Changeset 1097

Timestamp:

06/06/06 13:24:29 (15 years ago)

Author:

lawrence

Message:

This has the code to support a temporary patch for
the unicode and orphan html characters problems. Further
details are in ticket:311

Location:

TI07-MOLES/trunk/StubB/XSLT/browse/portal/cgi

Files:

• DIF.py (modified) (2 diffs)
• ETxmlView.py (modified) (2 diffs)
• browseCGI.py (modified) (1 diff)
• renderDiscoverySet.py (modified) (1 diff)
• sub_orphan.py (added)

TI07-MOLES/trunk/StubB/XSLT/browse/portal/cgi/DIF.py

@@ -8 +8 @@
 from ServiceBinding import Service
 from renderEntity import renderEntity
+from ETxmlView import loadET
 
 class DIF:
@@ -16 +17 @@
         '''Initialise a python dif instance based on an xml document '''
         self.metadataType='DIF'
-        #self.xml=xml.decode('utf-8','replace')
-        #self.xml=unicode(xml,'latin-1')
-        #self.xml=xml.encode('latin-1','replace')
-        encodings=['ascii',]#['latin-1','ascii','iso-8859-1']
         try:
-            for option in encodings:
-                try:
-                    s=xml.encode(option,'replace')
-                    
-                    self.elem=ET.fromstring(s)
-                    self.xml=s
-                except UnicodeError:
-                    pass
-                else:
-                    break
+            self.elem=loadET(xml)
+            self.xml=xml
         except:
             # for some reason we can't parse the document ...
-            #print '<p>%s</p>'%xml[11385:11395]
-            #raise bnl
-            #import codecs
-            #f=codecs.open('unparseable.xml','a','utf-8')
-            #f.write(xml+'\n')
-            print '<p>'+s[11360:11420]+'</p>'
-            raise 
+            raise
             self.elem=None
             return

TI07-MOLES/trunk/StubB/XSLT/browse/portal/cgi/ETxmlView.py

@@ -8 +8 @@
 import ElementTree as ET
 import re
+from sub_orphan import *
 
 def et2text(elem,indent='',html=0,space='   '):
@@ -64 +65 @@
         result+='%s/%s%s'%(lt,span(elem.tag,'xmlElemTag'),gt)
     return div(result,'xmlElem')
+    
+def loadET(inputString):
+    ''' This method returns an elementtree object after some cleaning 
+    of the string, essentially a hack to make sure that xml doesn't contain any 
+    naughty & characters alone (typically from URL copies), or unescaped orphan 
+    < or > signs ... and that the unicode has been processed to something that 
+    might work'''
+     
+    inputString=re.sub(r'&(?!\w+;)', '&amp;', inputString)
+    
+    # first just try and do it so we don't waste time if we don't need to ...
+    try:
+        elem=ET.fromstring(inputString)
+        return elem
+    except:
+        pass
+        # and carry on 
 
-def xmlCleanup(string):
-    '''This is a hack to make sure that xml doesn't contain any naughty & characters alone
-    (typically from URL copies)'''
-    #match ? but not (?! if it's followed by alphanumeric \w characters + and a semicolon ;
-    #don't want to use urlencode cos we don't know where it came from ...
-    return re.sub(r'&(?!\w+;)', '&amp;', string)
+    #ok, let's deal with orphan > and < signs then ...
+    subtool=subAI()
+    s=subtool.sub(inputString)
+
+    #now let's sort out an encoding
+    encodings=['utf-8','latin-1','iso-8859-1','ascii',]
+    elem=None
+    for option in encodings:
+        try:
+            s=s.encode(option,'replace')
+            try:
+                elem=ET.fromstring(s)
+            except:
+                s=re.sub('\n','<br/>LINE: ',s)
+                print '<p>%s</p>'%s
+                raise
+        except UnicodeError:
+            pass
+        else:
+            break
+    return elem
 
 def xml2text(xmlString):
-    tree=ET.fromstring(xmlCleanup(xmlString))
+    tree=loadET(xmlString)
     return et2text(tree)
 
 def xml2HTML(xmlString,**kw):
-    tree=ET.fromstring(xmlCleanup(xmlString))
+    tree=loadET(xmlString)
     return et2HTML(tree,**kw)

TI07-MOLES/trunk/StubB/XSLT/browse/portal/cgi/browseCGI.py

@@ -230 +230 @@
                 results=ws.GetResults(offset=state.offset,number=state.stride)
                 difs=[]
-                for result in results:
-                    difs.append(xmlCleanup(result))
+                for result in results: difs.append(result)
                 html=renderDiscoverySet(difs,state,selector=self.selector,
                                summary=1,spatial=1,temporal=1,services=1)

TI07-MOLES/trunk/StubB/XSLT/browse/portal/cgi/renderDiscoverySet.py

@@ -132 +132 @@
     for result in results:
         #g.write(xmlCleanup(result)+'\n')
-        difs.append(xmlCleanup(result))
+        difs.append(result)
     html=renderDiscoverySet(difs,state,summary=1,spatial=1,temporal=1)
     f=file('output.html','wb')