From: Radek Czajka Date: Tue, 2 Nov 2010 13:45:04 +0000 (+0100) Subject: fixes #932: bookimport works with parent-child relations in unsorted xml files X-Git-Url: https://git.mdrn.pl/wolnelektury.git/commitdiff_plain/4175dca3688ed6ce25a270cdfe10f2814564821c?ds=inline;hp=-c fixes #932: bookimport works with parent-child relations in unsorted xml files --- 4175dca3688ed6ce25a270cdfe10f2814564821c diff --git a/apps/catalogue/management/commands/importbooks.py b/apps/catalogue/management/commands/importbooks.py index ffe0de8ec..be65bd9a1 100644 --- a/apps/catalogue/management/commands/importbooks.py +++ b/apps/catalogue/management/commands/importbooks.py @@ -44,7 +44,11 @@ class Command(BaseCommand): if not os.path.isdir(dir_name): print self.style.ERROR("%s: Not a directory. Skipping." % dir_name) else: - for file_name in sorted(os.listdir(dir_name)): + # files queue + files = sorted(os.listdir(dir_name)) + postponed = {} + while files: + file_name = files.pop(0) file_path = os.path.join(dir_name, file_name) file_base, ext = os.path.splitext(file_path) @@ -99,6 +103,17 @@ class Command(BaseCommand): file_path) files_skipped += 1 + except Book.DoesNotExist, e: + if file_name not in postponed or postponed[file_name] < files_imported: + # push it back into the queue, maybe the missing child will show up + if verbose: + print self.style.NOTICE('Waiting for missing children') + files.append(file_name) + postponed[file_name] = files_imported + else: + # we're in a loop, nothing's being imported - some child is really missing + raise e + # Print results print print "Results: %d files imported, %d skipped, %d total." % ( diff --git a/apps/catalogue/models.py b/apps/catalogue/models.py index eb55b985d..b1ffdb63d 100644 --- a/apps/catalogue/models.py +++ b/apps/catalogue/models.py @@ -395,6 +395,17 @@ class Book(models.Model): from markupstring import MarkupString from django.core.files.storage import default_storage + # check for parts before we do anything + children = [] + if hasattr(book_info, 'parts'): + for part_url in book_info.parts: + base, slug = part_url.rsplit('/', 1) + try: + children.append(Book.objects.get(slug=slug)) + except Book.DoesNotExist, e: + raise Book.DoesNotExist(_('Book with slug = "%s" does not exist.') % slug) + + # Read book metadata book_base, book_slug = book_info.url.rsplit('/', 1) book, created = Book.objects.get_or_create(slug=book_slug) @@ -435,16 +446,10 @@ class Book(models.Model): book_tag = book.book_tag() - if hasattr(book_info, 'parts'): - for n, part_url in enumerate(book_info.parts): - base, slug = part_url.rsplit('/', 1) - try: - child_book = Book.objects.get(slug=slug) - child_book.parent = book - child_book.parent_number = n - child_book.save() - except Book.DoesNotExist, e: - raise Book.DoesNotExist(_('Book with slug = "%s" does not exist.') % slug) + for n, child_book in enumerate(children): + child_book.parent = book + child_book.parent_number = n + child_book.save() # Save XML and HTML files book.xml_file.save('%s.xml' % book.slug, raw_file, save=False)