From f78911646cf53e37c47921f9dcd9702d6e946f54 Mon Sep 17 00:00:00 2001
From: yvesf <yvesf-git@xapek.org>
Date: Sat, 20 Nov 2010 18:20:32 +0100
Subject: ergebnisliste unbefriedigend

---
 indexer.py | 35 +++++++++++++++++++----------------
 1 file changed, 19 insertions(+), 16 deletions(-)

(limited to 'indexer.py')

diff --git a/indexer.py b/indexer.py
index 22f583e..ac14a9e 100644
--- a/indexer.py
+++ b/indexer.py
@@ -18,9 +18,11 @@ schema = fields.Schema(
     createtime=fields.NUMERIC() )
 
 if not os.path.exists("index"):
+    create = True
     os.mkdir("index")
     index = create_in(u"index", schema)
 else:
+    create = False
     index = open_dir("index")
 
 filepaths = Queue()
@@ -36,7 +38,7 @@ for path, directories, files in os.walk(directory):
     for filename in files:
         if filename.endswith(".pdf"):
             filepath = os.path.join(path, filename)
-            docnum = searcher.document_number(path=filepath)
+            docnum = create or searcher.document_number(path=filepath)
             if not docnum:
                 skipped += 1
             else:
@@ -45,19 +47,20 @@ for path, directories, files in os.walk(directory):
             print u"\r{0} files found {1} skipped".format(filecount+skipped, skipped),
 print ""
 
-writer = index.writer()
-deleted = 0
-processed = 0
-for fields in searcher.all_stored_fields():
-    path = fields['path']
-    processed += 1
-    if not os.path.exists(path):
-        writer.delete_by_term('path', path)
-        deleted += 1
-    print u"\r{0} pages processed. {1} deleted".format(processed, deleted),
-print ""
+if not create:
+    writer = index.writer()
+    deleted = 0
+    processed = 0
+    for fields in searcher.all_stored_fields():
+        path = fields['path']
+        processed += 1
+        if not os.path.exists(path):
+            writer.delete_by_term('path', path)
+            deleted += 1
+        print u"\r{0} pages processed. {1} deleted".format(processed, deleted),
+    print ""
+    writer.commit()
 
-writer.commit()
 searcher.close()
 
 class PDFWorker(Thread):
@@ -73,6 +76,7 @@ class PDFWorker(Thread):
                 title = inputfile.getDocumentInfo().title
                 pagenumber = 0
                 for page in inputfile.pages:
+                    print u"{0} processing {1} Page {2}".format(self.name, filepath, pagenumber)
                     pagenumber += 1
                     content = page.extractText()
                     documents.put( {"title":title, "path":filepath, "pagenumber":pagenumber, "content":content, "createtime":time.time() } )
@@ -87,7 +91,7 @@ class IndexWorker(Thread):
         while index != None:
             try:
                 doc = documents.get(True, 0.5)
-            except Empty:
+            except Empty,e:
                 continue
             print u"{0} adding     {1} page {2}".format(self.name, doc['path'], doc['pagenumber'])
             writer = index.writer()
@@ -107,10 +111,9 @@ print "all running"
 for thread in threads:
     thread.join()
 
-idx.join()
-
 oldindex = index
 index = None
+idx.join()
 print "optimize index"
 oldindex.optimize()
 oldindex.close()
-- 
cgit v1.2.1