--- documentViewer/MpdlXmlTextServer.py	2011/07/19 18:46:34	1.238.2.4
+++ documentViewer/MpdlXmlTextServer.py	2011/08/03 19:04:18	1.238.2.6
@@ -227,18 +227,13 @@ 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 +242,16 @@ 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':
@@ -260,8 +259,8 @@ class MpdlXmlTextServer(SimpleItem):
                 if np > 0:
                     docinfo['numTextPages'] = np
                     if docinfo.get('numPages', 0) == 0:
-                        # seems to be text-only
-                        docinfo['numTextPages'] = np
+                        # seems to be text-only - update page count
+                        docinfo['numPages'] = np
                         pageinfo['end'] = min(pageinfo['end'], np)
                         pageinfo['numgroups'] = int(np / pageinfo['groupsize'])
                         if np % pageinfo['groupsize'] > 0:
@@ -311,7 +310,8 @@ 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