Fixed some syntax. Renamed abbreviated variables with verbose names. Made sure...

Fixed some syntax. Renamed abbreviated variables with verbose names. Made sure that if conversion fails, then an error is raised (rather than return empty strings).

git-svn-id: https://svn.erp5.org/repos/public/erp5/trunk@17512 20353a03-c40f-0410-a6d1-a30d3c3de9de

product/ERP5/Document/Document.py

@@ -547,12 +547,12 @@ class Document(XMLObject, UrlMixIn, ConversionCacheMixin, SnapshotMixin):
       later stage of the implementation.
     """
     # XXX results should be cached as volatile attributes
-    refs = [r[1] for r in self.getSearchableReferenceList()]
-    res = self.Base_getImplicitSuccessorValueList(refs)
+    reference_list = [r[1] for r in self.getSearchableReferenceList()]
+    result = self.Base_getImplicitSuccessorValueList(reference_list)
     # get unique latest (most relevant) versions
-    res = [r.getObject().getLatestVersionValue() for r in res]
-    res_dict = dict.fromkeys(res)
-    return res_dict.keys()
+    result = [r.getObject().getLatestVersionValue() for r in result]
+    result_dict = dict.fromkeys(result)
+    return result_dict.keys()
 
   security.declareProtected(Permissions.AccessContentsInformation, 'getImplicitPredecessorValueList')
   def getImplicitPredecessorValueList(self):
@@ -871,7 +871,6 @@ class Document(XMLObject, UrlMixIn, ConversionCacheMixin, SnapshotMixin):
     # XXX this should be security-unaware - delegate to script with proxy roles
     return res[0].getLanguage() # XXX what happens if it is empty?
 
-
   ### Property getters
   # Property Getters are document dependent so that we can
   # handle the weird cases in which needed properties change with the type of document
@@ -1070,7 +1069,7 @@ class Document(XMLObject, UrlMixIn, ConversionCacheMixin, SnapshotMixin):
       is the one to override in subclasses.
     """
     if not self.hasBaseData():
-      return ''
+      raise ConversionError('This document has not been processed yet.')
     if self.hasConversion(format='base-html'):
       mime, data = self.getConversion(format='base-html')
       return data
@@ -1098,7 +1097,7 @@ class Document(XMLObject, UrlMixIn, ConversionCacheMixin, SnapshotMixin):
     # find charset and convert to utf-8
     charset_list = self.charset_parser.findall(str(html)) # XXX - Not efficient is datastream 
                                                           # instance but hard to do better
-    if charset_list:
+    if charset_list and charset_list[0] not in ('utf-8', 'UTF-8'):
       stripped_html = unicode(str(stripped_html), charset_list[0]).encode('utf-8')
     return stripped_html