Merge pull request #31 from chetan/fix/unicode-decode-err

feat: update mtime script (release 2025.08)

Author: chetan

README.md

@@ -1,6 +1,6 @@
 # git-restore-mtime action
 
-A GitHub Workflow Action which restores timestamps of files in the current tree based on their last commit times. Uses the [git-restore-mtime](https://github.com/MestreLion/git-tools) script (v2022.12 release) by [@MestreLion](https://github.com/MestreLion).
+A GitHub Workflow Action which restores timestamps of files in the current tree based on their last commit times. Uses the [git-restore-mtime](https://github.com/MestreLion/git-tools) script (v2025.08 release) by [@MestreLion](https://github.com/MestreLion).
 
 ## Usage
 

git-restore-mtime

@@ -30,8 +30,7 @@ assuming the actual modification date and its commit date are close.
 """
 
 # TODO:
-# - Add -z on git whatchanged/ls-files, so we don't deal with filename decoding
-# - When Python is bumped to 3.7, use text instead of universal_newlines on subprocess
+# - Add -z on git log/ls-files, so we don't deal with filename decoding
 # - Update "Statistics for some large projects" with modern hardware and repositories.
 # - Create a README.md for git-restore-mtime alone. It deserves extensive documentation
 #   - Move Statistics there
@@ -76,10 +75,13 @@ import subprocess
 import sys
 import time
 
-__version__ = "2022.12"
+if sys.version_info < (3, 8):
+    sys.exit("Python 3.8 or later required.")
+
+__version__ = "2025.08"
 
 # Update symlinks only if the platform supports not following them
-UPDATE_SYMLINKS = bool(os.utime in getattr(os, 'supports_follow_symlinks', []))
+UPDATE_SYMLINKS = bool({os.utime, os.stat} <= getattr(os, 'supports_follow_symlinks', set()))
 
 # Call os.path.normpath() only if not in a POSIX platform (Windows)
 NORMALIZE_PATHS = (os.path.sep != '/')
@@ -241,17 +243,23 @@ def normalize(path):
     normalize(r'"Back\\slash_double\"quote_a\303\247a\303\255"') =>
         r'Back\slash_double"quote_açaí')
 
-    See notes on `windows/non-ascii-paths.txt` about path encodings on non-UTF-8
-    platforms and filesystems.
+    Paths with invalid UTF-8 encoding, such as single 0x80-0xFF bytes (e.g, from
+    Latin1/Windows-1251 encoding) are decoded using surrogate escape, the same
+    method used by Python for filesystem paths. So 0xE6 ("æ" in Latin1, r'\\346'
+    from Git) is decoded as "\udce6". See https://peps.python.org/pep-0383/ and
+    https://vstinner.github.io/painful-history-python-filesystem-encoding.html
+
+    Also see notes on `windows/non-ascii-paths.txt` about path encodings on
+    non-UTF-8 platforms and filesystems.
     """
     if path and path[0] == '"':
         # Python 2: path = path[1:-1].decode("string-escape")
         # Python 3: https://stackoverflow.com/a/46650050/624066
         path = (path[1:-1]                 # Remove enclosing double quotes
                 .encode('latin1')          # Convert to bytes, required by 'unicode-escape'
                 .decode('unicode-escape')  # Perform the actual octal-escaping decode
-                .encode('latin1')          # 1:1 mapping to bytes, forming UTF-8 encoding
-                .decode('utf8'))           # Decode from UTF-8
+                .encode('latin1')          # 1:1 mapping to bytes, UTF-8 encoded
+                .decode('utf8', 'surrogateescape'))  # Decode from UTF-8
     if NORMALIZE_PATHS:
         # Make sure the slash matches the OS; for Windows we need a backslash
         path = os.path.normpath(path)
@@ -292,7 +300,7 @@ def get_mtime_ns(secs: int, idx: int):
 
 
 def get_mtime_path(path):
-    return os.path.getmtime(path)
+    return os.stat(path, **UTIME_KWS).st_mtime
 
 
 # Git class and parse_log(), the heart of the script ##########################
@@ -318,8 +326,8 @@ class Git:
 
     def log(self, merge=False, first_parent=False, commit_time=False,
             reverse_order=False, paths: list = None):
-        cmd = 'log --raw --pretty={}'.format('%ct' if commit_time else '%at')
-        if merge:         cmd += ' -m'
+        cmd = 'log --raw --no-show-signature --pretty={}'.format('%ct' if commit_time else '%at')
+        cmd += ' -m' if merge else ' --no-merges'
         if first_parent:  cmd += ' --first-parent'
         if reverse_order: cmd += ' --reverse'
         return self._run(cmd, paths)
@@ -345,7 +353,7 @@ class Git:
         if paths:
             cmdlist.append('--')
             cmdlist.extend(paths)
-        popen_args = dict(universal_newlines=True, encoding='utf8')
+        popen_args = dict(text=True, encoding='utf8')
         if not self.errors:
             popen_args['stderr'] = subprocess.DEVNULL
         log.trace("Executing: %s", ' '.join(cmdlist))