--- documentViewer/MpdlXmlTextServer.py	2011/07/29 18:36:04	1.238.2.5
+++ documentViewer/MpdlXmlTextServer.py	2011/08/04 14:57:02	1.238.2.7
@@ -1,4 +1,3 @@
-
 from OFS.SimpleItem import SimpleItem
 from Products.PageTemplates.PageTemplateFile import PageTemplateFile 
 
@@ -227,18 +226,13 @@ class MpdlXmlTextServer(SimpleItem):
           
     def processPageInfo(self, dom, docinfo, pageinfo):
         """processes page info divs from dom and stores in docinfo and pageinfo"""
-        # process all toplevel divs
-        alldivs = dom.findall(".//div")
-        pagediv = None
+        # assume first second level div is pageMeta
+        alldivs = dom.find("div")
         for div in alldivs:
             dc = div.get('class')
             
-            # page content div
-            if dc == 'pageContent':
-                pagediv = div
-            
             # pageNumberOrig  
-            elif dc == 'pageNumberOrig':
+            if dc == 'pageNumberOrig':
                 pageinfo['pageNumberOrig'] = div.text
                 
             # pageNumberOrigNorm
@@ -247,12 +241,16 @@ class MpdlXmlTextServer(SimpleItem):
                 
             # pageNumberOrigNorm
             elif dc == 'countFigureEntries':
-                docinfo['countFigureEntries'] = getInt(div.text)
+                docinfo['numFigureEntries'] = getInt(div.text)
                 
             # pageNumberOrigNorm
             elif dc == 'countTocEntries':
                 # WTF: s1 = int(s)/30+1
-                docinfo['countTocEntries'] = getInt(div.text)
+                docinfo['numTocEntries'] = getInt(div.text)
+                
+            # pageHeaderTitle
+            elif dc == 'pageHeaderTitle':
+                docinfo['pageHeaderTitle'] = div.text
                 
             # numTextPages
             elif dc == 'countPages':
@@ -270,7 +268,7 @@ class MpdlXmlTextServer(SimpleItem):
         return
          
            
-    def getTextPage(self, mode="text_dict", pn=1, docinfo=None, pageinfo=None):
+    def getTextPage(self, mode="text", pn=1, docinfo=None, pageinfo=None):
         """returns single page from fulltext"""
         logging.debug("getTextPage mode=%s, pn=%s"%(mode,pn))
         # check for cached text -- but this shouldn't be called twice
@@ -293,9 +291,12 @@ class MpdlXmlTextServer(SimpleItem):
         characterNormalization = pageinfo.get('characterNormalization', None)
         selfurl = docinfo['viewerUrl']   
         
-        if mode == "text_dict":
+        if mode == "dict":
             # text_dict is called textPollux in the backend
             textmode = "textPollux"
+        elif not mode:
+            # default is text
+            textmode = "text"
         else:
             textmode = mode
         
@@ -311,7 +312,8 @@ class MpdlXmlTextServer(SimpleItem):
         # page content is in <div class="pageContent">
         pagediv = None
         # ElementTree 1.2 in Python 2.6 can't do div[@class='pageContent']
-        alldivs = dom.findall(".//div")
+        # so we look at the second level divs
+        alldivs = dom.findall("div")
         for div in alldivs:
             dc = div.get('class')
             # page content div