`configure_file`: update \@ escape logic

When configuring a 'meson' or 'cmake@' style file,
add a case for escaped variables using matched pairs of
`\@` i.e. `\@foo\@ -> @foo@`.

The match for @var@ has been amended with a negative lookbehind
to ensure that any occurrances of `\@foo@` are not evaluated to
`\bar`.

The previous behaviour, matching `\@` and escaping only that character,
had undesirable side effects including mangling valid perl when
configuring files.

Closes: https://github.com/mesonbuild/meson/issues/7165

mesonbuild/utils/universal.py

@@ -1182,24 +1182,21 @@ def do_replacement(regex: T.Pattern[str], line: str,
                    variable_format: Literal['meson', 'cmake', 'cmake@'],
                    confdata: T.Union[T.Dict[str, T.Tuple[str, T.Optional[str]]], 'ConfigurationData']) -> T.Tuple[str, T.Set[str]]:
     missing_variables: T.Set[str] = set()
-    if variable_format == 'cmake':
-        start_tag = '${'
-        backslash_tag = '\\${'
-    else:
-        start_tag = '@'
-        backslash_tag = '\\@'
 
     def variable_replace(match: T.Match[str]) -> str:
-        # Pairs of escape characters before '@' or '\@'
+        # Pairs of escape characters before '@', '\@', '${' or '\${'
         if match.group(0).endswith('\\'):
             num_escapes = match.end(0) - match.start(0)
             return '\\' * (num_escapes // 2)
-        # Single escape character and '@'
-        elif match.group(0) == backslash_tag:
-            return start_tag
-        # Template variable to be replaced
+        # Handle cmake escaped \${} tags
+        elif variable_format == 'cmake' and match.group(0) == '\\${':
+            return '${'
+        # \@escaped\@ variables
+        elif match.groupdict().get('escaped') is not None:
+            return match.group('escaped')[1:-2]+'@'
         else:
-            varname = match.group(1)
+            # Template variable to be replaced
+            varname = match.group('variable')
             var_str = ''
             if varname in confdata:
                 var, _ = confdata.get(varname)
@@ -1280,11 +1277,23 @@ def do_define(regex: T.Pattern[str], line: str, confdata: 'ConfigurationData',
 
 def get_variable_regex(variable_format: Literal['meson', 'cmake', 'cmake@'] = 'meson') -> T.Pattern[str]:
     # Only allow (a-z, A-Z, 0-9, _, -) as valid characters for a define
-    # Also allow escaping '@' with '\@'
     if variable_format in {'meson', 'cmake@'}:
-        regex = re.compile(r'(?:\\\\)+(?=\\?@)|\\@|@([-a-zA-Z0-9_]+)@')
+        # Also allow escaping pairs of '@' with '\@'
+        regex = re.compile(r'''
+            (?:\\\\)+(?=\\?@)  # Matches multiple backslashes followed by an @ symbol
+            |                  # OR
+            (?<!\\)@(?P<variable>[-a-zA-Z0-9_]+)@  # Match a variable enclosed in @ symbols and capture the variable name; no matches beginning with '\@'
+            |                  # OR
+            (?P<escaped>\\@[-a-zA-Z0-9_]+\\@)  # Match an escaped variable enclosed in @ symbols
+        ''', re.VERBOSE)
     else:
-        regex = re.compile(r'(?:\\\\)+(?=\\?\$)|\\\${|\${([-a-zA-Z0-9_]+)}')
+        regex = re.compile(r'''
+            (?:\\\\)+(?=\\?\$)  # Match multiple backslashes followed by a dollar sign
+            |                  # OR
+            \\\${              # Match a backslash followed by a dollar sign and an opening curly brace
+            |                  # OR
+            \${(?P<variable>[-a-zA-Z0-9_]+)}  # Match a variable enclosed in curly braces and capture the variable name
+        ''', re.VERBOSE)
     return regex
 
 def do_conf_str(src: str, data: T.List[str], confdata: 'ConfigurationData',

test cases/common/14 configure file/config6.h.in

@@ -1,19 +1,40 @@
 /* No escape */
 #define MESSAGE1 "@var1@"
 
-/* Single escape means no replace */
-#define MESSAGE2 "\@var1@"
+/* Escaped whole variable */
+#define MESSAGE2 "\\@var1\\@"
 
 /* Replace pairs of escapes before '@' or '\@' with escape characters
  * (note we have to double number of pairs due to C string escaping)
  */
 #define MESSAGE3 "\\\\@var1@"
 
-/* Pairs of escapes and then single escape to avoid replace */
-#define MESSAGE4 "\\\\\@var1@"
+/* Pairs of escapes and then an escaped variable */
+#define MESSAGE4 "\\\\\@var1\@"
 
-/* Check escaped variable does not overlap following variable */
-#define MESSAGE5 "\@var1@var2@"
+/* We don't gobble \@ prefixing some text */
+#define MESSAGE5 "\\\\@var1"
 
-/* Check escape character outside variables */
-#define MESSAGE6 "\\ @ \@ \\\\@ \\\\\@"
+/* Check escape character outside variables
+                  \  @    \@ */
+#define MESSAGE6 "\\ @ \\\\@"
+
+/* Catch any edge cases */
+
+/* no substitution - not a variable */
+#define MESSAGE7 "@var1"
+
+/* Escaped variable followed by another variable */
+#define MESSAGE8 "\\\\@var1@var2@"
+
+/* Variable followed by another variable */
+#define MESSAGE9 "@var1@var2@"
+
+/* Variable followed by another variable and escaped */
+#define MESSAGE10 "@var1@var2\\\\@"
+
+/* Lots of substitutions in a row*/
+#define MESSAGE11 "@var1@@var2@@var3@@var4@"
+
+/* This should never happen in the real world, right? */
+#define MESSAGE12 "@var1@var2\\\\@var3@var4\\\\@"

test cases/common/14 configure file/meson.build

@@ -143,6 +143,8 @@ test('test5', executable('prog5', 'prog5.c'))
 conf6 = configuration_data()
 conf6.set('var1', 'foo')
 conf6.set('var2', 'bar')
+conf6.set('var3', 'baz')
+conf6.set('var4', 'qux')
 configure_file(
   input : 'config6.h.in',
   output : '@BASENAME@',

test cases/common/14 configure file/prog6.c

@@ -4,8 +4,14 @@
 int main(void) {
     return strcmp(MESSAGE1, "foo")
         || strcmp(MESSAGE2, "@var1@")
-        || strcmp(MESSAGE3, "\\foo")
+        || strcmp(MESSAGE3, "\\@var1@")
         || strcmp(MESSAGE4, "\\@var1@")
-        || strcmp(MESSAGE5, "@var1bar")
-        || strcmp(MESSAGE6, "\\ @ @ \\@ \\@");
+        || strcmp(MESSAGE5, "\\@var1")
+        || strcmp(MESSAGE6, "\\ @ \\@")
+        || strcmp(MESSAGE7, "@var1")
+        || strcmp(MESSAGE8, "\\@var1bar")
+        || strcmp(MESSAGE9, "foovar2@")
+        || strcmp(MESSAGE10, "foovar2\\@")
+        || strcmp(MESSAGE11, "foobarbazqux")
+        || strcmp(MESSAGE12, "foovar2\\@var3@var4\\@");
 }