Changeset 1689

Timestamp:

07/28/06 16:44:50 (2 years ago)

Author:

elvelind

Message:

fix #242. Problems with I18N and (X)HTML entities

Files:

• branches/1.0/turbogears/toolbox/admi18n/pygettext.py (modified) (2 diffs)
• branches/1.0/turbogears/util.py (modified) (2 diffs)

branches/1.0/turbogears/toolbox/admi18n/pygettext.py

@@ -168 +168 @@
-import tokenize
-import operator
-
-
+
+
+
+
-
-__version__ = '1.5'
@@ -517 +519 @@
-        f = None
-        try:
-
+
+
-        except Exception, e:
-            print 'Skip %s: %s' % (self.__curfile, e)
-            return
+
-        node = f.getroot()
-        self.get_text_node(node)

branches/1.0/turbogears/util.py

@@ -1 +1 @@
-import os
-import sys
+import re
+import htmlentitydefs
-from inspect import getargspec, getargvalues
-from itertools import izip, islice, chain, imap
@@ -395 +397 @@
-        return value
-
+def fixentities(htmltext):
+    # replace HTML character entities with numerical references
+    # note: this won't handle CDATA sections properly
+    def repl(m):
+        entity = htmlentitydefs.entitydefs.get(m.group(1).lower())
+        if not entity:
+            return m.group(0)
+        elif len(entity) == 1:
+            if entity in "&<>'\"":
+                return m.group(0)
+            return "&#%d;" % ord(entity)
+        else:
+            return entity
+    return re.sub("&(\w+);?", repl, htmltext)
+
+
-__all__ = ["Enum", "setlike", 
-           "get_package_name", "get_model", "load_project_config",