X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/05b19dad8ca44136e064abcc6beff5dac4489c60..565aa32ff31bdbe6e2aaf9f47b124ac1f883a990:/librarian/picture.py?ds=inline diff --git a/librarian/picture.py b/librarian/picture.py index 7830528..b665a34 100644 --- a/librarian/picture.py +++ b/librarian/picture.py @@ -1,5 +1,5 @@ - -from dcparser import (as_person, as_date, Field, WorkInfo, DCNS) +# -*- coding: utf-8 -*- +from dcparser import Field, WorkInfo, DCNS from librarian import (RDFNS, ValidationError, NoDublinCore, ParseError, WLURI) from xml.parsers.expat import ExpatError from os import path @@ -10,19 +10,16 @@ import re class WLPictureURI(WLURI): - _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/obraz/' - '(?P[-a-z0-9]+)(/(?P[a-z]{3}))?/?$') - - def __init__(self, *args, **kw): - super(WLPictureURI, self).__init__(*args, **kw) + _re_wl_uri = re.compile('http://wolnelektury.pl/katalog/obraz/(?P[-a-z0-9]+)/?$') @classmethod - def from_slug_and_lang(cls, slug, lang): + def from_slug(cls, slug): uri = 'http://wolnelektury.pl/katalog/obraz/%s/' % slug return cls(uri) - def filename_stem(self): - return self.slug + +def as_wlpictureuri_strict(text): + return WLPictureURI.strict(text) class PictureInfo(WorkInfo): @@ -30,24 +27,24 @@ class PictureInfo(WorkInfo): Dublin core metadata for a picture """ FIELDS = ( - Field(DCNS('format.dimensions.digital'), 'dimensions', required=False), - Field(DCNS('format.dimensions.original'), 'dimensions_original', required=False), - Field(DCNS('format.physical'), 'physical', required=False), + 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('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), - ) + Field(DCNS('identifier.url'), 'url', WLPictureURI, strict=as_wlpictureuri_strict), + ) - def validate(self): - """ - WorkInfo has a language validation code only, which we do not need. - """ - pass - class ImageStore(object): - EXT = ['gif', 'jpeg', 'png', 'swf', 'psd', 'bmp' - 'tiff', 'tiff', 'jpc', 'jp2', 'jpf', 'jb2', 'swc', - 'aiff', 'wbmp', 'xbm'] + 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', @@ -56,7 +53,7 @@ class ImageStore(object): def __init__(self, dir_): self.dir = dir_ - return super(ImageStore, self).__init__() + super(ImageStore, self).__init__() def path(self, slug, mime_type): """ @@ -97,20 +94,16 @@ class WLPicture(object): else: self.picture_info = None - @classmethod - def from_string(cls, xml, *args, **kwargs): - return cls.from_file(StringIO(xml), *args, **kwargs) - @classmethod def from_file(cls, xmlfile, parse_dublincore=True, image_store=None): # first, prepare for parsing if isinstance(xmlfile, basestring): - file = open(xmlfile, 'rb') + xmlfile = open(xmlfile, 'rb') try: - data = file.read() + data = xmlfile.read() finally: - file.close() + xmlfile.close() else: data = xmlfile.read() @@ -124,7 +117,7 @@ class WLPicture(object): image_store = ImageStore(path.dirname(xmlfile.name)) try: - parser = etree.XMLParser(remove_blank_text=False) + parser = etree.XMLParser() tree = etree.parse(StringIO(data.encode('utf-8')), parser) return cls(tree, parse_dublincore=parse_dublincore, image_store=image_store) @@ -149,3 +142,25 @@ class WLPicture(object): def image_file(self, *args, **kwargs): return open(self.image_path, *args, **kwargs) + + def partiter(self): + """ + Iterates the parts of this picture and returns them and their metadata + """ + for part in self.edoc.iter("div"): + pd = {'themes': [], 'object': None, 'type': part.get('type')} + if pd['type'] == 'area': + pd['coords'] = ((int(part.get('x1')), int(part.get('y1'))), + (int(part.get('x2')), int(part.get('y2')))) + + parent = part + while True: + parent = parent.getparent() + if parent is None: + break + if parent.tag == 'sem': + if parent.get('type') == 'theme': + pd['themes'] += map(unicode.strip, unicode(parent.get('theme')).split(',')) + elif parent.get('type') == 'object' and pd['object'] is None: + pd['object'] = parent.get('object') + yield pd