--- documentViewer/MpdlXmlTextServer.py	2011/07/15 19:45:52	1.238.2.3
+++ documentViewer/MpdlXmlTextServer.py	2011/08/03 19:04:18	1.238.2.6
@@ -12,28 +12,8 @@ import xml.etree.ElementTree as ET
 import re
 import logging
 import urllib
-import documentViewer
-#from documentViewer import getTextFromNode, serializeNode
 
-def intOr0(s, default=0):
-    """convert s to int or return default"""
-    try:
-        return int(s)
-    except:
-        return default
-
-def getText(node):
-    """get the cdata content of a node"""
-    if node is None:
-        return ""
-    # ET:
-    text = node.text or ""
-    for e in node:
-        text += gettext(e)
-        if e.tail:
-            text += e.tail
-
-    return text
+from SrvTxtUtils import getInt, getText, getHttpData
 
 def serialize(node):
     """returns a string containing an XML snippet of node"""
@@ -90,7 +70,6 @@ class MpdlXmlTextServer(SimpleItem):
     manage_changeMpdlXmlTextServerForm = PageTemplateFile("zpt/manage_changeMpdlXmlTextServer", globals())
         
     def __init__(self,id,title="",serverUrl="http://mpdl-text.mpiwg-berlin.mpg.de/mpdl/interface/", serverName=None, timeout=40):
-        
         """constructor"""
         self.id=id
         self.title=title
@@ -102,12 +81,12 @@ class MpdlXmlTextServer(SimpleItem):
         
     def getHttpData(self, url, data=None):
         """returns result from url+data HTTP request"""
-        return documentViewer.getHttpData(url,data,timeout=self.timeout)
+        return getHttpData(url,data,timeout=self.timeout)
     
     def getServerData(self, method, data=None):
         """returns result from text server for method+data"""
         url = self.serverUrl+method
-        return documentViewer.getHttpData(url,data,timeout=self.timeout)
+        return getHttpData(url,data,timeout=self.timeout)
 
     # WTF: what does this really do? can it be integrated in getPage?
     def getSearch(self, pageinfo=None,  docinfo=None):
@@ -248,18 +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
@@ -268,21 +242,25 @@ class MpdlXmlTextServer(SimpleItem):
                 
             # pageNumberOrigNorm
             elif dc == 'countFigureEntries':
-                docinfo['countFigureEntries'] = intOr0(div.text)
+                docinfo['numFigureEntries'] = getInt(div.text)
                 
             # pageNumberOrigNorm
             elif dc == 'countTocEntries':
                 # WTF: s1 = int(s)/30+1
-                docinfo['countTocEntries'] = intOr0(div.text)
+                docinfo['numTocEntries'] = getInt(div.text)
+                
+            # pageHeaderTitle
+            elif dc == 'pageHeaderTitle':
+                docinfo['pageHeaderTitle'] = div.text
                 
             # numTextPages
             elif dc == 'countPages':
-                np = intOr0(div.text)                    
+                np = getInt(div.text)                    
                 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:
@@ -332,7 +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
@@ -504,17 +483,9 @@ class MpdlXmlTextServer(SimpleItem):
                 pagediv = div
                 
             elif dc == 'queryResultHits':
-                docinfo['tocSize_%s'%mode] = intOr0(div.text)
+                docinfo['tocSize_%s'%mode] = getInt(div.text)
 
         if pagediv:
-#            # split xml in chunks
-#            tocs = []
-#            tocdivs = pagediv.findall('div')
-#            for p in zip(tocdivs[::2], tocdivs[1::2]):
-#                toc = serialize(p[0])
-#                toc += serialize(p[1])
-#                tocs.append(toc)
-#                logging.debug("pair: %s"%(toc))
             # store XML in docinfo
             docinfo['tocXML_%s'%mode] = ET.tostring(pagediv, 'UTF-8')