Autodetect if a zip/rar file is actually a comic. Fixes #4880 (RAR/ZIP file Autodetection to treat them as CBR/CBZ)

src/calibre/ebooks/metadata/archive.py

@@ -6,17 +6,21 @@ __license__   = 'GPL v3'
 __copyright__ = '2010, Kovid Goyal <kovid@kovidgoyal.net>'
 __docformat__ = 'restructuredtext en'
 
-import textwrap, os
+import os
 from contextlib import closing
 
 from calibre.customize import FileTypePlugin
 
+def is_comic(list_of_names):
+    extensions = set([x.rpartition('.')[-1].lower() for x in list_of_names])
+    return len(extensions) == 1 and iter(extensions).next() in ('jpg', 'jpeg', 'png')
+
 class ArchiveExtract(FileTypePlugin):
     name = 'Archive Extract'
     author = 'Kovid Goyal'
-    description = textwrap.dedent(_('''\
+    description = _('Extract common e-book formats from archives '
-        Extract common e-book formats from archives (zip/rar) files.
+        '(zip/rar) files. Also try to autodetect if they are actually '
-    '''))
+        'cbz/cbr files.')
     file_types = set(['zip', 'rar'])
     supported_platforms = ['windows', 'osx', 'linux']
     on_import = True
@@ -35,6 +39,13 @@ class ArchiveExtract(FileTypePlugin):
             fnames = zf.namelist()
 
         fnames = [x for x in fnames if '.' in x]
+        if is_comic(fnames):
+            ext = '.cbr' if is_rar else '.cbz'
+            of = self.temporary_file('_archive_extract'+ext)
+            with open(archive, 'rb') as f:
+                of.write(f.read())
+            of.close()
+            return of.name
         if len(fnames) > 1 or not fnames:
             return archive
         fname = fnames[0]

src/calibre/ebooks/metadata/rar.py

@@ -13,6 +13,9 @@ from calibre.ptempfile import PersistentTemporaryFile
 from calibre.libunrar import extract_member, names
 
 def get_metadata(stream):
+    from calibre.ebooks.metadata.archive import is_comic
+    from calibre.ebooks.metadata.meta import get_metadata
+
     path = getattr(stream, 'name', False)
     if not path:
         pt = PersistentTemporaryFile('_rar-meta.rar')
@@ -21,6 +24,8 @@ def get_metadata(stream):
         path = pt.name
     path = os.path.abspath(path)
     file_names = list(names(path))
+    if is_comic(file_names):
+        return get_metadata(stream, 'cbr')
     for f in file_names:
         stream_type = os.path.splitext(f)[1].lower()
         if stream_type:
@@ -29,7 +34,6 @@ def get_metadata(stream):
                                'rb', 'imp', 'pdf', 'lrf'):
                 data = extract_member(path, match=None, name=f)[1]
                 stream = StringIO(data)
-                from calibre.ebooks.metadata.meta import get_metadata
                 return get_metadata(stream, stream_type)
     raise ValueError('No ebook found in RAR archive')
 

src/calibre/ebooks/metadata/zip.py

@@ -8,15 +8,21 @@ from cStringIO import StringIO
 
 
 def get_metadata(stream):
+    from calibre.ebooks.metadata.meta import get_metadata
+    from calibre.ebooks.metadata.archive import is_comic
     stream_type = None
     zf = ZipFile(stream, 'r')
-    for f in zf.namelist():
+    names = zf.namelist()
+    if is_comic(names):
+        # Is probably a comic
+        return get_metadata(stream, 'cbz')
+
+    for f in names:
         stream_type = os.path.splitext(f)[1].lower()
         if stream_type:
             stream_type = stream_type[1:]
             if stream_type in ('lit', 'opf', 'prc', 'mobi', 'fb2', 'epub',
                                'rb', 'imp', 'pdf', 'lrf'):
-                from calibre.ebooks.metadata.meta import get_metadata
                 stream = StringIO(zf.read(f))
                 return get_metadata(stream, stream_type)
     raise ValueError('No ebook found in ZIP archive')