--- ECHO_content/ECHO_xslt.py	2006/09/10 22:57:38	1.9
+++ ECHO_content/ECHO_xslt.py	2006/09/11 14:43:23	1.10
@@ -24,6 +24,7 @@ import urllib
 try:
     from Ft.Xml.Xslt.Processor import Processor
     from Ft.Xml import InputSource, EMPTY_NAMESPACE,Parse
+    from Ft.Xml.Domlette import NonvalidatingReader
 except:
     print "4suite has to be installed"
 
@@ -144,7 +145,13 @@ class ECHO_xslt(ECHO_pageTemplate,ECHO_l
 
     meta_type="ECHO_xslt"
     
-    cache=ECHO_cache()
+    cache=ECHO_cache() # cache for analysed pages
+    caching="yes"
+    
+    appendQueryString=True # add query string to the cgiUrl can be changed with addChanges
+    
+    passURL=False #use url from querystring parameter fn to retrieve the text and not the url in cgi-url can be changed with addChanges
+        
     
     results={}
     manage_options=ECHO_pageTemplate.manage_options+(
@@ -164,13 +171,30 @@ class ECHO_xslt(ECHO_pageTemplate,ECHO_l
         pt=zptFile(self, 'zpt/ChangeECHO_xsltForm.zpt')
         return pt()
 
-    def addChanges(self,cgiUrl,RESPONSE=None):
+    def addChanges(self,cgiUrl,appendQueryString=False,passURL=False,caching=False,RESPONSE=None):
         """change the xslt, ueberschriebt addChanges in ECHO_PageTemplate"""
+    
         if urlparse.urlparse(cgiUrl)[0]=="":#relative url in absolute
             self.cgiUrl=urlparse.urljoin(self.absolute_url(), cgiUrl)
         else:
             self.cgiUrl=cgiUrl
-            
+        
+        if appendQueryString: 
+            self.appendQueryString=True
+        else:
+            self.appendQueryString=False
+        
+        if passURL:
+            self.passURL=True
+        else:
+            self.passURL=False
+       
+        if caching:
+            self.caching="yes"
+        else:
+            self.caching="No"
+        
+
         if RESPONSE:
             RESPONSE.redirect("manage_main")
         
@@ -218,30 +242,38 @@ class ECHO_xslt(ECHO_pageTemplate,ECHO_l
                 return self._v_xmltrans.getResult()
 
     
+    def getText(self):
+        """print nur den text"""
+        qs,baseUri=self.getTextInput()
+        self.REQUEST.RESPONSE.redirect(qs)
 
     def deleteCache(self):
         """deletefrom cache"""
         fn=self.REQUEST['fn']
         self.cache.deleteObject(fn)
         
-    def getPageLex(self,_pn="1",_caching="yes"):
+    def getPageLex(self,_pn="1",_caching=None):
         """getpage mit lexikalischer analyse und xslt transform
         if _caching=yes dann wird die lwxikalisch analysierte seite in einem cache abgespeichert
         """
-       
+
+        if not _caching:
+            _caching=self.caching
+            
         fn=self.REQUEST['fn']
 
         
         fromCache=self.cache.retrieveObject(fn,_pn)
-
+ 
         if fromCache and _caching=="yes":
-            print "retrieve",fn,_pn 
+          
             txt = fromCache
         else:
             txt=self.tagLex(nr=_pn)   
-            print "store",fn,_pn 
+         
             self.cache.storeObject(fn,_pn,txt[0:])
         
+       
         xsl=self.xslt()
         
         xsltproc=Processor()
@@ -255,33 +287,65 @@ class ECHO_xslt(ECHO_pageTemplate,ECHO_l
         
         return tmp[0:]
             
-
-    def getPage(self,_pn,REQUEST=None):
+    def getTextInput(self):
+        """get the text
+        wie der text geholt wird liegt an der konfiguration,
+        is appendQueryString gesetzt, dann wir jeweils der Querystring an vorgebenen url gesetzt, erwartet wird fn=
+        fźr den Pfad, is passURL gesetzt, dann wird falls fn= eine vollstŠndige url enthŠlt, diese anstelle der in cgiurl definierten genommen.
+        """
+        
+        if getattr(self,'passURL',False) and self.REQUEST.has_key('fn') and (urlparse.urlparse(self.REQUEST['fn'])[0]=='http'):
+            qs=self.REQUEST['fn']
+            baseUri=qs
+        elif getattr(self,'pappendQueryString',True):
+            qs="%s%s"%(self.cgiUrl,self.REQUEST['QUERY_STRING'])
+            baseUri=self.cgiUrl
+        else:
+            qs="%s"%(self.cgiUrl)
+            baseUri=self.cgiUrl
+        
+        #fact= InputSource.DefaultFactory.fromUri(qs)
+        return qs,baseUri
+        #return InputSource.InputSource(fact)
+        #xmlt=urllib.urlopen(qs).read()
+        
+    def getPage(self,_pn,REQUEST=None,_caching=None):
         """get a page from an xml"""
+        
+        if not _caching:
+            _caching=self.caching
+            
         pn=int(_pn)-1
         if pn<0:
             if REQUEST:
                 return "Sorry, pagenumbers have to be greater than 0"
             else:
                 return None
-        impl=xml.dom.getDOMImplementation()
-        newDoc=impl.createDocument(EMPTY_NAMESPACE,None,None
-                                   )
-        root=newDoc.createElement('page')
-        newDoc.appendChild(root)
+       
+        xmlt,self.baseUri=self.getTextInput()
         
+        #get the text from cache, if existing
+        fromCache=self.cache.retrieveObject(self.baseUri,"-1")
+        if fromCache and _caching=="yes":
+          
+            txt = fromCache
+        else:
+
+            txt=urllib.urlopen(xmlt).read()
+            
+            self.cache.storeObject(self.baseUri,"-1",txt)
         
-        qs="%s%s"%(self.cgiUrl,self.REQUEST['QUERY_STRING'])
+        dom=NonvalidatingReader.parseString(txt,self.baseUri)
         
+        #pb should have a namespache
         
-     
-        xmlt=urllib.urlopen(qs).read()
-        dom=Parse(xmlt)
-       
-        pbs=dom.xpath("//pb")
+        pbs=dom.xpath("//mpiwg:pb",explicitNss={'mpiwg':'http://www.mpiwg-berlin.mpg.de/namespace'})
+        if len(pbs)==0: # versuche nochmal ohne
+            pbs=dom.xpath("//pb")
+
         if pn > len(pbs):
             if REQUEST:
-                return "Sorry, pagenumber %s does not exit"%pn+1
+                return "Sorry, pagenumber %s does not exit"%(pn+1)
             else:
                 return None
             
@@ -292,30 +356,6 @@ class ECHO_xslt(ECHO_pageTemplate,ECHO_l
         else:
             endNode=None
         
-           
-#        follower=beginNode.xpath('following::node()')
-#        #root.appendChild(beginNode)
-#        found=[]
-#       
-#        for node in endNode.xpath('preceding::node()'):
-#            if node in follower:
-#                found.append(node)
-#    
-        
-   
-        #rempove all children of parents which are also in the list
-#        for x in found:
-#            for child in x.xpath(".//*"):
-#                try:
-#                  
-#                    nr=found.index(child)
-#               
-#                    del(found[nr])
-#                except:
-#                    pass
-#
-#        found.append(endNode)
-#       
         deleteNodes=beginNode.xpath('preceding::node()')
         if endNode:
             deleteNodes+=endNode.xpath('following::node()')
@@ -330,7 +370,7 @@ class ECHO_xslt(ECHO_pageTemplate,ECHO_l
         strio = StringIO()
         PrettyPrint(dom,strio) 
         xmlstr = strio.getvalue()
-
+        
         return xmlstr