Changeset 3910 for TI01-discovery

Timestamp:

19/05/08 17:15:57 (11 years ago)

Author:

cbyrom

Message:

Fix problems with molesReadWrite - including escaping of special
characters + avoiding outputting abstractText element twice.

File:

• TI01-discovery/branches/ingestAutomation-upgrade/OAIBatch/molesReadWrite.py (modified) (15 diffs)

TI01-discovery/branches/ingestAutomation-upgrade/OAIBatch/molesReadWrite.py

@@ -1 @@
-#Dominic Lowe, BADC  18 October 2006
-#Updated Sue Latham, BADC 6 December 2006 - add dgStructuredKeyword class
-
-import cElementTree as ET
-import sys
+try: #python 2.5
+    from xml.etree import cElementTree as ET
+except ImportError:
+    try:
+        # if you've installed it yourself it comes this way
+        import cElementTree as ET
+    except ImportError:
+        # if you've egged it this is the way it comes
+        from ndgUtils.elementtree import cElementTree as ET
+import sys, re
 
 
@@ -24 +29 @@
         #if this element is a parent of other elements (excluding annotation and documentation)
         #then return True else return false
-        if elem[:]==[]:
+        if elem.getchildren()==[]:
             return False
-        for child in elem[:]:
+        for child in elem.getchildren():
             if child.tag == '{http://www.w3.org/2001/XMLSchema}element':
                 return True
@@ -59 +64 @@
 
         return False
-        
-        
-        
+
         
     def __checkChildren(self, parentclassname, elem):     
         #if parentclassname == 'dgMetadataProvenance':
             #pdb.set_trace()
-        for child in elem[:]:
+        for child in elem.getchildren():
             childname = None
             if child.attrib.has_key('name'):        
@@ -89 +92 @@
                     tmp= self.sequences[parentclassname]                    
                     tmp.append(childname)
-                    self.sequences[parentclassname]=tmp             
+                    self.sequences[parentclassname]=tmp
                 if self.__isParent(child):
                     if not self.sequences.has_key(childname):
@@ -105 +108 @@
                 pass
             
-    def setup(self):    
+    def setup(self):
         tree = ET.parse(self.schema)
         root = tree.getroot()
@@ -170 +173 @@
             setattr(self,childname, childobj)
     
-    def toXML(self,molesFrag, schema=None):
+    def toXML(self, molesFrag, schema=None):
         if schema != None:
             self.schema=schema
         else:
             self.schema=None
-        #print molesFrag.tag
         orderedAttribs=schema.lookupOrder(self.__dict__,molesFrag.tag)
         for item in orderedAttribs:
-           #print "item in orderedAttribs is %s" %item
            if type(item) is xmlSchema:
                         continue
            if item == '{http://ndg.nerc.ac.uk/moles}':
                 continue
+                    
+           # avoid abstractText being output twice
+           # NB, this is a hack to fix the problem of this class not handling
+           # namespaces correctly - ideally this is what should be fixed here
+           if str(molesFrag.tag) == "abstract":
+                if len(molesFrag) > 0:
+                    continue
+            
+           frag = None
            if hasattr(self, item):
                 if isinstance(self.__dict__[item], molesElement):
@@ -193 +203 @@
                             frag=ET.Element(item)
                             it.toXML(frag, schema=self.schema)
-                            molesFrag.append(frag)
                         else:
                             frag=ET.Element(item)
                             frag.text=it                            
-                            molesFrag.append(frag)
+                        
+                        molesFrag.append(frag)
                 else:
                     frag=ET.Element(item)
                     frag.text=self.__dict__[item]
                     molesFrag.append(frag)
+            
         return molesFrag
             
@@ -224 +235 @@
                 else:
                     childWithoutNS, ns=self._stripNS(child.tag)
-                    setattr(self,childWithoutNS, child.text)
+                    
+                    # escape any special characters - mainly required for
+                    # >, etc which get expanded out whilst processing the
+                    # elements here
+                    value = self.escapeSpecialCharacters(child.text)
+                    setattr(self,childWithoutNS, value)
+                    
+    def escapeSpecialCharacters(self, text):
+        '''
+        Escape any special characters included in the input text string
+        @param text: string to escape special characters from
+        @return: string with special characters escaped 
+        '''
+        subtool=subAI()
+        s=subtool.sub(text)
+        #if text != s:
+        #    print "NB, adjusted string content: %s -> %s" %(text, s)
+        return s
+        
+
+class subAI:
+    ''' This is Alan Iwi's substitute and replace orphan <> code '''
+    def __init__(self):
+        self.r1=re.compile('<([^>]*(<|$))')
+        self.r2=re.compile('((^|>)[^<]*)>')
+    def sub(self,s):
+        if s is None:
+            return s
+        
+        old=''
+        while s != old:
+            old=s
+            s=self.r1.sub(r'&lt;\1',s)
+            s=self.r2.sub(r'\1&gt;',s)
+        return s
         
                     
@@ -253 +298 @@
         ['dataModelID', \
         'dgDataGranule', \
+        'dgGranuleSummary',\
+        'dgGranuleName',\
+        'simpleCondition',\
+        'dgSecurityCondition',\
+        'accessControlPolicy',\
         'dgDataEntity', \
         'dgMetadataRecord', \
@@ -263 +313 @@
         'dgTemporalCoverage', \
         'dgBoundingBox', \
+        'dgArea',\
         'DateRange', \
         'dgDataSummary', \
@@ -279 +330 @@
         'dgValidTermParentID', \
         'dgValidSubterm', \
+        'ListLevel',\
         'metadataDescriptionID', \
         'dgMetadataDescription', \
@@ -284 +336 @@
         'abstract',\
         'descriptionSection',\
+        'dgReferenceClass',\
         'descriptionOnlineReference',\
         'dgSimpleLink',\
@@ -318 +371 @@
         'DataProductionToolID',\
         'ObservationStationID',\
+        'ObsStationDeployment',\
         'dgPrincipalInvestigator',\
         'dgInvestigator',\
@@ -323 +377 @@
         'dgCruise',\
         'RelatedDeployment',\
+        'dgModel',\
+        'DPTDeployment',\
+        'dgInstrument',\
+        'dgDataProductionTool',\
+        'dgDPTRole',\
+        'dgObservationStation',\
+        'dgStationaryPlatform',\
+        'dgMovingPlatform',\
+        'dgLandStation',\
+        'dgMooring',\
+        'position',\
+        'dgStationGroup',\
+        'dgShip',\
+        'vesselType',\
+        'RecordCreation',\
+        'RecordUpdate',\
+        'dgMetadataProvenance',\
         ]
 
@@ -338 +409 @@
         print key
         print schema.sequences[key]
-    
-
  
-   
+
 if __name__=='__main__':
     main()