From 5c63521f2478f650400859f4fbb100db25adfadc Mon Sep 17 00:00:00 2001
From: Jean Boussier <jean.boussier@gmail.com>
Date: Mon, 19 Jan 2026 06:24:17 +0100
Subject: [PATCH 1/2] File.dirname: add a spec for Shift JIS handling

While trying to speedup various `File.*` methods, I realized they
were way slower and complicated than they should for no apparent
reason. However after asking Nobu he explained that Shift JIS encoded
text can contain `0x5C` (ASCII backslash) as the second byte of a
two byte character sequence.

Since on Windows `0x5C` is `File::ALT_SEPARATOR`, this can easily
break naive path related algorithms searching for directory separators.
---
 core/file/dirname_spec.rb | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/core/file/dirname_spec.rb b/core/file/dirname_spec.rb
index 63436e19c..d215e3e03 100644
--- a/core/file/dirname_spec.rb
+++ b/core/file/dirname_spec.rb
@@ -78,6 +78,28 @@ def object.to_int; 2; end
     File.dirname("foo/../").should == "foo"
   end
 
+  it "rejects strings encoded with non ASCII-compatible encodings" do
+    Encoding.list.reject(&:ascii_compatible?).reject(&:dummy?).each do |enc|
+      path = "/foo/bar".encode(enc)
+      -> {
+        File.dirname(path)
+      }.should raise_error(Encoding::CompatibilityError)
+    end
+  end
+
+  it "works with all ASCII-compatible encodings" do
+    Encoding.list.select(&:ascii_compatible?).each do |enc|
+      File.dirname("/foo/bar".encode(enc)).should == "/foo".encode(enc)
+    end
+  end
+
+  it "handles Shift JIS 0x5C (\\) as second byte of a multi-byte sequence" do
+    # dir/fileソname.txt
+    path = "dir/file\x83\x5cname.txt".b.force_encoding(Encoding::SHIFT_JIS)
+    path.valid_encoding?.should be_true
+    File.dirname(path).should == "dir"
+  end
+
   platform_is_not :windows do
     it "ignores repeated leading / (edge cases on non-windows)" do
       File.dirname("/////foo/bar/").should == "/foo"
@@ -98,6 +120,13 @@ def object.to_int; 2; end
       File.dirname("//foo//").should == "//foo"
       File.dirname('/////').should == '//'
     end
+
+    it "handles Shift JIS 0x5C (\\) as second byte of a multi-byte sequence (windows)" do
+      # dir/fileソname.txt
+      path = "dir\\file\x83\x5cname.txt".b.force_encoding(Encoding::SHIFT_JIS)
+      path.valid_encoding?.should be_true
+      File.dirname(path).should == "dir"
+    end
   end
 
   it "accepts an object that has a #to_path method" do

From 1724265c7317dce4666a1fd39deffc08101f6fe7 Mon Sep 17 00:00:00 2001
From: Benoit Daloze <eregontp@gmail.com>
Date: Tue, 20 Jan 2026 13:17:42 +0100
Subject: [PATCH 2/2] Update core/file/dirname_spec.rb

---
 core/file/dirname_spec.rb | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/core/file/dirname_spec.rb b/core/file/dirname_spec.rb
index d215e3e03..1b006af78 100644
--- a/core/file/dirname_spec.rb
+++ b/core/file/dirname_spec.rb
@@ -122,7 +122,7 @@ def object.to_int; 2; end
     end
 
     it "handles Shift JIS 0x5C (\\) as second byte of a multi-byte sequence (windows)" do
-      # dir/fileソname.txt
+      # dir\fileソname.txt
       path = "dir\\file\x83\x5cname.txt".b.force_encoding(Encoding::SHIFT_JIS)
       path.valid_encoding?.should be_true
       File.dirname(path).should == "dir"