apps/oai/handlers.py

   1 from oaipmh import server, common, metadata, error
   2 from catalogue.models import Book, Tag
   3 from api.models import Deleted
   4 from api.handlers import WL_BASE
   5 from librarian.dcparser import BookInfo
   6 from librarian import WLURI
   7 from django.contrib.contenttypes.models import ContentType
   8 from django.contrib.auth.models import User
   9 from datetime import datetime
  10 from lxml import etree
  11 from lxml.etree import ElementTree
  12 from django.db.models import Q
  13 from django.conf import settings
  14
  15 wl_dc_reader = metadata.MetadataReader(
  16     fields={
  17     'title':       ('textList', 'rdf:RDF/rdf:Description/dc:title/text()'),
  18     'creator':     ('textList', 'rdf:RDF/rdf:Description/dc:creator/text()'),
  19     'subject':     ('textList', 'rdf:RDF/rdf:Description/dc:subject.period/text() | rdf:RDF/rdf:Description/dc:subject.type/text() | rdf:RDF/rdf:Description/dc:subject.genre/text()'),
  20     'description': ('textList', 'rdf:RDF/rdf:Description/dc:description/text()'),
  21     'publisher':   ('textList', 'rdf:RDF/rdf:Description/dc:publisher/text()'),
  22     'contributor': ('textList', 'rdf:RDF/rdf:Description/dc:contributor.editor/text() | rdf:RDF/rdf:Description/dc:contributor.translator/text() | rdf:RDF/rdf:Description/dc:contributor.technical_editor/text()'),
  23     'date':        ('textList', 'rdf:RDF/rdf:Description/dc:date/text()'),
  24     'type':        ('textList', 'rdf:RDF/rdf:Description/dc:type/text()'),
  25     'format':      ('textList', 'rdf:RDF/rdf:Description/dc:format/text()'),
  26     'identifier':  ('textList', 'rdf:RDF/rdf:Description/dc:identifier.url/text()'),
  27     'source':      ('textList', 'rdf:RDF/rdf:Description/dc:source/text()'),
  28     'language':    ('textList', 'rdf:RDF/rdf:Description/dc:language/text()'),
  29     #    'relation':    ('textList', 'rdf:RDF/rdf:Description/dc:relation/text()'),
  30     #    'coverage':    ('textList', 'rdf:RDF/rdf:Description/dc:coverage/text()'),
  31     'rights':      ('textList', 'rdf:RDF/rdf:Description/dc:rights/text()')
  32     },
  33     namespaces={
  34     'dc': 'http://purl.org/dc/elements/1.1/',
  35     'rdf': 'http://www.w3.org/1999/02/22-rdf-syntax-ns#'}
  36     )
  37
  38
  39 class Catalogue(common.ResumptionOAIPMH):
  40     TAG_CATEGORIES = ['author', 'epoch', 'kind', 'genre']
  41
  42     def __init__(self):
  43         super(Catalogue, self).__init__()
  44
  45         # earliest change
  46         year_zero = datetime(1990, 1, 1, 0, 0, 0)
  47
  48         try:
  49             earliest_change = \
  50                 Book.objects.order_by('changed_at')[0].changed_at
  51         except: earliest_change = year_zero
  52
  53         try:
  54             earliest_delete = \
  55                 Deleted.objects.exclude(slug__exact=u'').ordery_by('deleted_at')[0].deleted_at
  56         except: earliest_delete = year_zero
  57
  58         self.earliest_datestamp = earliest_change <= earliest_delete and \
  59             earliest_change or earliest_delete
  60
  61     def metadata(self, book):
  62         xml = etree.parse(book.xml_file)
  63         md = wl_dc_reader(xml)
  64         return md.getMap()
  65
  66     def record_for_book(self, book, headers_only=False):
  67         meta = None
  68         identifier = str(WLURI.from_slug(book.slug))
  69         if isinstance(book, Book):
  70             #            setSpec = map(self.tag_to_setspec, book.tags.filter(category__in=self.TAG_CATEGORIES))
  71             header = common.Header(identifier, book.changed_at, [], False)
  72             if not headers_only:
  73                 meta = common.Metadata(self.metadata(book))
  74             about = None
  75         elif isinstance(book, Deleted):
  76             header = common.Header(identifier, book.deleted_at, [], True)
  77             if not headers_only:
  78                 meta = common.Metadata({})
  79             about = None
  80         if headers_only:
  81             return header
  82         return header, meta, about
  83
  84     def identify(self, **kw):
  85         ident = common.Identify(
  86             'Wolne Lektury',  # generate
  87             '%s/oaipmh' % WL_BASE,  # generate
  88             '2.0',  # version
  89             [m[1] for m in settings.MANAGERS],  # adminEmails
  90             self.earliest_datestamp,  # earliest datestamp of any change
  91             'persistent',  # deletedRecord
  92             'YYYY-MM-DDThh:mm:ssZ',  # granularity
  93             ['identity'],  # compression
  94             []  # descriptions
  95             )
  96         return ident
  97
  98     def books(self, tag, from_, until):
  99         if tag:
 100             # we do not support sets, since they are problematic for deleted books.
 101             raise errror.NoSetHierarchyError("Wolne Lektury does not support sets.")
 102             # books = Book.tagged.with_all([tag])
 103         else:
 104             books = Book.objects.all()
 105         deleted = Deleted.objects.exclude(slug__exact=u'')
 106
 107         books = books.order_by('changed_at')
 108         deleted = deleted.order_by('deleted_at')
 109         if from_:
 110             books = books.filter(changed_at__gte=from_)
 111             deleted = deleted.filter(deleted_at__gte=from_)
 112         if until:
 113             books = books.filter(changed_at__lte=until)
 114             deleted = deleted.filter(deleted_at__lte=until)
 115         return list(books) + list(deleted)
 116
 117     @staticmethod
 118     def tag_to_setspec(tag):
 119         return "%s:%s" % (tag.category, tag.slug)
 120
 121     @staticmethod
 122     def setspec_to_tag(s):
 123         if not s: return None
 124         cs = s.split(':')
 125         if len(cs) == 2:
 126             if not cs[0] in Catalogue.TAG_CATEGORIES:
 127                 raise error.NoSetHierarchyError("No category part in set")
 128             tag = Tag.objects.get(slug=cs[1], category=cs[0])
 129             return tag
 130         raise error.NoSetHierarchyError("Setspec should have two components: category:slug")
 131
 132     def getRecord(self, **kw):
 133         """
 134 Returns (header, metadata, about) for given record.
 135         """
 136         slug = WLURI(kw['identifier']).slug
 137         try:
 138             book = Book.objects.get(slug=slug)
 139             return self.record_for_book(book)
 140         except Book.DoesNotExist:
 141             book_type = ContentType.objects.get_for_model(Book)
 142             try:
 143                 deleted_book = Deleted.objects.get(content_type=book_type,
 144                                                   slug=slug)
 145             except:
 146                 raise error.IdDoesNotExistError("No item for this identifier")
 147             return self.record_for_book(deleted_book)
 148
 149     def listIdentifiers(self, **kw):
 150         records = [self.record_for_book(book, headers_only=True) for
 151                    book in self.books(None,
 152                            kw.get('from_', None),
 153                            kw.get('until', None))]
 154         return records, None
 155
 156     def listRecords(self, **kw):
 157         """
 158 can get a resumptionToken kw.
 159 returns result, token
 160         """
 161         records = [self.record_for_book(book) for
 162                    book in self.books(None,
 163                            kw.get('from_', None),
 164                            kw.get('until', None))]
 165
 166         return records, None
 167
 168     def listMetadataFormats(self, **kw):
 169         return [('oai_dc',
 170                  'http://www.openarchives.org/OAI/2.0/oai_dc.xsd',
 171                  server.NS_OAIDC)]
 172
 173     def listSets(self, **kw):
 174         raise error.NoSetHierarchyError("Wolne Lektury does not support sets.")
 175         # tags = []
 176         # for category in Catalogue.TAG_CATEGORIES:
 177         #     for tag in Tag.objects.filter(category=category):
 178         #         tags.append(("%s:%s" % (tag.category, tag.slug),
 179         #                      tag.name,
 180         #                      tag.description))
 181         # return tags, None
 182
 183