Librarian in regular requirements.

[redakcja.git] / apps / catalogue / management / commands / import_wl.py

diff --git a/apps/catalogue/management/commands/import_wl.py b/apps/catalogue/management/commands/import_wl.py
old mode 100755 (executable)
new mode 100644 (file)
index 6836d36..45c9e33
--- a/apps/catalogue/management/commands/import_wl.py
+++ b/apps/catalogue/management/commands/import_wl.py
@@ -1,5 +1,6 @@
 # -*- coding: utf-8 -*-
 
+from collections import defaultdict
 import json
 from optparse import make_option
 import urllib2
@@ -30,28 +31,27 @@ class Command(BaseCommand):
         verbose = options.get('verbose')
 
         # Start transaction management.
-        transaction.commit_unless_managed()
         transaction.enter_transaction_management()
-        transaction.managed(True)
 
         if verbose:
-            print 'Reading currently managed files.'
-        slugs = {}
-        for b in Book.objects.all():
+            print 'Reading currently managed files (skipping hidden ones).'
+        slugs = defaultdict(list)
+        for b in Book.objects.exclude(slug__startswith='.').all():
             if verbose:
                 print b.slug
             text = b.materialize().encode('utf-8')
             try:
-                info = BookInfo.from_string(text)
+                info = BookInfo.from_bytes(text)
             except (ParseError, ValidationError):
                 pass
             else:
-                slugs[info.slug] = b
+                slugs[info.slug].append(b)
 
         book_count = 0
         commit_args = {
             "author_name": 'Platforma',
-            "description": 'Import from WL',
+            "description": 'Automatycznie zaimportowane z Wolnych Lektur',
+            "publishable": True,
         }
 
         if verbose:
@@ -59,15 +59,21 @@ class Command(BaseCommand):
         for book in json.load(urllib2.urlopen(WL_API)):
             book_detail = json.load(urllib2.urlopen(book['href']))
             xml_text = urllib2.urlopen(book_detail['xml']).read()
-            info = BookInfo.from_string(xml_text)
-            previous_book = slugs.get(info.slug, None)
-            if previous_book:
+            info = BookInfo.from_bytes(xml_text)
+            previous_books = slugs.get(info.slug)
+            if previous_books:
+                if len(previous_books) > 1:
+                    print self.style.ERROR("There is more than one book "
+                        "with slug %s:"), 
+                previous_book = previous_books[0]
                 comm = previous_book.slug
             else:
+                previous_book = None
                 comm = '*'
             print book_count, info.slug , '-->', comm
-            Book.import_xml_text(xml_text, title=info.title,
-                slug=info.slug, previous_book=slugs.get(info.slug, None))
+            Book.import_xml_text(xml_text, title=info.title[:255],
+                slug=info.slug[:128], previous_book=previous_book,
+                commit_args=commit_args)
             book_count += 1
 
         # Print results