Fix the comment parsing: allow more than simply alphanums in the text. Now

it allows everything but the closing bracket.

Add some commentary for the regex since it is central to the operation of
the templating engine.


git-svn-id: http://viewvc.tigris.org/svn/viewvc/trunk@215 8cb11bc2-c004-0410-86c3-e597b4017df7

lib/ezt.py

@@ -35,7 +35,19 @@ import string
 import re
 from types import StringType, IntType, FloatType
 
-_re_parse = re.compile('(\[[-\w.# ]+\])|(\[\[\])')
+#
+# This regular expression matches three alternatives:
+#   expr: DIRECTIVE | BRACKET | COMMENT
+#   DIRECTIVE: '[' ('-' | '.' | ' ' | alphanum)+ ']
+#   BRACKET: '[[]'
+#   COMMENT: '[#' not-rbracket* ']'
+#
+# When used with the split() method, the return value will be composed of
+# non-matching text and the two paren groups (DIRECTIVE and BRACKET). Since
+# the COMMENT matches are not placed into a group, they are considered a
+# "splitting" value and simply dropped.
+#
+_re_parse = re.compile('(\[[-\w. ]+\])|(\[\[\])|\[#[^\]]*\]')
 
 # block commands and their argument counts
 _block_cmd_specs = { 'if-any':1, 'if-index':2, 'for':1 }
@@ -54,6 +66,7 @@ class Template:
     # parse the program into: (TEXT DIRECTIVE BRACKET)* TEXT
     # DIRECTIVE will be '[directive]' or None
     # BRACKET will be '[[]' or None
+    # note that comments are automatically dropped
     parts = _re_parse.split(text)
 
     program = [ ]
@@ -74,9 +87,6 @@ class Template:
         # DIRECTIVE is present.
         args = string.split(piece[1:-1])
         cmd = args[0]
-        if cmd == '#':
-          # comment
-          continue
         if cmd == 'else':
           if len(args) > 1:
             raise ArgCountSyntaxError()