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 django.contrib.contenttypes.models import ContentType
   7 from django.contrib.auth.models import User
   8 from datetime import datetime
   9 from lxml import etree
  10 from lxml.etree import ElementTree
  11 from django.db.models import Q
  12 from django.conf import settings
  13 from django.contrib.sites.models import Site
  14
  15
  16 wl_dc_reader = metadata.MetadataReader(
  17     fields={
  18     'title':       ('textList', 'rdf:RDF/rdf:Description/dc:title/text()'),
  19     'creator':     ('textList', 'rdf:RDF/rdf:Description/dc:creator/text()'),
  20     '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()'),
  21     'description': ('textList', 'rdf:RDF/rdf:Description/dc:description/text()'),
  22     'publisher':   ('textList', 'rdf:RDF/rdf:Description/dc:publisher/text()'),
  23     '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()'),
  24     'date':        ('textList', 'rdf:RDF/rdf:Description/dc:date/text()'),
  25     'type':        ('textList', 'rdf:RDF/rdf:Description/dc:type/text()'),
  26     'format':      ('textList', 'rdf:RDF/rdf:Description/dc:format/text()'),
  27     'identifier':  ('textList', 'rdf:RDF/rdf:Description/dc:identifier.url/text()'),
  28     'source':      ('textList', 'rdf:RDF/rdf:Description/dc:source/text()'),
  29     'language':    ('textList', 'rdf:RDF/rdf:Description/dc:language/text()'),
  30     'isPartOf':     ('textList', 'rdf:RDF/rdf:Description/dc:relation.isPartOf/text()'),
  31     'hasPart':     ('textList', 'rdf:RDF/rdf:Description/dc:relation.hasPart/text()'),
  32     #    'relation':    ('textList', 'rdf:RDF/rdf:Description/dc:relation/text()'),
  33     #    'coverage':    ('textList', 'rdf:RDF/rdf:Description/dc:coverage/text()'),
  34     'rights':      ('textList', 'rdf:RDF/rdf:Description/dc:rights/text()')
  35     },
  36     namespaces={
  37     'dc': 'http://purl.org/dc/elements/1.1/',
  38     'rdf': 'http://www.w3.org/1999/02/22-rdf-syntax-ns#'}
  39     )
  40
  41
  42 NS_DCTERMS = "http://purl.org/dc/terms/"
  43
  44
  45 def nsdcterms(name):
  46     return '{%s}%s' % (NS_DCTERMS, name)
  47
  48
  49 class Catalogue(common.ResumptionOAIPMH):
  50     TAG_CATEGORIES = ['author', 'epoch', 'kind', 'genre']
  51
  52     def __init__(self, metadata_registry):
  53         super(Catalogue, self).__init__()
  54         self.metadata_registry = metadata_registry
  55
  56         self.oai_id = "oai:" + Site.objects.get_current().domain + ":%s"
  57
  58         # earliest change
  59         year_zero = datetime(1990, 1, 1, 0, 0, 0)
  60
  61         try:
  62             earliest_change = \
  63                 Book.objects.order_by('changed_at')[0].changed_at
  64         except: earliest_change = year_zero
  65
  66         try:
  67             earliest_delete = \
  68                 Deleted.objects.exclude(slug__exact=u'').ordery_by('deleted_at')[0].deleted_at
  69         except: earliest_delete = year_zero
  70
  71         self.earliest_datestamp = earliest_change <= earliest_delete and \
  72             earliest_change or earliest_delete
  73
  74     def metadata(self, book):
  75         try:
  76             xml = etree.parse(book.xml_file)
  77         finally:
  78             book.xml_file.close()
  79         md = wl_dc_reader(xml)
  80         return md.getMap()
  81
  82     def record_for_book(self, book, headers_only=False):
  83         meta = None
  84         identifier = self.slug_to_identifier(book.slug)
  85         if isinstance(book, Book):
  86             #            setSpec = map(self.tag_to_setspec, book.tags.filter(category__in=self.TAG_CATEGORIES))
  87             header = common.Header(identifier, book.changed_at, [], False)
  88             if not headers_only:
  89                 meta = common.Metadata(self.metadata(book))
  90             about = None
  91         elif isinstance(book, Deleted):
  92             header = common.Header(identifier, book.deleted_at, [], True)
  93             if not headers_only:
  94                 meta = common.Metadata({})
  95             about = None
  96         if headers_only:
  97             return header
  98         return header, meta, about
  99
 100     def identify(self, **kw):
 101         ident = common.Identify(
 102             'Wolne Lektury',  # generate
 103             '%s/oaipmh' % WL_BASE,  # generate
 104             '2.0',  # version
 105             [m[1] for m in settings.MANAGERS],  # adminEmails
 106             self.earliest_datestamp,  # earliest datestamp of any change
 107             'persistent',  # deletedRecord
 108             'YYYY-MM-DDThh:mm:ssZ',  # granularity
 109             ['identity'],  # compression
 110             []  # descriptions
 111             )
 112         return ident
 113
 114     def books(self, tag, from_, until):
 115         if tag:
 116             # we do not support sets, since they are problematic for deleted books.
 117             raise error.NoSetHierarchyError("Wolne Lektury does not support sets.")
 118             # books = Book.tagged.with_all([tag])
 119         else:
 120             books = Book.objects.all()
 121         deleted = Deleted.objects.exclude(slug__exact=u'')
 122
 123         books = books.order_by('changed_at')
 124         deleted = deleted.order_by('deleted_at')
 125         if from_:
 126             books = books.filter(changed_at__gte=from_)
 127             deleted = deleted.filter(deleted_at__gte=from_)
 128         if until:
 129             books = books.filter(changed_at__lte=until)
 130             deleted = deleted.filter(deleted_at__lte=until)
 131         return list(books) + list(deleted)
 132
 133     @staticmethod
 134     def tag_to_setspec(tag):
 135         return "%s:%s" % (tag.category, tag.slug)
 136
 137     @staticmethod
 138     def setspec_to_tag(s):
 139         if not s: return None
 140         cs = s.split(':')
 141         if len(cs) == 2:
 142             if not cs[0] in Catalogue.TAG_CATEGORIES:
 143                 raise error.NoSetHierarchyError("No category part in set")
 144             tag = Tag.objects.get(slug=cs[1], category=cs[0])
 145             return tag
 146         raise error.NoSetHierarchyError("Setspec should have two components: category:slug")
 147
 148     def getRecord(self, **kw):
 149         """
 150 Returns (header, metadata, about) for given record.
 151         """
 152         slug = self.identifier_to_slug(kw['identifier'])
 153         try:
 154             book = Book.objects.get(slug=slug)
 155             return self.record_for_book(book)
 156         except Book.DoesNotExist:
 157             book_type = ContentType.objects.get_for_model(Book)
 158             try:
 159                 deleted_book = Deleted.objects.get(content_type=book_type,
 160                                                   slug=slug)
 161             except:
 162                 raise error.IdDoesNotExistError("No item for this identifier")
 163             return self.record_for_book(deleted_book)
 164
 165     def validate_kw(self, kw):
 166         if 'resumptionToken' in kw:
 167             raise error.BadResumptionTokenError("No resumption token support at this point")
 168         if 'metadataPrefix' in kw and not self.metadata_registry.hasWriter(kw['metadataPrefix']):
 169             raise error.CannotDisseminateFormatError("This format is not supported")
 170
 171     def identifier_to_slug(self, ident):
 172         return ident.split(':')[-1]
 173
 174     def slug_to_identifier(self, slug):
 175         return self.oai_id % slug
 176
 177     def listIdentifiers(self, **kw):
 178         self.validate_kw(kw)
 179         records = [self.record_for_book(book, headers_only=True) for
 180                    book in self.books(None,
 181                            kw.get('from_', None),
 182                            kw.get('until', None))]
 183         return records, None
 184
 185     def listRecords(self, **kw):
 186         """
 187 can get a resumptionToken kw.
 188 returns result, token
 189         """
 190         self.validate_kw(kw)
 191         records = [self.record_for_book(book) for
 192                    book in self.books(None,
 193                            kw.get('from_', None),
 194                            kw.get('until', None))]
 195
 196         return records, None
 197
 198     def listMetadataFormats(self, **kw):
 199         formats = [
 200             ('oai_dc',
 201              'http://www.openarchives.org/OAI/2.0/oai_dc.xsd',
 202              server.NS_OAIDC),
 203             ('qdc',
 204              'http://dublincore.org/schemas/xmls/qdc/2006/01/06/dcterms.xsd',
 205              NS_DCTERMS)]
 206         if 'identifier' in kw:
 207             slug = self.identifier_to_slug(kw['identifier'])
 208             try:
 209                 b = Book.objects.get(slug=slug)
 210                 return formats
 211             except:
 212                 try:
 213                     d = Deleted.objects.get(slug=slug)
 214                     return []
 215                 except:
 216                     raise error.IdDoesNotExistError("This id does not exist")
 217         else:
 218             return formats
 219
 220     def listSets(self, **kw):
 221         raise error.NoSetHierarchyError("Wolne Lektury does not support sets.")
 222         # tags = []
 223         # for category in Catalogue.TAG_CATEGORIES:
 224         #     for tag in Tag.objects.filter(category=category):
 225         #         tags.append(("%s:%s" % (tag.category, tag.slug),
 226         #                      tag.name,
 227         #                      tag.description))
 228         # return tags, None