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