feat: introduce .convertPathToPattern method

mrmlnc · May 9, 2023 · 6b0624a · 6b0624a
1 parent 00cf05b
commit 6b0624a
Show file tree

Hide file tree

Showing 5 changed files with 211 additions and 27 deletions.
diff --git a/README.md b/README.md
@@ -26,6 +26,7 @@ This package provides methods for traversing the file system and returning pathn
     * [generateTasks](#generatetaskspatterns-options)
     * [isDynamicPattern](#isdynamicpatternpattern-options)
     * [escapePath](#escapepathpath)
+	* [convertPathToPattern](#convertpathtopatternpath)
 * [Options](#options-3)
   * [Common](#common)
     * [concurrency](#concurrency)
@@ -291,6 +292,31 @@ fg.win32.escapePath('C:\\Program Files (x86)\\**\\*');
 // Windows: C:\\Program Files \\(x86\\)\\**\\*
 ```
 
+#### `convertPathToPattern(path)`
+
+Converts a path to a pattern depending on the platform, including special character escaping.
+
+* Posix. Works similarly to the `fg.posix.escapePath` method.
+* Windows. Works similarly to the `fg.win32.escapePath` method, additionally converting backslashes to forward slashes in cases where they are not escape characters (`!()+@{}`).
+
+```js
+fg.convertPathToPattern('[OpenSource] mrmlnc – fast-glob (Deluxe Edition) 2014') + '/*.flac';
+// \\[OpenSource\\] mrmlnc – fast-glob \\(Deluxe Edition\\) 2014/*.flac
+
+fg.convertPathToPattern('C:/Program Files (x86)/**/*');
+// Posix: C:/Program Files \\(x86\\)/\\*\\*/\\*
+// Windows: C:/Program Files \\(x86\\)/**/*
+
+fg.convertPathToPattern('C:\\Program Files (x86)\\**\\*');
+// Posix: C:\\\\Program Files \\(x86\\)\\*\\*\\*
+// Windows: C:/Program Files \\(x86\\)/**/*
+
+fg.posix.convertPathToPattern('\\\\?\\c:\\Program Files (x86)') + '/**/*';
+// Posix: \\\\\\?\\\\c:\\\\Program Files \\(x86\\)/**/* (broken pattern)
+fg.win32.convertPathToPattern('\\\\?\\c:\\Program Files (x86)') + '/**/*';
+// Windows: //?/c:/Program Files \\(x86\\)/**/*
+```
+
 ## Options
 
 ### Common options
@@ -680,11 +706,11 @@ Always use forward-slashes in glob expressions (patterns and [`ignore`](#ignore)
 ```ts
 [
 	'directory/*',
-	path.join(process.cwd(), '**').replace(/\\/g, '/')
+	fg.convertPathToPattern(process.cwd()) + '/**'
 ]
 ```
 
-> :book: Use the [`normalize-path`][npm_normalize_path] or the [`unixify`][npm_unixify] package to convert Windows-style path to a Unix-style path.
+> :book: Use the [`.convertPathToPattern`](#convertpathtopatternpath) package to convert Windows-style path to a Unix-style path.
 
 Read more about [matching with backslashes][micromatch_backslashes].
 
@@ -705,7 +731,7 @@ Refers to Bash. You need to escape special characters:
 fg.sync(['\\(special-*file\\).txt']) // ['(special-*file).txt']
 ```
 
-Read more about [matching special characters as literals][picomatch_matching_special_characters_as_literals].
+Read more about [matching special characters as literals][picomatch_matching_special_characters_as_literals]. Or use the [`.escapePath`](#escapepathpath).
 
 ## How to exclude directory from reading?
 
@@ -731,11 +757,15 @@ You have to understand that if you write the pattern to exclude directories, the
 
 ## How to use UNC path?
 
-You cannot use [Uniform Naming Convention (UNC)][unc_path] paths as patterns (due to syntax), but you can use them as [`cwd`](#cwd) directory.
+You cannot use [Uniform Naming Convention (UNC)][unc_path] paths as patterns (due to syntax) directly, but you can use them as [`cwd`](#cwd) directory or use the `fg.convertPathToPattern` method.
 
 ```ts
+// cwd
 fg.sync('*', { cwd: '\\\\?\\C:\\Python27' /* or //?/C:/Python27 */ });
 fg.sync('Python27/*', { cwd: '\\\\?\\C:\\' /* or //?/C:/ */ });
+
+// .convertPathToPattern
+fg.sync(fg.convertPathToPattern('\\\\?\\c:\\Python27') + '/*');
 ```
 
 ## Compatible with `node-glob`?

diff --git a/src/index.spec.ts b/src/index.spec.ts
@@ -221,7 +221,20 @@ describe('Package', () => {
 		});
 	});
 
-	describe('.posix', () => {
+	describe('.convertPathToPattern', () => {
+		it('should return a pattern', () => {
+			// In posix system \\ is a escaping character and it will be escaped before non-special characters.
+			const posix = 'C:\\\\Program Files \\(x86\\)\\*\\*\\*';
+			const windows = 'C:/Program Files \\(x86\\)/**/*';
+			const expected = tests.platform.isWindows() ? windows : posix;
+
+			const actual = fg.convertPathToPattern('C:\\Program Files (x86)\\**\\*');
+
+			assert.strictEqual(actual, expected);
+		});
+	});
+
+	describe('posix', () => {
 		describe('.escapePath', () => {
 			it('should return escaped path', () => {
 				const expected = '/directory/\\*\\*/\\*';
@@ -231,9 +244,19 @@ describe('Package', () => {
 				assert.strictEqual(actual, expected);
 			});
 		});
+
+		describe('.convertPathToPattern', () => {
+			it('should return a pattern', () => {
+				const expected = 'a\\*.txt';
+
+				const actual = fg.posix.convertPathToPattern('a\\*.txt');
+
+				assert.strictEqual(actual, expected);
+			});
+		});
 	});
 
-	describe('.win32', () => {
+	describe('win32', () => {
 		describe('.escapePath', () => {
 			it('should return escaped path', () => {
 				const expected = 'C:\\Program Files \\(x86\\)\\**\\*';
@@ -243,5 +266,15 @@ describe('Package', () => {
 				assert.strictEqual(actual, expected);
 			});
 		});
+
+		describe('.convertPathToPattern', () => {
+			it('should return a pattern', () => {
+				const expected = 'C:/Program Files \\(x86\\)/**/*';
+
+				const actual = fg.win32.convertPathToPattern('C:\\Program Files (x86)\\**\\*');
+
+				assert.strictEqual(actual, expected);
+			});
+		});
 	});
 });
diff --git a/src/index.ts b/src/index.ts
@@ -79,12 +79,24 @@ namespace FastGlob {
 		return utils.path.escape(source);
 	}
 
+	export function convertPathToPattern(source: string): PatternInternal {
+		assertPatternsInput(source);
+
+		return utils.path.convertPathToPattern(source);
+	}
+
 	export namespace posix {
 		export function escapePath(source: string): PatternInternal {
 			assertPatternsInput(source);
 
 			return utils.path.escapePosixPath(source);
 		}
+
+		export function convertPathToPattern(source: string): PatternInternal {
+			assertPatternsInput(source);
+
+			return utils.path.convertPosixPathToPattern(source);
+		}
 	}
 
 	export namespace win32 {
@@ -93,6 +105,12 @@ namespace FastGlob {
 
 			return utils.path.escapeWindowsPath(source);
 		}
+
+		export function convertPathToPattern(source: string): PatternInternal {
+			assertPatternsInput(source);
+
+			return utils.path.convertWindowsPathToPattern(source);
+		}
 	}
 }
 

diff --git a/src/utils/path.spec.ts b/src/utils/path.spec.ts
@@ -24,26 +24,7 @@ describe('Utils → Path', () => {
 		});
 	});
 
-	describe('.removeLeadingDotCharacters', () => {
-		it('should return path without changes', () => {
-			assert.strictEqual(util.removeLeadingDotSegment('../a/b'), '../a/b');
-			assert.strictEqual(util.removeLeadingDotSegment('~/a/b'), '~/a/b');
-			assert.strictEqual(util.removeLeadingDotSegment('/a/b'), '/a/b');
-			assert.strictEqual(util.removeLeadingDotSegment('a/b'), 'a/b');
-
-			assert.strictEqual(util.removeLeadingDotSegment('..\\a\\b'), '..\\a\\b');
-			assert.strictEqual(util.removeLeadingDotSegment('~\\a\\b'), '~\\a\\b');
-			assert.strictEqual(util.removeLeadingDotSegment('\\a\\b'), '\\a\\b');
-			assert.strictEqual(util.removeLeadingDotSegment('a\\b'), 'a\\b');
-		});
-
-		it('should return path without leading dit characters', () => {
-			assert.strictEqual(util.removeLeadingDotSegment('./a/b'), 'a/b');
-			assert.strictEqual(util.removeLeadingDotSegment('.\\a\\b'), 'a\\b');
-		});
-	});
-
-	describe('.escapePattern', () => {
+	describe('.escape', () => {
 		it('should return pattern without additional escape characters', () => {
 			assert.strictEqual(util.escape('\\!abc'), '\\!abc');
 			assert.strictEqual(util.escape('\\*'), '\\*');
@@ -87,4 +68,103 @@ describe('Utils → Path', () => {
 			assert.strictEqual(util.escapeWindowsPath('+('), '\\+\\(');
 		});
 	});
+
+	describe('.removeLeadingDotCharacters', () => {
+		it('should return path without changes', () => {
+			assert.strictEqual(util.removeLeadingDotSegment('../a/b'), '../a/b');
+			assert.strictEqual(util.removeLeadingDotSegment('~/a/b'), '~/a/b');
+			assert.strictEqual(util.removeLeadingDotSegment('/a/b'), '/a/b');
+			assert.strictEqual(util.removeLeadingDotSegment('a/b'), 'a/b');
+
+			assert.strictEqual(util.removeLeadingDotSegment('..\\a\\b'), '..\\a\\b');
+			assert.strictEqual(util.removeLeadingDotSegment('~\\a\\b'), '~\\a\\b');
+			assert.strictEqual(util.removeLeadingDotSegment('\\a\\b'), '\\a\\b');
+			assert.strictEqual(util.removeLeadingDotSegment('a\\b'), 'a\\b');
+		});
+
+		it('should return path without leading dit characters', () => {
+			assert.strictEqual(util.removeLeadingDotSegment('./a/b'), 'a/b');
+			assert.strictEqual(util.removeLeadingDotSegment('.\\a\\b'), 'a\\b');
+		});
+	});
+
+	describe('.convertPathToPattern', () => {
+		it('should return a pattern', () => {
+			assert.strictEqual(util.convertPathToPattern('.{directory}'), '.\\{directory\\}');
+		});
+	});
+
+	describe('.convertPosixPathToPattern', () => {
+		it('should escape special characters', () => {
+			assert.strictEqual(util.convertPosixPathToPattern('./**\\*'), './\\*\\*\\*');
+		});
+	});
+
+	describe('.convertWindowsPathToPattern', () => {
+		it('should escape special characters', () => {
+			assert.strictEqual(util.convertPosixPathToPattern('.{directory}'), '.\\{directory\\}');
+		});
+
+		it('should do nothing with escaped glob symbols', () => {
+			assert.strictEqual(util.convertWindowsPathToPattern('\\!\\'), '\\!/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\+\\'), '\\+/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\@\\'), '\\@/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\(\\'), '\\(/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\)\\'), '\\)/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\{\\'), '\\{/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\}\\'), '\\}/');
+
+			assert.strictEqual(util.convertWindowsPathToPattern('.\\*'), './*');
+			assert.strictEqual(util.convertWindowsPathToPattern('.\\**'), './**');
+			assert.strictEqual(util.convertWindowsPathToPattern('.\\**\\*'), './**/*');
+
+			assert.strictEqual(util.convertWindowsPathToPattern('a\\{b,c\\d,{b,c}}'), 'a\\{b,c/d,\\{b,c\\}\\}');
+		});
+
+		it('should convert slashes', () => {
+			assert.strictEqual(util.convertWindowsPathToPattern('/'), '/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\'), '/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\\\'), '//');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\/'), '//');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\/\\'), '///');
+		});
+
+		it('should convert relative paths', () => {
+			assert.strictEqual(util.convertWindowsPathToPattern('file.txt'), 'file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('./file.txt'), './file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('.\\file.txt'), './file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('../file.txt'), '../file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('..\\file.txt'), '../file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('.\\file.txt'), './file.txt');
+		});
+
+		it('should convert absolute paths', () => {
+			assert.strictEqual(util.convertWindowsPathToPattern('/.file.txt'), '/.file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('/root/.file.txt'), '/root/.file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\.file.txt'), '/.file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\root\\.file.txt'), '/root/.file.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\root/.file.txt'), '/root/.file.txt');
+		});
+
+		it('should convert traditional DOS paths', () => {
+			assert.strictEqual(util.convertWindowsPathToPattern('D:ShipId.txt'), 'D:ShipId.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('D:/ShipId.txt'), 'D:/ShipId.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('D://ShipId.txt'), 'D://ShipId.txt');
+
+			assert.strictEqual(util.convertWindowsPathToPattern('D:\\ShipId.txt'), 'D:/ShipId.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('D:\\\\ShipId.txt'), 'D://ShipId.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('D:\\/ShipId.txt'), 'D://ShipId.txt');
+		});
+
+		it('should convert UNC paths', () => {
+			assert.strictEqual(util.convertWindowsPathToPattern('\\\\system07\\'), '//system07/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\\\system07\\c$\\'), '//system07/c$/');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\\\Server02\\Share\\Foo.txt'), '//Server02/Share/Foo.txt');
+
+			assert.strictEqual(util.convertWindowsPathToPattern('\\\\127.0.0.1\\c$\\File.txt'), '//127.0.0.1/c$/File.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\\\.\\c:\\File.txt'), '//./c:/File.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\\\?\\c:\\File.txt'), '//?/c:/File.txt');
+			assert.strictEqual(util.convertWindowsPathToPattern('\\\\.\\UNC\\LOCALHOST\\c$\\File.txt'), '//./UNC/LOCALHOST/c$/File.txt');
+		});
+	});
 });
diff --git a/src/utils/path.ts b/src/utils/path.ts
@@ -5,13 +5,24 @@ import { Pattern } from '../types';
 
 const IS_WINDOWS_PLATFORM = os.platform() === 'win32';
 const LEADING_DOT_SEGMENT_CHARACTERS_COUNT = 2; // ./ or .\\
-/*
+/**
  * All non-escaped special characters.
  * Posix: ()*?[\]{|}, !+@ before (, ! at the beginning, \\ before non-special characters.
  * Windows: (){}, !+@ before (, ! at the beginning.
  */
 const POSIX_UNESCAPED_GLOB_SYMBOLS_RE = /(\\?)([()*?[\]{|}]|^!|[!+@](?=\()|\\(?![!()*+?@[\]{|}]))/g;
 const WINDOWS_UNESCAPED_GLOB_SYMBOLS_RE = /(\\?)([(){}]|^!|[!+@](?=\())/g;
+/**
+ * The device path (\\.\ or \\?\).
+ * https://learn.microsoft.com/en-us/dotnet/standard/io/file-path-formats#dos-device-paths
+ */
+const DOS_DEVICE_PATH_RE = /^\\\\([.?])/;
+/**
+ * All backslashes except those escaping special characters.
+ * Windows: !()+@{}
+ * https://learn.microsoft.com/en-us/windows/win32/fileio/naming-a-file#naming-conventions
+ */
+const WINDOWS_BACKSLASHES_RE = /\\(?![!()+@{}])/g;
 
 /**
  * Designed to work only with simple paths: `dir\\file`.
@@ -47,3 +58,15 @@ export function escapeWindowsPath(pattern: Pattern): Pattern {
 export function escapePosixPath(pattern: Pattern): Pattern {
 	return pattern.replace(POSIX_UNESCAPED_GLOB_SYMBOLS_RE, '\\$2');
 }
+
+export const convertPathToPattern = IS_WINDOWS_PLATFORM ? convertWindowsPathToPattern : convertPosixPathToPattern;
+
+export function convertWindowsPathToPattern(filepath: string): Pattern {
+	return escapeWindowsPath(filepath)
+		.replace(DOS_DEVICE_PATH_RE, '//$1')
+		.replace(WINDOWS_BACKSLASHES_RE, '/');
+}
+
+export function convertPosixPathToPattern(filepath: string): Pattern {
+	return escapePosixPath(filepath);
+}