linguist-js 2.3.2 → 2.4.0

package/dist/cli.js

@@ -26,6 +26,7 @@ commander_1.program
     .option('-I|--checkIgnored [bool]', 'Force the checking of gitignore files', true)
     .option('-H|--checkHeuristics [bool]', 'Apply heuristics to ambiguous languages', true)
     .option('-S|--checkShebang [bool]', 'Check shebang lines for explicit classification', true)
+    .option('-M|--checkModeline [bool]', 'Check modelines for explicit classification', true)
     .helpOption(`-h|--help`, 'Display this help message')
     .version(VERSION, '-v|--version', 'Display the installed version of linguist-js');
 commander_1.program.parse(process.argv);

package/dist/index.js

@@ -15,8 +15,8 @@ const read_file_1 = __importDefault(require("./helpers/read-file"));
 const convert_pcre_1 = __importDefault(require("./helpers/convert-pcre"));
 const convert_glob_1 = __importDefault(require("./helpers/convert-glob"));
 async function analyse(input, opts = {}) {
-    var _a, _b, _c, _d, _e, _f, _g, _h, _j, _k, _l, _m, _o, _p, _q;
-    var _r, _s, _t;
+    var _a, _b, _c, _d, _e, _f, _g, _h, _j, _k, _l, _m, _o, _p, _q, _r, _s;
+    var _t, _u, _v;
     const useRawContent = opts.fileContent !== undefined;
     input = [input !== null && input !== void 0 ? input : []].flat();
     opts.fileContent = [(_a = opts.fileContent) !== null && _a !== void 0 ? _a : []].flat();
@@ -29,6 +29,7 @@ async function analyse(input, opts = {}) {
     const vendorPaths = [...vendorData, ...docData, ...generatedData];
     // Setup main variables
     const fileAssociations = {};
+    const definiteness = {};
     const extensions = {};
     const overrides = {};
     const results = {
@@ -61,7 +62,14 @@ async function analyse(input, opts = {}) {
         folders = data.folders;
     }
     // Apply aliases
-    opts = { checkIgnored: !opts.quick, checkAttributes: !opts.quick, checkHeuristics: !opts.quick, checkShebang: !opts.quick, ...opts };
+    opts = {
+        checkIgnored: !opts.quick,
+        checkAttributes: !opts.quick,
+        checkHeuristics: !opts.quick,
+        checkShebang: !opts.quick,
+        checkModeline: !opts.quick,
+        ...opts
+    };
     // Ignore specific languages
     for (const lang of (_b = opts.ignoredLanguages) !== null && _b !== void 0 ? _b : []) {
         for (const key in langData) {
@@ -152,14 +160,30 @@ async function analyse(input, opts = {}) {
         // Skip if file is unreadable
         if (firstLine === null)
             continue;
-        // Check shebang line for explicit classification
-        if (!opts.quick && opts.checkShebang && firstLine.startsWith('#!')) {
-            // Find matching interpreters
-            const matches = Object.entries(langData).filter(([, data]) => { var _a; return (_a = data.interpreters) === null || _a === void 0 ? void 0 : _a.some(interpreter => firstLine.match('\\b' + interpreter + '\\b')); });
+        // Check first line for explicit classification
+        const hasShebang = opts.checkShebang && /^#!/.test(firstLine);
+        const hasModeline = opts.checkModeline && /-\*-|(syntax|filetype|ft)\s*=/.test(firstLine);
+        if (!opts.quick && (hasShebang || hasModeline)) {
+            const matches = [];
+            for (const [lang, data] of Object.entries(langData)) {
+                const langMatcher = (lang) => `\\b${lang.toLowerCase().replace(/\W/g, '\\$&')}\\b`;
+                // Check for interpreter match
+                const matchesInterpretor = (_f = data.interpreters) === null || _f === void 0 ? void 0 : _f.some(interpreter => firstLine.match(`\\b${interpreter}\\b`));
+                // Check modeline declaration
+                const matchesLang = firstLine.toLowerCase().match(langMatcher(lang));
+                const matchesAlias = (_g = data.aliases) === null || _g === void 0 ? void 0 : _g.some(lang => firstLine.toLowerCase().match(langMatcher(lang)));
+                // Add language
+                if (opts.checkShebang && matchesInterpretor)
+                    matches.push(lang);
+                if (opts.checkModeline && (matchesLang || matchesAlias))
+                    matches.push(lang);
+            }
             if (matches.length) {
                 // Add explicitly-identified language
-                const forcedLang = matches[0][0];
+                const forcedLang = matches[0];
                 addResult(file, forcedLang);
+                definiteness[file] = true;
+                continue;
             }
         }
         // Check override for manual language classification
@@ -168,6 +192,7 @@ async function analyse(input, opts = {}) {
             if (match) {
                 const forcedLang = match[1];
                 addResult(file, forcedLang);
+                definiteness[file] = true;
                 continue;
             }
         }
@@ -175,7 +200,7 @@ async function analyse(input, opts = {}) {
         let skipExts = false;
         for (const lang in langData) {
             // Check if filename is a match
-            const matchesName = (_f = langData[lang].filenames) === null || _f === void 0 ? void 0 : _f.some(name => path_1.default.basename(file.toLowerCase()) === name.toLowerCase());
+            const matchesName = (_h = langData[lang].filenames) === null || _h === void 0 ? void 0 : _h.some(name => path_1.default.basename(file.toLowerCase()) === name.toLowerCase());
             if (matchesName) {
                 addResult(file, lang);
                 skipExts = true;
@@ -184,7 +209,7 @@ async function analyse(input, opts = {}) {
         if (!skipExts)
             for (const lang in langData) {
                 // Check if extension is a match
-                const matchesExt = (_g = langData[lang].extensions) === null || _g === void 0 ? void 0 : _g.some(ext => file.toLowerCase().endsWith(ext.toLowerCase()));
+                const matchesExt = (_j = langData[lang].extensions) === null || _j === void 0 ? void 0 : _j.some(ext => file.toLowerCase().endsWith(ext.toLowerCase()));
                 if (matchesExt)
                     addResult(file, lang);
             }
@@ -194,6 +219,11 @@ async function analyse(input, opts = {}) {
     }
     // Narrow down file associations to the best fit
     for (const file in fileAssociations) {
+        // Skip if file has explicit association
+        if (definiteness[file]) {
+            results.files.results[file] = fileAssociations[file][0];
+            continue;
+        }
         // Skip binary files
         if (!useRawContent && !opts.keepBinary) {
             const isCustomText = customText.some(path => RegExp(path).test(file));
@@ -217,7 +247,7 @@ async function analyse(input, opts = {}) {
                         heuristic.language = heuristic.language[0];
                     }
                     // Make sure the results includes this language
-                    const languageGroup = (_h = langData[heuristic.language]) === null || _h === void 0 ? void 0 : _h.group;
+                    const languageGroup = (_k = langData[heuristic.language]) === null || _k === void 0 ? void 0 : _k.group;
                     const matchesLang = fileAssociations[file].includes(heuristic.language);
                     const matchesParent = languageGroup && fileAssociations[file].includes(languageGroup);
                     if (!matchesLang && !matchesParent)
@@ -240,10 +270,10 @@ async function analyse(input, opts = {}) {
                 }
             }
         // If no heuristics, assign a language
-        (_j = (_r = results.files.results)[file]) !== null && _j !== void 0 ? _j : (_r[file] = fileAssociations[file][0]);
+        (_l = (_t = results.files.results)[file]) !== null && _l !== void 0 ? _l : (_t[file] = fileAssociations[file][0]);
     }
     // Skip specified categories
-    if ((_k = opts.categories) === null || _k === void 0 ? void 0 : _k.length) {
+    if ((_m = opts.categories) === null || _m === void 0 ? void 0 : _m.length) {
         const categories = ['data', 'markup', 'programming', 'prose'];
         const hiddenCategories = categories.filter(cat => !opts.categories.includes(cat));
         for (const [file, lang] of Object.entries(results.files.results)) {
@@ -275,21 +305,21 @@ async function analyse(input, opts = {}) {
     for (const [file, lang] of Object.entries(results.files.results)) {
         if (lang && !langData[lang])
             continue;
-        const fileSize = (_o = (_m = (_l = opts.fileContent) === null || _l === void 0 ? void 0 : _l[files.indexOf(file)]) === null || _m === void 0 ? void 0 : _m.length) !== null && _o !== void 0 ? _o : fs_1.default.statSync(file).size;
+        const fileSize = (_q = (_p = (_o = opts.fileContent) === null || _o === void 0 ? void 0 : _o[files.indexOf(file)]) === null || _p === void 0 ? void 0 : _p.length) !== null && _q !== void 0 ? _q : fs_1.default.statSync(file).size;
         results.files.bytes += fileSize;
         // If no language found, add extension in other section
         if (!lang) {
             const ext = path_1.default.extname(file);
             const unknownType = ext === '' ? 'filenames' : 'extensions';
             const name = ext === '' ? path_1.default.basename(file) : ext;
-            (_p = (_s = results.unknown[unknownType])[name]) !== null && _p !== void 0 ? _p : (_s[name] = 0);
+            (_r = (_u = results.unknown[unknownType])[name]) !== null && _r !== void 0 ? _r : (_u[name] = 0);
             results.unknown[unknownType][name] += fileSize;
             results.unknown.bytes += fileSize;
             continue;
         }
         // Add language and bytes data to corresponding section
         const { type } = langData[lang];
-        (_q = (_t = results.languages.results)[lang]) !== null && _q !== void 0 ? _q : (_t[lang] = { type, bytes: 0, color: langData[lang].color });
+        (_s = (_v = results.languages.results)[lang]) !== null && _s !== void 0 ? _s : (_v[lang] = { type, bytes: 0, color: langData[lang].color });
         if (opts.childLanguages)
             results.languages.results[lang].parent = langData[lang].group;
         results.languages.results[lang].bytes += fileSize;

package/dist/types.d.ts

@@ -18,6 +18,7 @@ export interface Options {
     checkAttributes?: boolean;
     checkHeuristics?: boolean;
     checkShebang?: boolean;
+    checkModeline?: boolean;
 }
 export interface Results {
     files: {

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "linguist-js",
-	"version": "2.3.2",
+	"version": "2.4.0",
 	"description": "Analyse languages used in a folder. Powered by GitHub Linguist, although it doesn't need to be installed.",
 	"main": "dist/index.js",
 	"bin": {

package/readme.md

@@ -113,21 +113,26 @@ const { files, languages, unknown } = linguist(folder, options);
       Whether to display sub-languages instead of their parents when possible (defaults to `false`).
     - `quick` (boolean):
       Whether to skip complex language analysis such as the checking of heuristics and gitattributes statements (defaults to `false`).
-      Alias for `checkAttributes:false, checkIgnored:false, checkHeuristics:false, checkShebang:false`.
+      Alias for `checkAttributes:false, checkIgnored:false, checkHeuristics:false, checkShebang:false, checkModeline:false`.
     - `keepVendored` (boolean):
       Whether to keep vendored files (dependencies, etc) (defaults to `false`).
+      Does nothing when `fileContent` is set.
     - `keepBinary` (boolean):
       Whether binary files should be included in the output (defaults to `false`).
     - `relativePaths` (boolean):
       Change the absolute file paths in the output to be relative to the current working directory (defaults to `false`).
     - `checkAttributes` (boolean):
       Force the checking of `.gitattributes` files (defaults to `true` unless `quick` is set).
+      Does nothing when `fileContent` is set.
     - `checkIgnored` (boolean):
       Force the checking of `.gitignore` files (defaults to `true` unless `quick` is set).
+      Does nothing when `fileContent` is set.
     - `checkHeuristics` (boolean):
       Apply heuristics to ambiguous languages (defaults to `true` unless `quick` is set).
     - `checkShebang` (boolean):
       Check shebang (`#!`) lines for explicit language classification (defaults to `true` unless `quick` is set).
+    - `checkModeline` (boolean):
+      Check modelines for explicit language classification (defaults to `true` unless `quick` is set).
 
 ### Command-line
 
@@ -155,7 +160,7 @@ linguist --help
     Requires `--json` to be specified.
   - `--quick`:
     Whether to skip the checking of `.gitattributes` and `.gitignore` files for manual language classifications.
-    Alias for `--checkAttributes=false --checkIgnored=false --checkHeuristics=false --checkShebang=false`.
+    Alias for `--checkAttributes=false --checkIgnored=false --checkHeuristics=false --checkExplicit=false --checkModeline=false`.
   - `--keepVendored`:
     Whether to include vendored files (auto-generated files, dependencies folder, etc).
   - `--keepBinary`:
@@ -170,6 +175,8 @@ linguist --help
     Apply heuristics to ambiguous languages (use alongside `--quick` to overwrite).
   - `--checkShebang`:
     Check shebang (`#!`) lines for explicit classification (use alongside `--quick` to overwrite).
+  - `--checkModeline`:
+    Check modelines for explicit classification (use alongside `--quick` to overwrite).
 - `--help`:
   Display a help message.
 - `--version`: