X-Git-Url: https://git.phdru.name/?a=blobdiff_plain;f=m_librarian%2Finp.py;h=76ab97be9883c4cd240ae725acc3300acf897e98;hb=ce323cc1847cdb57a6fc6df2c685e68fd76be522;hp=56f85e4d1dcdfb4561fe0ac08cc4d364c291f25a;hpb=991627dd1d4fcdecc04a091ec9c0c8b9e8e8b27c;p=m_librarian.git

diff --git a/m_librarian/inp.py b/m_librarian/inp.py
index 56f85e4..76ab97b 100644
--- a/m_librarian/inp.py
+++ b/m_librarian/inp.py
@@ -1,22 +1,23 @@
 
-__all__ = ['import_inpx']
-
 import os
 from zipfile import ZipFile
-from sqlobject import sqlhub, SQLObjectNotFound
+from sqlobject import sqlhub
+from sqlobject.sqlbuilder import Select
 from .db import Author, Book, Extension, Genre, Language, \
     insert_name, insert_author
 
+__all__ = ['import_inpx']
+
 
 EOT = chr(4)  # INP field separator
 
 
 def split_line(line):
     parts = line.strip().split(EOT)
-    l = len(parts)
-    if l < 11:
+    _l = len(parts)
+    if _l < 11:
         raise ValueError('Unknown INP structure: "%s"' % line)
-    if l == 11:  # Standard structure
+    if _l == 11:  # Standard structure
         parts.append(None)  # Emulate lang
     else:  # New structure
         parts = parts[:12]
@@ -26,12 +27,6 @@ def split_line(line):
 def import_inp_line(archive, parts):
     authors, genres, title, series, ser_no, file, size, lib_id, deleted, \
         extension, date, language = parts
-    try:
-        Book.archive_file_idx.get(archive, file)
-    except SQLObjectNotFound:
-        pass
-    else:
-        return
     try:
         ser_no = int(ser_no)
     except ValueError:
@@ -45,8 +40,13 @@ def import_inp_line(archive, parts):
                 lib_id=lib_id, deleted=deleted,
                 extension=extension_row, date=date,
                 language=language_row)
-    for author in authors.split(':'):
+    authors = authors.split(':')
+    seen_authors = set()
+    for author in authors:
         if author:
+            if author in seen_authors:
+                continue
+            seen_authors.add(author)
             alist = author.split(',', 2)
             surname = alist[0]
             if len(alist) > 1:
@@ -66,20 +66,45 @@ def import_inp_line(archive, parts):
 
 
 def import_inp(archive, inp):
+    files = set()
+    connection = sqlhub.processConnection
+    for file, in connection.queryAll(connection.sqlrepr(
+            Select(Book.q.file, Book.q.archive == archive))):
+        files.add(file)
     for line in inp:
-        import_inp_line(archive, split_line(line))
+        line = line.decode('utf-8')
+        parts = split_line(line)
+        file = parts[5]
+        if file not in files:
+            files.add(file)
+            import_inp_line(archive, parts)
 
 
-def import_inpx(path):
+def import_inpx(path, pbar_cb=None):
     inpx = ZipFile(path)
+    if pbar_cb:
+        inp_count = 0
+        for name in inpx.namelist():
+            ext = os.path.splitext(name)[1]
+            if ext == '.inp':
+                inp_count += 1
+        pbar_cb.set_max(inp_count)
+    inp_count = 0
     for name in inpx.namelist():
         archive, ext = os.path.splitext(name)
         if ext != '.inp':
             continue
+        if pbar_cb:
+            inp_count += 1
+            pbar_cb.display(inp_count)
         inp = inpx.open(name)
         sqlhub.doInTransaction(import_inp, archive + '.zip', inp)
         inp.close()
     connection = sqlhub.processConnection
-    if connection.dbName in ('postgres', 'sqlite'):
+    if connection.dbName == 'postgres':
         for table in Author, Book, Extension, Genre, Language:
             connection.query("VACUUM %s" % table.sqlmeta.table)
+    elif connection.dbName == 'sqlite':
+        connection.query("VACUUM")
+    if pbar_cb:
+        pbar_cb.close()