Lookahead to ignore path separators in ranges which span path separators in fnmatch._translate

Author: dmitrin9

Lib/fnmatch.py

@@ -87,6 +87,7 @@ def _translate(pat, star, question_mark):
     res = []
     add = res.append
     star_indices = []
+    inside_range=False
 
     i, n = 0, len(pat)
     while i < n:
@@ -135,18 +136,31 @@ def _translate(pat, star, question_mark):
                         if chunks[k-1][-1] > chunks[k][0]:
                             chunks[k-1] = chunks[k-1][:-1] + chunks[k][1:]
                             del chunks[k]
+
+                    if len(chunks)>1:
+                        char_range=set(range(ord(chunks[0][-1]), ord(chunks[-1][0])))
+
+                        question_mark_char=question_mark.replace('\\', '').replace('[', '').replace(']', '').replace('^', '')
+                        question_mark_char=set(map(ord, question_mark_char))
+
+                        if question_mark_char.intersection(char_range):
+                            inside_range=True
+
                     # Escape backslashes and hyphens for set difference (--).
                     # Hyphens that create ranges shouldn't be escaped.
                     stuff = '-'.join(s.replace('\\', r'\\').replace('-', r'\-')
                                      for s in chunks)
+
                 i = j+1
                 if not stuff:
                     # Empty range: never match.
                     add('(?!)')
                 elif stuff == '!':
                     # Negated empty range: match any character.
-                    add('.')
+                    add(question_mark)
                 else:
+                    if question_mark != '.' and inside_range:
+                        add(f'(?={question_mark})')
                     # Escape set operations (&&, ~~ and ||).
                     stuff = _re_setops_sub(r'\\\1', stuff)
                     if stuff[0] == '!':

Lib/glob.py

@@ -263,55 +263,6 @@ def escape(pathname):
 _dir_open_flags = os.O_RDONLY | getattr(os, 'O_DIRECTORY', 0)
 _no_recurse_symlinks = object()
 
-def escape_regex_range_including_seps(pat, seps):
-    """Escape ranges containing seperators in a path
-    """
-    pat = list(pat)
-    ordinal_seps=set(map(ord, seps))
-
-    insideRange = False
-    ds=[]
-
-    buf=''
-    idx1=0
-    idx2=0
-    rangeIncludesSep=False
-
-    for path_idx, path_ch in enumerate(pat):
-        if path_idx > 0:
-            if path_ch == '[' and pat[path_idx-1] != '\\':
-                insideRange = True
-                idx1=path_idx
-                continue
-            if path_ch == ']' and pat[path_idx-1] != '\\':
-                insideRange = False
-                idx2=path_idx+1
-
-        if insideRange:
-            buf+=path_ch
-            if path_ch == '-':
-                glob_range = list(range(ord(pat[path_idx-1]), ord(pat[path_idx+1])))
-                if ordinal_seps.intersection(glob_range):
-                    rangeIncludesSep = True
-
-        elif len(buf)>0:
-            ds.append([idx1, idx2, rangeIncludesSep])
-
-            buf=''
-            idx1=1
-            idx2=2
-            rangeIncludesSep=False
-
-    for ds_idx, ds_elem in enumerate(ds):
-        idx1=ds_elem[0]
-        idx2=ds_elem[1]
-        rangeIncludesSep=ds_elem[2]
-        if rangeIncludesSep:
-            pat.insert(idx1, '\\')
-            pat.insert(idx2, '\\')
-
-    return ''.join(pat)
-
 def translate(pat, *, recursive=False, include_hidden=False, seps=None):
     """Translate a pathname with shell wildcards to a regular expression.
 
@@ -369,7 +320,6 @@ def translate(pat, *, recursive=False, include_hidden=False, seps=None):
                 results.append(any_sep)
 
     res = ''.join(results)
-    res=escape_regex_range_including_seps(res, seps=seps)
     return fr'(?s:{res})\Z'
 
 

Lib/test/test_glob.py

@@ -514,8 +514,7 @@ def fn(pat):
         self.assertEqual(fn('foo/bar\\baz'), r'(?s:foo[/\\]bar[/\\]baz)\Z')
         self.assertEqual(fn('**/*'), r'(?s:(?:.+[/\\])?[^/\\]+)\Z')
 
-        self.assertEqual(fn('foo[%-0]bar'), r'(?s:foo\[%-0\]bar)\Z')
-        self.assertEqual(fn('foo[U-d]bar'), r'(?s:foo\[U-d\]bar)\Z')
+        self.assertEqual(fn('foo[%-0]bar'), r'(?s:foo(?=[^/\\])[%-0]bar)\Z')
 
 
 if __name__ == "__main__":