allow multiple dc:publisher

[librarian.git] / librarian / picture.py

diff --git a/librarian/picture.py b/librarian/picture.py
index 5d644d7..1aa1d07 100644 (file)
--- a/librarian/picture.py
+++ b/librarian/picture.py
@@ -1,5 +1,7 @@
+# -*- coding: utf-8 -*-
+from operator import and_
 
-from dcparser import (as_person, as_date, Field, WorkInfo, DCNS)
+from dcparser import Field, WorkInfo, DCNS
 from librarian import (RDFNS, ValidationError, NoDublinCore, ParseError, WLURI)
 from xml.parsers.expat import ExpatError
 from os import path
@@ -7,18 +9,17 @@ from StringIO import StringIO
 from lxml import etree
 from lxml.etree import (XMLSyntaxError, XSLTApplyError, Element)
 import re
-from functools import *
-from operator import *
+
 
 class WLPictureURI(WLURI):
-    _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/obraz/'
-            '(?P<slug>[-a-z0-9]+)/?$')
+    _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/obraz/(?P<slug>[-a-z0-9]+)/?$')
 
     @classmethod
     def from_slug(cls, slug):
         uri = 'http://wolnelektury.pl/katalog/obraz/%s/' % slug
         return cls(uri)
 
+
 def as_wlpictureuri_strict(text):
     return WLPictureURI.strict(text)
 
@@ -32,21 +33,21 @@ class PictureInfo(WorkInfo):
         Field(DCNS('subject.period'), 'epochs', salias='epoch', multiple=True),
         Field(DCNS('subject.type'), 'kinds', salias='kind', multiple=True),
         Field(DCNS('subject.genre'), 'genres', salias='genre', multiple=True, required=False),
+        Field(DCNS('subject.style'), 'styles', salias='style', multiple=True, required=False),
 
         Field(DCNS('format.dimensions'), 'dimensions', required=False),
         Field(DCNS('format.checksum.sha1'), 'sha1', required=True),
         Field(DCNS('description.medium'), 'medium', required=False),
         Field(DCNS('description.dimensions'), 'original_dimensions', required=False),
         Field(DCNS('format'), 'mime_type', required=False),
-        Field(DCNS('identifier.url'), 'url', WLPictureURI,
-            strict=as_wlpictureuri_strict),
-        )
+        Field(DCNS('identifier.url'), 'url', WLPictureURI, strict=as_wlpictureuri_strict)
+    )
 
 
 class ImageStore(object):
     EXT = ['gif', 'jpeg', 'png', 'swf', 'psd', 'bmp'
-            'tiff', 'tiff', 'jpc', 'jp2', 'jpf', 'jb2', 'swc',
-            'aiff', 'wbmp', 'xbm']
+           'tiff', 'tiff', 'jpc', 'jp2', 'jpf', 'jb2', 'swc',
+           'aiff', 'wbmp', 'xbm']
     MIME = ['image/gif', 'image/jpeg', 'image/png',
             'application/x-shockwave-flash', 'image/psd', 'image/bmp',
             'image/tiff', 'image/tiff', 'application/octet-stream',
@@ -54,8 +55,8 @@ class ImageStore(object):
             'application/x-shockwave-flash', 'image/iff', 'image/vnd.wap.wbmp', 'image/xbm']
 
     def __init__(self, dir_):
+        super(ImageStore, self).__init__()
         self.dir = dir_
-        return super(ImageStore, self).__init__()
 
     def path(self, slug, mime_type):
         """
@@ -95,6 +96,7 @@ class WLPicture(object):
             self.picture_info = PictureInfo.from_element(self.rdf_elem)
         else:
             self.picture_info = None
+        self.frame = None
 
     @classmethod
     def from_string(cls, xml, *args, **kwargs):
@@ -119,14 +121,14 @@ class WLPicture(object):
         data = data.replace(u'\ufeff', '')
 
         # assume images are in the same directory
-        if image_store is None and xmlfile.name is not None:
+        if image_store is None and getattr(xmlfile, 'name', None):
             image_store = ImageStore(path.dirname(xmlfile.name))
 
         try:
             parser = etree.XMLParser(remove_blank_text=False)
             tree = etree.parse(StringIO(data.encode('utf-8')), parser)
 
-            me =  cls(tree, parse_dublincore=parse_dublincore, image_store=image_store)
+            me = cls(tree, parse_dublincore=parse_dublincore, image_store=image_store)
             me.load_frame_info()
             return me
         except (ExpatError, XMLSyntaxError, XSLTApplyError), e:
@@ -161,29 +163,31 @@ class WLPicture(object):
         def has_all_props(node, props):
             return reduce(and_, map(lambda prop: prop in node.attrib, props))
 
-        if has_all_props(area,  ['x1', 'x2', 'y1', 'y2']) == False:
+        if not has_all_props(area, ['x1', 'x2', 'y1', 'y2']):
             return None
-            
+
         def n(prop): return int(area.get(prop))
         return [[n('x1'), n('y1')], [n('x2'), n('y2')]]
-        
 
     def partiter(self):
         """
         Iterates the parts of this picture and returns them and their metadata
         """
         # omg no support for //sem[(@type='theme') or (@type='object')] ?
-        for part in list(self.edoc.iterfind("//sem[@type='theme']")) + list(self.edoc.iterfind("//sem[@type='object']")):
-            pd = {}
-            pd['type'] = part.get('type')
+        for part in list(self.edoc.iterfind("//sem[@type='theme']")) +\
+                list(self.edoc.iterfind("//sem[@type='object']")):
+            pd = {'type': part.get('type')}
 
             coords = self.get_sem_coords(part)
-            if coords is None: continue
+            if coords is None:
+                continue
             pd['coords'] = coords
 
             def want_unicode(x):
-                if not isinstance(x, unicode): return x.decode('utf-8')
-                else: return x
+                if not isinstance(x, unicode):
+                    return x.decode('utf-8')
+                else:
+                    return x
             pd['object'] = part.attrib['type'] == 'object' and want_unicode(part.attrib.get('object', u'')) or None
             pd['themes'] = part.attrib['type'] == 'theme' and [part.attrib.get('theme', u'')] or []
             yield pd