Changeset 4196 for exist

Timestamp:

15/09/08 10:41:44 (11 years ago)

Author:

cbyrom

Message:

Extend xmlHandler to allow specification of different namespaces when
using the _distrbutens method to decorate xpath expressions with namespaces
+ add logging + add namespace info to ndgObject - for NSs used in the
various doc types + fix bug in ndgSearch to check response Hits properly.

Location:

exist/trunk/python/ndgUtils

Files:

• ndgObject.py (modified) (4 diffs)
• ndgSearch.py (modified) (3 diffs)
• xmlHandler2.py (modified) (5 diffs)

exist/trunk/python/ndgUtils/ndgObject.py

@@ -6 +6 @@
                 (1) a downloadable xml representation from a repository,
                 (2) a printable xml representation '''
-    
+
     # The various different document types:
     MOLES_DOC_TYPE = 'NDG-B0'
@@ -19 +19 @@
     ATOM_DOC_TYPE = 'ATOM'
     ATOM_BACKUP_DOC_TYPE = 'ATOM-BACKUP'
+
+        # various namespaces used in the docs
+    ATOM_NS = 'http://www.w3.org/2005/Atom'
+    MOLES_NS = 'http://ndg.nerc.ac.uk/schema/moles2alpha'
+    GEOSS_NS = 'http://www.georss.org/georss'
+    GML_NS = 'http://www.opengis.net/gml'
         
     # Group the doc types according to the source they should be retrieved from
@@ -26 +32 @@
                       NUMSIM_DOC_TYPE, ATOM_DOC_TYPE, \
                       ATOM_BACKUP_DOC_TYPE]
-
+    
     def __init__(self,uri,config=None):
         ''' Parse the uri and prepare for obtaining the actual content'''
@@ -145 +151 @@
         #now, we'll build a stub-B url as well, in case that comes in handy 
         if self.schema!='NumSim' and self.gettable<>-1:
-            self.BURL=discoveryBASE.replace(self.schema,'NDG-B1')
+            if self.schema:
+                self.BURL=discoveryBASE.replace(self.schema,'NDG-B1')
             if server:
                 self.BURL=self.BURL.replace(server,servicehost)

exist/trunk/python/ndgUtils/ndgSearch.py

@@ -34 +34 @@
 #
 
-default_HostAndPort="http://localhost:8080/axis2/services/DiscoveryService"
+default_HostAndPort="http://ndg.badc.rl.ac.uk:8080/axis2/services/DiscoveryService"
 
 class ndgSearch:
@@ -98 +98 @@
         etime=time.time()-itime
         logging.debug('Search Request [%s] took [%ss]'%(term,etime))
-        
+
+        # NB, set default and only change this if response lists Hits - this
+        # can return 'None' - so isn't reliable var to set to without checking
+        self.hits = 0
+        if response.Hits:
+            self.hits = response.Hits
+            
         if response._status:
             self.serverSessionID=response._resultId
-            self.hits=response.Hits
             self.documents=response.Documents.Document
-        else:
-            self.hits=response.Hits
         
         logging.info("Search returned %s results: %s" %(self.hits, self.documents))
@@ -225 +228 @@
         #so we know that the following call is the problem ...
         responses=self.getAllDocs(format)
+        
         i=len(self.documents)
         j=len(responses)

exist/trunk/python/ndgUtils/xmlHandler2.py

@@ -13 +13 @@
         
 from xml.parsers.expat import ExpatError
-import StringIO, re
+import StringIO, re, logging
 XMLHDR='<?xml version="1.0"'
 from ETxmlView import subAI
@@ -24 +24 @@
            - load up an element-tree
            - collect a namespace map '''
-        
+        logging.info("Reading in XML file - to create an elementtree object")
         self.cleanup=subAI()
 
@@ -52 +52 @@
         self.__getns()
         
-        
         self.tree=ET.XML(self.xmls)
         self.__updatens()
+        logging.info("XML file read into elementtree object")
     
     def __getns(self):
@@ -61 +61 @@
         #ought to do this with a regular expression, but needs must
         # or bettter yet, use iterparse in the first place, but that seemed slow.
-        
-
         self.defns=None
         if self.xmls[0:19]==XMLHDR:
@@ -176 +174 @@
         return h
      
-    def _distributens(self,xpathExpression):
-        ''' Actually we only support tag finding in this '''
-        if self.defns is None: return xpathExpression
+    def _distributens(self, xpathExpression, alternativeNS=None):
+        ''' 
+        Update xpath expression to include namespaces.  NB, the
+        default doc NS is used if available and if the keyword, 'alternativeNS'
+        has not been set to specify a different namespace
+        @param xpathExpression: xpath expression to update with namespace
+        @type xpathExpression: str
+        @keyword alternativeNS: alternative namespace to decorate the xpath
+        expression with
+        @type alternativeNS: str
+        @return: xpath expression with name space decoration  
+        '''
+        logging.debug("Adding namespace to xpath expression, '%s'" \
+                      %xpathExpression)
+        if self.defns is None and not alternativeNS:
+            logging.debug("- no namespace specified - returning")
+            return xpathExpression
+        
         tags=xpathExpression.split('/')
         new=''
+        ns = (alternativeNS or self.defns)
         for t in tags: 
             if t[1]<>'{': 
-                new+='{%s}%s/'%(self.defns,t)
+                new+='{%s}%s/'%(ns, t)
             else:
                 new+=t+'/'
         new=new[0:-1]
+        logging.debug("- added namespace, '%s'" %ns)
+        logging.debug("- final xpath expression, '%s'" %new)
         return new