.tokenCount,\n          startTokenIndex: start,\n          endTokenIndex: end,\n          startIndex: candidate.startIndex,\n          endIndex: candidate.endIndex,\n          startLine: candidate.startLine,\n          endLine: candidate.endLine,\n        };\n      })\n      .toSorted((left, right) => left.startTokenIndex - right.startTokenIndex)\n  );\n  return mergeAdjacentGroups(occurrenceGroups, maxGapTokens, spansMultipleFilesAfterMerge);\n}\n\nfunction summarize(\n  groups: CrossFileOccurrence[][],\n  files: CrossFileDuplicationSourceFile[],\n  tokenOffsets: number[]\n): CrossFileDuplicationMetrics {\n  const reported: CrossFileDuplicateBlockGroup[] = [];\n  // Accumulated in Maps: file names are arbitrary strings, and a plain object would read\n  // inherited properties for names like \"constructor\".\n  const groupCountByFile = new Map<string, number>();\n  const fileDataByName = new Map(\n    files.map((file, index) => [\n      file.file,\n      { tokens: file.tokens, codeLineNumbers: file.codeLineNumbers, offset: tokenOffsets[index] ?? 0 },\n    ])\n  );\n  const lineNumbersByFile = new Map<string, Set<number>>();\n  let duplicateBlockCount = 0;\n  for (const group of groups) {\n    // Mirrors within-file counting: each redundant occurrence contributes one count per matched\n    // fragment, gapped merging consolidates the grouping without halving the count, and spans a\n    // partial merge shares between a retained group and the merged group count once.\n    duplicateBlockCount += countRedundantFragments(group);\n    const occurrences = group\n      .map((occurrence) => {\n        const { file, startLine, endLine, tokenCount } = occurrence;\n        const lineNumbers = collectOccurrenceLines(occurrence, fileDataByName, lineNumbersByFile);\n        return { file, startLine, endLine, ...(lineNumbers && { lineNumbers }), tokenCount };\n      })\n      .toSorted((left, right) => left.file.localeCompare(right.file) || left.startLine - right.startLine);\n    const files = [...new Set(occurrences.map(({ file }) => file))];\n    for (const file of files) {\n      groupCountByFile.set(file, (groupCountByFile.get(file) ?? 0) + 1);\n    }\n    reported.push({ files, occurrences, tokenCount: Math.min(...group.map(({ tokenCount }) => tokenCount)) });\n  }\n  reported.sort(\n    (left, right) =>\n      right.tokenCount - left.tokenCount ||\n      (left.occurrences[0]?.file ?? '').localeCompare(right.occurrences[0]?.file ?? '') ||\n      (left.occurrences[0]?.startLine ?? 0) - (right.occurrences[0]?.startLine ?? 0)\n  );\n  return {\n    duplicateBlockCount,\n    duplicateBlockGroupCountByFile: Object.fromEntries(groupCountByFile),\n    duplicateLineNumbersByFile: Object.fromEntries(\n      [...lineNumbersByFile].map(([file, lines]) => [file, [...lines].toSorted((left, right) => left - right)])\n    ),\n    groups: reported,\n  };\n}\n\n/**\n * Adds the code lines an occurrence's segment tokens cover (matched tokens of an exact or gapped\n * occurrence, the whole block or matched cores of a near-miss one) to its file's line set, mapping the\n * project-wide token segments back into the file's own token stream, and returns them in order.\n * A file that supplied only candidates (no token stream) is skipped rather than approximated from\n * the bounding line range, which would include gap and comment/blank lines and break the field's\n * exactness contract.\n */\nfunction collectOccurrenceLines(\n  occurrence: CrossFileOccurrence,\n  fileDataByName: Map<\n    string,\n    { tokens?: CrossFileDuplicationSourceFile['tokens']; codeLineNumbers?: Set<number>; offset: number }\n  >,\n  lineNumbersByFile: Map<string, Set<number>>\n): number[] | undefined {\n  const fileData = fileDataByName.get(occurrence.file);\n  if (!fileData?.tokens) {\n    return undefined;\n  }\n  let fileLines = lineNumbersByFile.get(occurrence.file);\n  if (!fileLines) {\n    fileLines = new Set();\n    lineNumbersByFile.set(occurrence.file, fileLines);\n  }\n  const lines = new Set<number>();\n  for (const segment of occurrence.segments) {\n    collectSegmentLines(\n      {\n        startTokenIndex: segment.startTokenIndex - fileData.offset,\n        endTokenIndex: segment.endTokenIndex - fileData.offset,\n      },\n      fileData.tokens,\n      fileData.codeLineNumbers,\n      lines\n    );\n  }\n  for (const line of lines) {\n    fileLines.add(line);\n  }\n  return [...lines].toSorted((left, right) => left - right);\n}\n"],"mappings":"6HA6FA,SAAgB,EACd,EACA,EAC6B,CAC7B,GAAM,CAAE,YAAW,eAAc,wBAAyBA,EAAAA,0BAA0B,CAAO,EACrF,EAAoC,EAAM,SAAS,CAAE,OAAM,cAAc,IAC7E,EAAW,IAAK,IAAe,CAAE,GAAG,EAAW,aAAc,EAAW,MAAK,EAAE,CACjF,EAGA,IAAK,IAAM,KAAa,EAAwB,EAAO,CAAS,EAC9D,EAAW,KAAK,CAAS,EAE3B,IAAM,EAAUC,EAAAA,oBACd,EACA,GAEC,EAAM,IAAU,EAAK,aAAe,EAAM,cAAgB,EAAK,WAAa,EAAM,UACrF,EACM,EAAe,EAAoB,EAAO,CAAY,EACtD,EAAS,EAAuB,CAAC,GAAG,EAAQ,OAAO,CAAC,EAAG,EAAc,CAAY,EACvF,IAAK,IAAM,KAAS,EAAsB,EAAO,EAAQ,EAAc,EAAW,CAAoB,EACpG,EAAO,KAAK,CAAK,EAEnB,OAAO,EAAU,EAAQ,EAAO,CAAY,CAC9C,CAGA,SAAS,EACP,EACA,EACA,EACA,EACA,EACyB,CACzB,IAAM,EAA8E,EAAM,QAAU,CAAC,CAAC,EACtG,IAAK,GAAM,CAAE,YAAW,kBAAiB,mBAAmB,EAAY,KAAK,EAAG,CAC9E,IAAM,EAAS,EAAa,IAAc,EAC1C,EAAoB,EAAU,EAAE,KAAK,CACnC,gBAAiB,EAAkB,EACnC,cAAe,EAAgB,CACjC,CAAC,CACH,CACA,OAAOC,EAAAA,+BAA+B,EAAO,EAAqB,EAAW,CAAoB,CAAC,CAAC,IAAK,GACtG,EAAM,IAAK,GAAe,CACxB,IAAM,EAAS,EAAa,EAAW,YAAc,EACrD,MAAO,CACL,GAAG,EACH,KAAM,EAAM,EAAW,UAAU,EAAE,MAAQ,GAC3C,SAAU,EAAW,SAAS,IAAK,IAAa,CAC9C,gBAAiB,EAAQ,gBAAkB,EAC3C,cAAe,EAAQ,cAAgB,CACzC,EAAE,EACF,gBAAiB,EAAW,gBAAkB,EAC9C,cAAe,EAAW,cAAgB,CAC5C,CACF,CAAC,CACH,CACF,CAGA,SAAS,EAAwB,EAAyC,EAA0C,CAClH,IAAM,EAA+B,CAAC,EAChC,EAAoC,CAAC,EAC3C,IAAK,GAAM,CAAC,EAAW,CAAE,SAAQ,0BAA0B,EAAM,QAAQ,EACnE,GAAU,IACZ,EAAmB,KAAK,CAAS,EACjC,EAAS,KAAK,CAAE,SAAQ,mBAAoBC,EAAAA,wBAAwB,CAAM,EAAG,WAAY,CAAoB,CAAC,GAMlH,OAHI,EAAS,OAAS,EACb,CAAC,EAEHC,EAAAA,gCAAgC,EAAU,EAAW,EAAI,CAAC,CAAC,SAAS,CAAE,YAAW,kBAAmB,CACzG,IAAM,EAAY,EAAmB,GAC/B,EAAO,IAAc,IAAA,GAAY,IAAA,GAAY,EAAM,GACzD,OAAO,IAAc,IAAA,IAAa,IAAS,IAAA,GACvC,CAAC,EACD,CAAC,CAAE,GAAG,EAAW,aAAc,EAAW,KAAM,EAAK,IAAK,CAAC,CACjE,CAAC,CACH,CAGA,SAAS,EAA6B,EAAuC,CAC3E,OAAO,IAAI,IAAI,EAAM,IAAK,GAAe,EAAW,IAAI,CAAC,CAAC,CAAC,MAAQ,CACrE,CAEA,SAAS,EAAmB,EAAuC,CACjE,OAAO,EAAM,QAAU,GAAK,IAAI,IAAI,EAAM,IAAK,GAAc,EAAU,YAAY,CAAC,CAAC,CAAC,MAAQ,CAChG,CAOA,SAAS,EAAoB,EAAyC,EAAgC,CACpG,IAAM,EAAyB,CAAC,EAC5B,EAAS,EACb,IAAK,GAAM,CAAE,SAAQ,gBAAgB,EAAO,CAC1C,EAAa,KAAK,CAAM,EAGxB,IAAI,EAAa,GAAQ,QAAU,EACnC,GAAI,CAAC,EACH,IAAK,IAAM,KAAa,EACtB,EAAa,KAAK,IAAI,EAAY,EAAU,aAAa,EAG7D,GAAU,EAAa,EAAe,CACxC,CACA,OAAO,CACT,CAGA,SAAS,EACP,EACA,EACA,EACyB,CACzB,IAAM,EAAmB,EAAO,IAAK,GACnC,EACG,IAAK,GAAmC,CACvC,IAAM,EAAQ,EAAU,iBAAmB,EAAa,EAAU,eAAiB,GAC7E,EAAM,EAAU,eAAiB,EAAa,EAAU,eAAiB,GAC/E,MAAO,CACL,KAAM,EAAU,KAChB,UAAW,EAAU,aACrB,qBAAsB,EAAU,oBAChC,oBAAqB,EAAU,oBAC/B,SAAU,CAAC,CAAE,gBAAiB,EAAO,cAAe,CAAI,CAAC,EACzD,WAAY,EAAU,WACtB,gBAAiB,EACjB,cAAe,EACf,WAAY,EAAU,WACtB,SAAU,EAAU,SACpB,UAAW,EAAU,UACrB,QAAS,EAAU,OACrB,CACF,CAAC,CAAC,CACD,UAAU,EAAM,IAAU,EAAK,gBAAkB,EAAM,eAAe,CAC3E,EACA,OAAOC,EAAAA,oBAAoB,EAAkB,EAAc,CAA4B,CACzF,CAEA,SAAS,EACP,EACA,EACA,EAC6B,CAC7B,IAAM,EAA2C,CAAC,EAG5C,kBAAmB,IAAI,IACvB,EAAiB,IAAI,IACzB,EAAM,KAAK,EAAM,IAAU,CACzB,EAAK,KACL,CAAE,OAAQ,EAAK,OAAQ,gBAAiB,EAAK,gBAAiB,OAAQ,EAAa,IAAU,CAAE,CACjG,CAAC,CACH,EACM,kBAAoB,IAAI,IAC1B,EAAsB,EAC1B,IAAK,IAAM,KAAS,EAAQ,CAI1B,GAAuBC,EAAAA,wBAAwB,CAAK,EACpD,IAAM,EAAc,EACjB,IAAK,GAAe,CACnB,GAAM,CAAE,OAAM,YAAW,UAAS,cAAe,EAC3C,EAAc,EAAuB,EAAY,EAAgB,CAAiB,EACxF,MAAO,CAAE,OAAM,YAAW,UAAS,GAAI,GAAe,CAAE,aAAY,EAAI,YAAW,CACrF,CAAC,CAAC,CACD,UAAU,EAAM,IAAU,EAAK,KAAK,cAAc,EAAM,IAAI,GAAK,EAAK,UAAY,EAAM,SAAS,EAC9F,EAAQ,CAAC,GAAG,IAAI,IAAI,EAAY,KAAK,CAAE,UAAW,CAAI,CAAC,CAAC,EAC9D,IAAK,IAAM,KAAQ,EACjB,EAAiB,IAAI,GAAO,EAAiB,IAAI,CAAI,GAAK,GAAK,CAAC,EAElE,EAAS,KAAK,CAAE,QAAO,cAAa,WAAY,KAAK,IAAI,GAAG,EAAM,KAAK,CAAE,gBAAiB,CAAU,CAAC,CAAE,CAAC,CAC1G,CAOA,OANA,EAAS,MACN,EAAM,IACL,EAAM,WAAa,EAAK,aACvB,EAAK,YAAY,EAAE,EAAE,MAAQ,GAAA,CAAI,cAAc,EAAM,YAAY,EAAE,EAAE,MAAQ,EAAE,IAC/E,EAAK,YAAY,EAAE,EAAE,WAAa,IAAM,EAAM,YAAY,EAAE,EAAE,WAAa,EAChF,EACO,CACL,sBACA,+BAAgC,OAAO,YAAY,CAAgB,EACnE,2BAA4B,OAAO,YACjC,CAAC,GAAG,CAAiB,CAAC,CAAC,KAAK,CAAC,EAAM,KAAW,CAAC,EAAM,CAAC,GAAG,CAAK,CAAC,CAAC,UAAU,EAAM,IAAU,EAAO,CAAK,CAAC,CAAC,CAC1G,EACA,OAAQ,CACV,CACF,CAUA,SAAS,EACP,EACA,EAIA,EACsB,CACtB,IAAM,EAAW,EAAe,IAAI,EAAW,IAAI,EACnD,GAAI,CAAC,GAAU,OACb,OAEF,IAAI,EAAY,EAAkB,IAAI,EAAW,IAAI,EAChD,IACH,kBAAY,IAAI,IAChB,EAAkB,IAAI,EAAW,KAAM,CAAS,GAElD,IAAM,kBAAQ,IAAI,IAClB,IAAK,IAAM,KAAW,EAAW,SAC/B,EAAA,oBACE,CACE,gBAAiB,EAAQ,gBAAkB,EAAS,OACpD,cAAe,EAAQ,cAAgB,EAAS,MAClD,EACA,EAAS,OACT,EAAS,gBACT,CACF,EAEF,IAAK,IAAM,KAAQ,EACjB,EAAU,IAAI,CAAI,EAEpB,MAAO,CAAC,GAAG,CAAK,CAAC,CAAC,UAAU,EAAM,IAAU,EAAO,CAAK,CAC1D"}

package/dist/crossFileDuplication.d.ts

@@ -7,7 +7,18 @@ export interface CrossFileDuplicationSourceFile extends Partial<CrossFileDuplica
 export interface CrossFileDuplicateOccurrence {
     endLine: number;
     file: string;
+    /**
+     * The code lines carrying the occurrence's matched tokens (see
+     * DuplicateBlockOccurrence.lineNumbers). Absent for a file that supplied only candidates (no
+     * `tokens`), whose matched lines are unknowable.
+     */
+    lineNumbers?: number[];
     startLine: number;
+    /**
+     * The tokens of its matched runs, or of a near-miss block or its matched cores with the edited
+     * tokens among them; the group's `tokenCount` is the smallest of its occurrences'.
+     */
+    tokenCount: number;
 }
 export interface CrossFileDuplicateBlockGroup {
     files: string[];