X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/fefdce4e24f9e397df5538fe6e7f54b5ece4d841..HEAD:/src/librarian/picture.py diff --git a/src/librarian/picture.py b/src/librarian/picture.py index 10d2ae7..8b2c882 100644 --- a/src/librarian/picture.py +++ b/src/librarian/picture.py @@ -1,8 +1,9 @@ -# -*- coding: utf-8 -*- -from __future__ import unicode_literals - +# This file is part of Librarian, licensed under GNU Affero GPLv3 or later. +# Copyright © Fundacja Wolne Lektury. See NOTICE for more information. +# from operator import and_ - +import functools +import io from .dcparser import Field, WorkInfo, DCNS from librarian import (RDFNS, ValidationError, NoDublinCore, ParseError, WLURI) from xml.parsers.expat import ExpatError @@ -10,16 +11,13 @@ from os import path from lxml import etree from lxml.etree import (XMLSyntaxError, XSLTApplyError, Element) import re -import six class WLPictureURI(WLURI): - _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/obraz/(?P[-a-z0-9]+)/?$') - - @classmethod - def from_slug(cls, slug): - uri = 'http://wolnelektury.pl/katalog/obraz/%s/' % slug - return cls(uri) + _re_wl_uri = re.compile( + 'http://wolnelektury.pl/katalog/obraz/(?P[-a-z0-9]+)/?$' + ) + template = 'http://wolnelektury.pl/katalog/obraz/%s/' def as_wlpictureuri_strict(text): @@ -34,27 +32,32 @@ class PictureInfo(WorkInfo): Field(DCNS('language'), 'language', required=False), 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('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('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): +class ImageStore: EXT = ['gif', 'jpeg', 'png', 'swf', 'psd', 'bmp' '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', - 'image/jp2', 'application/octet-stream', 'application/octet-stream', - 'application/x-shockwave-flash', 'image/iff', 'image/vnd.wap.wbmp', 'image/xbm'] + 'image/jp2', 'application/octet-stream', + 'application/octet-stream', 'application/x-shockwave-flash', + 'image/iff', 'image/vnd.wap.wbmp', 'image/xbm'] def __init__(self, dir_): super(ImageStore, self).__init__() @@ -68,7 +71,10 @@ class ImageStore(object): try: i = self.MIME.index(mime_type) except ValueError: - err = ValueError("Picture %s has unknown mime type: %s" % (slug, mime_type)) + err = ValueError( + "Picture %s has unknown mime type: %s" + % (slug, mime_type) + ) err.slug = slug err.mime_type = mime_type raise err @@ -77,7 +83,7 @@ class ImageStore(object): return path.join(self.dir, slug + '.' + ext) -class WLPicture(object): +class WLPicture: def __init__(self, edoc, parse_dublincore=True, image_store=None): self.edoc = edoc self.image_store = image_store @@ -87,13 +93,18 @@ class WLPicture(object): dc_path = './/' + RDFNS('RDF') if root_elem.tag != 'picture': - raise ValidationError("Invalid root element. Found '%s', should be 'picture'" % root_elem.tag) + raise ValidationError( + "Invalid root element. Found '%s', should be 'picture'" + % root_elem.tag + ) if parse_dublincore: self.rdf_elem = root_elem.find(dc_path) if self.rdf_elem is None: - raise NoDublinCore('Document has no DublinCore - which is required.') + raise NoDublinCore( + "Document has no DublinCore - which is required." + ) self.picture_info = PictureInfo.from_element(self.rdf_elem) else: @@ -102,13 +113,13 @@ class WLPicture(object): @classmethod def from_bytes(cls, xml, *args, **kwargs): - return cls.from_file(six.BytesIO(xml), *args, **kwargs) + return cls.from_file(io.BytesIO(xml), *args, **kwargs) @classmethod def from_file(cls, xmlfile, parse_dublincore=True, image_store=None): # first, prepare for parsing - if isinstance(xmlfile, six.text_type): + if isinstance(xmlfile, str): file = open(xmlfile, 'rb') try: data = file.read() @@ -117,10 +128,10 @@ class WLPicture(object): else: data = xmlfile.read() - if not isinstance(data, six.text_type): + if not isinstance(data, str): data = data.decode('utf-8') - data = data.replace(u'\ufeff', '') + data = data.replace('\ufeff', '') # assume images are in the same directory if image_store is None and getattr(xmlfile, 'name', None): @@ -128,9 +139,10 @@ class WLPicture(object): try: parser = etree.XMLParser(remove_blank_text=False) - tree = etree.parse(six.BytesIO(data.encode('utf-8')), parser) + tree = etree.parse(io.BytesIO(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) as e: @@ -139,7 +151,9 @@ class WLPicture(object): @property def mime_type(self): if self.picture_info is None: - raise ValueError('DC is not loaded, hence we don\'t know the image type') + raise ValueError( + "DC is not loaded, hence we don't know the image type." + ) return self.picture_info.mime_type @property @@ -163,7 +177,9 @@ class WLPicture(object): return [[0, 0], [-1, -1]] def has_all_props(node, props): - return reduce(and_, map(lambda prop: prop in node.attrib, props)) + return functools.reduce( + and_, map(lambda prop: prop in node.attrib, props) + ) if not has_all_props(area, ['x1', 'x2', 'y1', 'y2']): return None @@ -173,7 +189,8 @@ class WLPicture(object): def partiter(self): """ - Iterates the parts of this picture and returns them and their metadata + 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']")) +\ @@ -186,22 +203,30 @@ class WLPicture(object): pd['coords'] = coords def want_unicode(x): - if not isinstance(x, six.text_type): + if not isinstance(x, str): 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 [] + pd['object'] = ( + part.attrib['type'] == 'object' + and want_unicode(part.attrib.get('object', '')) + or None + ) + pd['themes'] = ( + part.attrib['type'] == 'theme' + and [part.attrib.get('theme', '')] + or [] + ) yield pd def load_frame_info(self): k = self.edoc.find("//sem[@object='kadr']") - + if k is not None: clip = self.get_sem_coords(k) self.frame = clip frm = Element("sem", {"type": "frame"}) - frm.append(k.iter("div").next()) + frm.append(next(k.iter("div"))) self.edoc.getroot().append(frm) k.getparent().remove(k) else: