--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/web/ldt/ldt_utils/projectindexer.py	Tue Jun 08 15:44:35 2010 +0200
@@ -0,0 +1,110 @@
+import tempfile
+import os
+import os.path
+import shutil
+import ldt.utils.xml
+from ldt import settings
+import xml
+import xml.dom
+import xml.dom.minidom
+import xml.dom.ext
+import xml.xpath
+import lucene
+from ldt.ldt_utils import STORE
+from ldt.ldt_utils import ANALYZER
+
+def Property(func):
+    return property(**func()) 
+
+class ProjectIndexer(object):
+    def __init__(self, projectList, writer, decoupage_blackList = settings.DECOUPAGE_BLACKLIST):
+        self.__projectList = projectList
+        self.__decoupage_blacklist = decoupage_blackList
+        self.__writer = writer
+            
+    @Property
+    def decoupage_blacklist(): #@NoSelf
+        doc = """get blacklist""" #@UnusedVariable
+       
+        def fget(self):
+            if self.__decoupage_blacklist is None:
+                self.__decoupage_blacklist = ()
+            return self.__decoupage_blacklist
+           
+        def fset(self, value):
+            self.__decoupage_blacklist = value
+           
+        def fdel(self):
+            del self.__decoupage_blacklist
+           
+        return locals()
+               
+    def index_all(self):
+        for project in self.__projectList:
+            self.index_project(project)
+
+    def index_project(self, project):
+        # ldt.utils.log.debug("Indexing project : "+str(project.ldt_id))
+        ldt=project.ldt
+        doc = xml.dom.minidom.parseString(ldt.encode( "utf-8" ))
+        
+        self.__writer.deleteDocuments(lucene.Term("ldt_id", project.ldt_id))
+            
+        con = xml.xpath.Context.Context(doc, 1, 1, None)
+        res = xml.xpath.Evaluate("/iri/annotations/content", context=con)
+
+        for content in res:
+            contentId = content.getAttribute("id")
+            
+            res =xml.xpath.Evaluate("ensemble", content)
+            for ensemble in res:
+                ensembleId = ensemble.getAttribute("id")
+                
+                for decoupageNode in ensemble.childNodes:
+                    # ldt.utils.log.debug("Indexing project decoupage : "+ repr(decoupageNode.nodeType) + " in " + repr(self.decoupage_blacklist))
+                    if decoupageNode.nodeType != xml.dom.Node.ELEMENT_NODE or decoupageNode.tagName != "decoupage"  or decoupageNode.getAttribute("id") in self.decoupage_blacklist:
+                        continue
+                
+                    decoupId = decoupageNode.getAttribute("id")
+                    res = xml.xpath.Evaluate("elements/element", decoupageNode)
+                    for elementNode in res:
+                        doc = lucene.Document()
+                        elementId = elementNode.getAttribute("id")
+                        tags = elementNode.getAttribute("tags")
+                        
+                        if tags is not None:                            
+                            tags.replace(",", ";")
+                        
+                        if tags is None or len(tags) == 0:
+                            tags = ""
+                            restagnode = xml.xpath.Evaluate("tag/text()", elementNode)
+                            for tagnode in restagnode:
+                                tags = tags + " ; " + tagnode.data
+                                
+                        if tags is None or len(tags) == 0:
+                            tags = ""
+                            restagnode = xml.xpath.Evaluate("tags/tag/text()", elementNode)
+                            for tagnode in restagnode:
+                                tags = tags + " ; " + tagnode.data                            
+
+                        title = ""
+                        for txtRes in xml.xpath.Evaluate("title/text()", elementNode): 
+                            title = title + txtRes.data 
+                
+                        abstract = ""
+                        for txtRes in xml.xpath.Evaluate("abstract/text()", elementNode): 
+                            abstract = abstract + txtRes.data 
+                            
+                        doc.add(lucene.Field("ldt_id", project.ldt_id, lucene.Field.Store.YES, lucene.Field.Index.UN_TOKENIZED))              
+                        doc.add(lucene.Field("iri_id", contentId, lucene.Field.Store.YES, lucene.Field.Index.UN_TOKENIZED))
+                        doc.add(lucene.Field("ensemble_id", ensembleId, lucene.Field.Store.YES, lucene.Field.Index.NO))
+                        doc.add(lucene.Field("decoupage_id", decoupId, lucene.Field.Store.YES, lucene.Field.Index.NO))
+                        doc.add(lucene.Field("element_id", elementId, lucene.Field.Store.YES, lucene.Field.Index.NO))
+                        doc.add(lucene.Field("tags", tags, lucene.Field.Store.NO, lucene.Field.Index.TOKENIZED))
+                        doc.add(lucene.Field("title", title, lucene.Field.Store.NO, lucene.Field.Index.TOKENIZED))
+                        doc.add(lucene.Field("abstract", abstract, lucene.Field.Store.NO, lucene.Field.Index.TOKENIZED))
+                        doc.add(lucene.Field("all", " ".join([tags, title, abstract]), lucene.Field.Store.NO, lucene.Field.Index.TOKENIZED))
+            
+                        self.__writer.addDocument(doc)
+    
+        self.__writer.flush()
\ No newline at end of file