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
  13
  14 wl_dc_reader = metadata.MetadataReader(
  15     fields={
  16     'title':       ('textList', 'rdf:RDF/rdf:Description/dc:title/text()'),
  17     'creator':     ('textList', 'rdf:RDF/rdf:Description/dc:creator/text()'),
  18     '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()'),
  19     'description': ('textList', 'rdf:RDF/rdf:Description/dc:description/text()'),
  20     'publisher':   ('textList', 'rdf:RDF/rdf:Description/dc:publisher/text()'),
  21     '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()'),
  22     'date':        ('textList', 'rdf:RDF/rdf:Description/dc:date/text()'),
  23     'type':        ('textList', 'rdf:RDF/rdf:Description/dc:type/text()'),
  24     'format':      ('textList', 'rdf:RDF/rdf:Description/dc:format/text()'),
  25     'identifier':  ('textList', 'rdf:RDF/rdf:Description/dc:identifier.url/text()'),
  26     'source':      ('textList', 'rdf:RDF/rdf:Description/dc:source/text()'),
  27     'language':    ('textList', 'rdf:RDF/rdf:Description/dc:language/text()'),
  28     #    'relation':    ('textList', 'rdf:RDF/rdf:Description/dc:relation/text()'),
  29     #    'coverage':    ('textList', 'rdf:RDF/rdf:Description/dc:coverage/text()'),
  30     'rights':      ('textList', 'rdf:RDF/rdf:Description/dc:rights/text()')
  31     },
  32     namespaces={
  33     'dc': 'http://purl.org/dc/elements/1.1/',
  34     'rdf': 'http://www.w3.org/1999/02/22-rdf-syntax-ns#'}
  35     )
  36
  37
  38 class Catalogue(common.ResumptionOAIPMH):
  39     TAG_CATEGORIES = ['author', 'epoch', 'kind', 'genre']
  40
  41     def __init__(self):
  42         super(Catalogue, self).__init__()
  43
  44         # earliest change
  45         year_zero = datetime(1990, 1, 1, 0, 0, 0)
  46
  47         try:
  48             earliest_change = \
  49                 Book.objects.order_by('changed_at')[0].changed_at
  50         except: earliest_change = year_zero
  51
  52         try:
  53             earliest_delete = \
  54                 Deleted.objects.ordery_by('deleted_at')[0].deleted_at
  55         except: earliest_delete = year_zero
  56
  57         self.earliest_datestamp = earliest_change <= earliest_delete and \
  58             earliest_change or earliest_delete
  59
  60         # admins
  61         self.admin_emails = [u.email for u in User.objects.filter(is_superuser=True)]
  62
  63     def metadata(self, book):
  64         xml = etree.parse(book.xml_file)
  65         md = wl_dc_reader(xml)
  66         return md.getMap()
  67
  68     def record_for_book(self, book, headers_only=False):
  69         meta = None
  70         if isinstance(book, Book):
  71             header = common.Header(book.slug, 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(book.slug, book.deleted_at, [], True)
  77             if not headers_only:
  78                 meta = common.Metadata({})
  79             about = None
  80         return header, meta, about
  81
  82     def identify(self, **kw):
  83         ident = common.Identify(
  84             'Wolne Lektury',  # generate
  85             '%s/oaipmh' % WL_BASE,  # generate
  86             '1.1',  # version
  87             self.admin_emails,  # adminEmails
  88             self.earliest_datestamp,  # earliest datestamp of any change
  89             'persistent',  # deletedRecord
  90             'YYYY-MM-DDThh:mm:ssZ',  # granularity
  91             'identity',  # compression
  92             []  # descriptions
  93             )
  94         return ident
  95
  96     def books(self, tag, from_, until):
  97         if tag:
  98             books = Book.tagged.with_all([tag])
  99         else:
 100             books = Book.objects.all()
 101         deleted = Deleted.objects.all()
 102
 103         books = books.order_by('changed_at')
 104         if from_:
 105             books = books.filter(changed_at__gte=from_)
 106             deleted = deleted.filter(deleted_at__gte=from_)
 107         if until:
 108             books = books.filter(changed_at__lte=until)
 109             deleted = deleted.filter(deleted_at__gte=until)
 110         return list(books) + list(deleted)
 111
 112     @staticmethod
 113     def tag_to_setspec(tag):
 114         return "%s:%s" % (tag.category, tag.slug)
 115
 116     @staticmethod
 117     def setspec_to_tag(s):
 118         if not s: return None
 119         cs = s.split(':')
 120         if len(cs) == 2:
 121             if not cs[0] in Catalogue.TAG_CATEGORIES:
 122                 raise error.NoSetHierarchyError()
 123             tag = Tag.objects.get(slug=cs[1], category=cs[0])
 124             return tag
 125         raise error.NoSetHierarchyError()
 126
 127     def getRecord(self, **kw):
 128         """
 129 Returns (header, metadata, about) for given record.
 130         """
 131         slug = kw['record']
 132         try:
 133             book = Book.objects.get(slug=slug)
 134             return self.record_for_book(book)
 135         except Book.DoesNotExist:
 136             book_type = ContentType.objects.get_for_model(Book)
 137             try:
 138                 deleted_book = Deleted.objects.get(content_type=book_type,
 139                                                   slug=slug)
 140             except:
 141                 raise error.NoRecordsMatchError()
 142             return self.record_for_book(deleted_book)
 143
 144     def listIdentifiers(self, **kw):
 145         records = [self.record_for_book(book, headers_only=True) for
 146                    book in self.books(
 147                        self.setspec_to_tag(
 148                            kw.get('set', None)),
 149                            kw.get('from', None),
 150                            kw.get('until', None))]
 151         return records
 152
 153     def listRecords(self, **kw):
 154         """
 155 can get a resumptionToken kw.
 156 returns result, token
 157         """
 158         records = [self.record_for_book(book) for
 159                    book in self.books(
 160                        self.setspec_to_tag(
 161                            kw.get('set', None)),
 162                            kw.get('from', None),
 163                            kw.get('until', None))]
 164
 165         return records, None
 166
 167     def listMetadataFormats(self, **kw):
 168         return [('oaidc',
 169                  'http://www.openarchives.org/OAI/2.0/oai_dc.xsd',
 170                  server.NS_OAIDC)]
 171
 172     def listSets(self, **kw):
 173         tags = []
 174         for category in Catalogue.TAG_CATEGORIES:
 175             for tag in Tag.objects.filter(category=category):
 176                 tags.append(("%s:%s" % (tag.category, tag.slug),
 177                              tag.name,
 178                              tag.description))
 179         return tags