Add text file convention (#183)

Author: SimonCropp

readme.md

@@ -20,6 +20,7 @@ A collection of minimal binary files.
 
 All files: https://github.com/VerifyTests/EmptyFiles/tree/main/files
 
+
 <!-- include: extensions. path: /src/extensions.include.md -->
 ### Archive
 
@@ -37,7 +38,7 @@ All files: https://github.com/VerifyTests/EmptyFiles/tree/main/files
 
   * .docx (1.9 KB)
   * .odt (2.2 KB)
-  * .pdf (291 bytes)
+  * .pdf (280 bytes)
   * .rtf (6 bytes)
 
 ### Image
@@ -54,13 +55,13 @@ All files: https://github.com/VerifyTests/EmptyFiles/tree/main/files
   * .ico (70 bytes)
   * .j2c (270 bytes)
   * .jfif (734 bytes)
-  * .jp2 (357 bytes)
+  * .jp2 (354 bytes)
   * .jpc (270 bytes)
   * .jpe (734 bytes)
   * .jpeg (734 bytes)
   * .jpg (734 bytes)
   * .jxr (300 bytes)
-  * .pbm (10 bytes)
+  * .pbm (8 bytes)
   * .pcx (131 bytes)
   * .pgm (12 bytes)
   * .png (119 bytes)
@@ -253,7 +254,7 @@ FileExtensions.AddTextExtension(".ext1");
 True(FileExtensions.IsTextExtension(".ext1"));
 True(FileExtensions.IsTextFile("file.ext1"));
 ```
-<sup><a href='/src/Tests/ExtensionsTests.cs#L36-L42' title='Snippet source file'>snippet source</a> | <a href='#snippet-AddTextExtension' title='Start of snippet'>anchor</a></sup>
+<sup><a href='/src/Tests/ExtensionsTests.cs#L50-L56' title='Snippet source file'>snippet source</a> | <a href='#snippet-AddTextExtension' title='Start of snippet'>anchor</a></sup>
 <!-- endSnippet -->
 
 
@@ -267,10 +268,40 @@ True(FileExtensions.IsTextExtension(".ext1"));
 FileExtensions.RemoveTextExtension(".ext1");
 False(FileExtensions.IsTextExtension(".ext1"));
 ```
-<sup><a href='/src/Tests/ExtensionsTests.cs#L52-L59' title='Snippet source file'>snippet source</a> | <a href='#snippet-RemoveTextExtension' title='Start of snippet'>anchor</a></sup>
+<sup><a href='/src/Tests/ExtensionsTests.cs#L66-L73' title='Snippet source file'>snippet source</a> | <a href='#snippet-RemoveTextExtension' title='Start of snippet'>anchor</a></sup>
 <!-- endSnippet -->
 
 
+#### AddTextFileConvention
+
+`AddTextFileConvention` allows the use of a convention based text file detection via a callback.
+
+At app startup add a convention using `FileExtensions.AddTextFileConvention`:
+
+<!-- snippet: AddTextFileConvention -->
+<a id='snippet-AddTextFileConvention'></a>
+```cs
+[ModuleInitializer]
+public static void AddTextFileConvention() =>
+    // Treat files ending with .txtViaConvention as text files
+    FileExtensions.AddTextFileConvention(path => path.EndsWith(".txtViaConvention"));
+```
+<sup><a href='/src/Tests/ExtensionsTests.cs#L24-L30' title='Snippet source file'>snippet source</a> | <a href='#snippet-AddTextFileConvention' title='Start of snippet'>anchor</a></sup>
+<!-- endSnippet -->
+
+Then any call to `FileExtensions.IsTextFile` will, in addition to checking the known text extensions, also check if any of the added text contentions return true.
+
+<!-- snippet: TextViaConvention -->
+<a id='snippet-TextViaConvention'></a>
+```cs
+True(FileExtensions.IsTextFile("c:/path/file.txtViaConvention"));
+```
+<sup><a href='/src/Tests/ExtensionsTests.cs#L17-L21' title='Snippet source file'>snippet source</a> | <a href='#snippet-TextViaConvention' title='Start of snippet'>anchor</a></sup>
+<!-- endSnippet -->
+
+An alternative approach to a text file convention would be to check if a file has a preamble that matches an known text encoding.
+
+
 ## Icon
 
 [Hollow](https://thenounproject.com/term/hollow/51835/) designed by [Michael Senkow](https://thenounproject.com/mhsenkow/) from [The Noun Project](https://thenounproject.com).

src/EmptyFiles.sln

@@ -11,6 +11,7 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Solution Items", "Solution
 		Directory.Packages.props = Directory.Packages.props
 		..\readme.md = ..\readme.md
 		global.json = global.json
+		mdsnippets.json = mdsnippets.json
 	EndProjectSection
 EndProject
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "EmptyFiles", "EmptyFiles\EmptyFiles.csproj", "{A017D3FD-7DA5-4DF4-A169-72DE4AF76048}"

src/EmptyFiles/FileExtensions.cs

@@ -13,15 +13,24 @@ public static bool IsTextExtension(CharSpan extension) =>
         IsTextExtension(extension.ToString());
 
     public static bool IsTextFile(string path) =>
-        IsTextExtension(Path.GetExtension(path));
+        IsTextFile(path.AsSpan());
 
     public static bool IsTextFile(CharSpan path)
     {
+        foreach (var convention in textFileConventions)
+        {
+            if (convention(path))
+            {
+                return true;
+            }
+        }
+
 #if NET6_0_OR_GREATER
         var extension = Path.GetExtension(path);
         return IsTextExtension(extension);
 #else
-        return IsTextFile(path.ToString());
+        var extension = Path.GetExtension(path.ToString());
+        return IsTextExtension(extension);
 #endif
     }
 
@@ -81,6 +90,11 @@ public static void AddTextExtensions(IEnumerable<string> extensions)
         }
     }
 
+    public static void AddTextFileConvention(IsTextFile convention) =>
+        textFileConventions.Add(convention);
+
+    static List<IsTextFile> textFileConventions = [];
+
     //From https://github.com/sindresorhus/text-extensions/blob/master/text-extensions.json
     static IReadOnlyCollection<string> textExtensions = new HashSet<string>
         {

src/EmptyFiles/IsTextFile.cs

@@ -0,0 +1,3 @@
+namespace EmptyFiles;
+
+public delegate bool IsTextFile(CharSpan path);

src/Tests/ExtensionsTests.cs

@@ -13,8 +13,22 @@ public void IsText()
         False(FileExtensions.IsTextExtension("bin"));
 
         #endregion
+
+        #region TextViaConvention
+
+        True(FileExtensions.IsTextFile("c:/path/file.txtViaConvention"));
+
+        #endregion
     }
 
+    #region AddTextFileConvention
+    [ModuleInitializer]
+    public static void AddTextFileConvention() =>
+        // Treat files ending with .txtViaConvention as text files
+        FileExtensions.AddTextFileConvention(path => path.EndsWith(".txtViaConvention"));
+
+    #endregion
+
     [Test]
     public void IsTextLegacy()
     {

src/extensions.include.md

@@ -15,7 +15,7 @@
 
   * .docx (1.9 KB)
   * .odt (2.2 KB)
-  * .pdf (291 bytes)
+  * .pdf (280 bytes)
   * .rtf (6 bytes)
 
 ### Image
@@ -32,13 +32,13 @@
   * .ico (70 bytes)
   * .j2c (270 bytes)
   * .jfif (734 bytes)
-  * .jp2 (357 bytes)
+  * .jp2 (354 bytes)
   * .jpc (270 bytes)
   * .jpe (734 bytes)
   * .jpeg (734 bytes)
   * .jpg (734 bytes)
   * .jxr (300 bytes)
-  * .pbm (10 bytes)
+  * .pbm (8 bytes)
   * .pcx (131 bytes)
   * .pgm (12 bytes)
   * .png (119 bytes)

src/mdsnippets.json

@@ -1,6 +1,6 @@
 {
   "$schema": "https://raw.githubusercontent.com/SimonCropp/MarkdownSnippets/master/schema.json",
   "TocExcludes": [ "NuGet package", "Release Notes", "Icon" ],
-  "MaxWidth": 80,
+  "MaxWidth": 100,
   "Convention": "InPlaceOverwrite"
 }