Merge branch 'sunburnt' of git+ssh://san.fnp/home/staging/dev/application into sunburnt

[wolnelektury.git] / apps / oai / handlers.py

diff --git a/apps/oai/handlers.py b/apps/oai/handlers.py
index 9e00b21..824cca9 100644 (file)
--- a/apps/oai/handlers.py
+++ b/apps/oai/handlers.py
@@ -3,13 +3,14 @@ from catalogue.models import Book, Tag
 from api.models import Deleted
 from api.handlers import WL_BASE
 from librarian.dcparser import BookInfo
 from api.models import Deleted
 from api.handlers import WL_BASE
 from librarian.dcparser import BookInfo

-from librarian import WLURI

 from django.contrib.contenttypes.models import ContentType
 from django.contrib.auth.models import User
 from datetime import datetime
 from lxml import etree
 from lxml.etree import ElementTree
 from django.db.models import Q
 from django.contrib.contenttypes.models import ContentType
 from django.contrib.auth.models import User
 from datetime import datetime
 from lxml import etree
 from lxml.etree import ElementTree
 from django.db.models import Q

+from django.conf import settings
+from django.contrib.sites.models import Site

 
 
 wl_dc_reader = metadata.MetadataReader(
 
 
 wl_dc_reader = metadata.MetadataReader(


@@ -26,6 +27,8 @@ wl_dc_reader = metadata.MetadataReader(
     'identifier':  ('textList', 'rdf:RDF/rdf:Description/dc:identifier.url/text()'),
     'source':      ('textList', 'rdf:RDF/rdf:Description/dc:source/text()'),
     'language':    ('textList', 'rdf:RDF/rdf:Description/dc:language/text()'),
     'identifier':  ('textList', 'rdf:RDF/rdf:Description/dc:identifier.url/text()'),
     'source':      ('textList', 'rdf:RDF/rdf:Description/dc:source/text()'),
     'language':    ('textList', 'rdf:RDF/rdf:Description/dc:language/text()'),

+    'isPartOf':     ('textList', 'rdf:RDF/rdf:Description/dc:relation.isPartOf/text()'),
+    'hasPart':     ('textList', 'rdf:RDF/rdf:Description/dc:relation.hasPart/text()'),

     #    'relation':    ('textList', 'rdf:RDF/rdf:Description/dc:relation/text()'),
     #    'coverage':    ('textList', 'rdf:RDF/rdf:Description/dc:coverage/text()'),
     'rights':      ('textList', 'rdf:RDF/rdf:Description/dc:rights/text()')
     #    'relation':    ('textList', 'rdf:RDF/rdf:Description/dc:relation/text()'),
     #    'coverage':    ('textList', 'rdf:RDF/rdf:Description/dc:coverage/text()'),
     'rights':      ('textList', 'rdf:RDF/rdf:Description/dc:rights/text()')


@@ -39,8 +42,11 @@ wl_dc_reader = metadata.MetadataReader(
 class Catalogue(common.ResumptionOAIPMH):
     TAG_CATEGORIES = ['author', 'epoch', 'kind', 'genre']
     
 class Catalogue(common.ResumptionOAIPMH):
     TAG_CATEGORIES = ['author', 'epoch', 'kind', 'genre']
     

-    def __init__(self):
+    def __init__(self, metadata_registry):

         super(Catalogue, self).__init__()
         super(Catalogue, self).__init__()

+        self.metadata_registry = metadata_registry
+
+        self.oai_id = "oai:"+Site.objects.get_current().domain+":%s"

 
         # earliest change
         year_zero = datetime(1990, 1, 1, 0, 0, 0)
 
         # earliest change
         year_zero = datetime(1990, 1, 1, 0, 0, 0)


@@ -52,26 +58,26 @@ class Catalogue(common.ResumptionOAIPMH):
 
         try:
             earliest_delete = \
 
         try:
             earliest_delete = \

-                Deleted.objects.ordery_by('deleted_at')[0].deleted_at
+                Deleted.objects.exclude(slug__exact=u'').ordery_by('deleted_at')[0].deleted_at

         except: earliest_delete = year_zero
 
         self.earliest_datestamp = earliest_change <= earliest_delete and \
             earliest_change or earliest_delete
 
         except: earliest_delete = year_zero
 
         self.earliest_datestamp = earliest_change <= earliest_delete and \
             earliest_change or earliest_delete
 

-        # admins
-        self.admin_emails = [u.email for u in User.objects.filter(is_superuser=True)]
-

     def metadata(self, book):
     def metadata(self, book):

-        xml = etree.parse(book.xml_file)
+        try:
+            xml = etree.parse(book.xml_file)
+        finally:
+            book.xml_file.close()

         md = wl_dc_reader(xml)
         return md.getMap()
 
     def record_for_book(self, book, headers_only=False):
         meta = None
         md = wl_dc_reader(xml)
         return md.getMap()
 
     def record_for_book(self, book, headers_only=False):
         meta = None

-        identifier = str(WLURI.from_slug(book.slug))
+        identifier = self.slug_to_identifier(book.slug)

         if isinstance(book, Book):
         if isinstance(book, Book):

-            setSpec = map(self.tag_to_setspec, book.tags.filter(category__in=self.TAG_CATEGORIES))
-            header = common.Header(identifier, book.changed_at, setSpec, False)
+            #            setSpec = map(self.tag_to_setspec, book.tags.filter(category__in=self.TAG_CATEGORIES))
+            header = common.Header(identifier, book.changed_at, [], False)

             if not headers_only:
                 meta = common.Metadata(self.metadata(book))
             about = None
             if not headers_only:
                 meta = common.Metadata(self.metadata(book))
             about = None


@@ -89,7 +95,7 @@ class Catalogue(common.ResumptionOAIPMH):
             'Wolne Lektury',  # generate
             '%s/oaipmh' % WL_BASE,  # generate
             '2.0',  # version
             'Wolne Lektury',  # generate
             '%s/oaipmh' % WL_BASE,  # generate
             '2.0',  # version

-            self.admin_emails,  # adminEmails
+            [m[1] for m in settings.MANAGERS],  # adminEmails

             self.earliest_datestamp,  # earliest datestamp of any change
             'persistent',  # deletedRecord
             'YYYY-MM-DDThh:mm:ssZ',  # granularity
             self.earliest_datestamp,  # earliest datestamp of any change
             'persistent',  # deletedRecord
             'YYYY-MM-DDThh:mm:ssZ',  # granularity


@@ -100,17 +106,18 @@ class Catalogue(common.ResumptionOAIPMH):
 
     def books(self, tag, from_, until):
         if tag:
 
     def books(self, tag, from_, until):
         if tag:

-            books = Book.tagged.with_all([tag])
+            # we do not support sets, since they are problematic for deleted books.
+            raise errror.NoSetHierarchyError("Wolne Lektury does not support sets.")
+            # books = Book.tagged.with_all([tag])

         else:
             books = Book.objects.all()
         else:
             books = Book.objects.all()

-        deleted = Deleted.objects.filter(slug__isnull=False)
+        deleted = Deleted.objects.exclude(slug__exact=u'')

 
         books = books.order_by('changed_at')
         deleted = deleted.order_by('deleted_at')
         if from_:
             books = books.filter(changed_at__gte=from_)
             deleted = deleted.filter(deleted_at__gte=from_)
 
         books = books.order_by('changed_at')
         deleted = deleted.order_by('deleted_at')
         if from_:
             books = books.filter(changed_at__gte=from_)
             deleted = deleted.filter(deleted_at__gte=from_)

-            print "DELETED:%s" % deleted

         if until:
             books = books.filter(changed_at__lte=until)
             deleted = deleted.filter(deleted_at__lte=until)
         if until:
             books = books.filter(changed_at__lte=until)
             deleted = deleted.filter(deleted_at__lte=until)


@@ -126,16 +133,16 @@ class Catalogue(common.ResumptionOAIPMH):
         cs = s.split(':')
         if len(cs) == 2:
             if not cs[0] in Catalogue.TAG_CATEGORIES:
         cs = s.split(':')
         if len(cs) == 2:
             if not cs[0] in Catalogue.TAG_CATEGORIES:

-                raise error.NoSetHierarchyError()
+                raise error.NoSetHierarchyError("No category part in set")

             tag = Tag.objects.get(slug=cs[1], category=cs[0])
             return tag
             tag = Tag.objects.get(slug=cs[1], category=cs[0])
             return tag

-        raise error.NoSetHierarchyError()
+        raise error.NoSetHierarchyError("Setspec should have two components: category:slug")

 
     def getRecord(self, **kw):
         """
 Returns (header, metadata, about) for given record.
         """
 
     def getRecord(self, **kw):
         """
 Returns (header, metadata, about) for given record.
         """

-        slug = WLURI(kw['identifier']).slug
+        slug = self.identifier_to_slug(kw['identifier'])

         try:
             book = Book.objects.get(slug=slug)
             return self.record_for_book(book)
         try:
             book = Book.objects.get(slug=slug)
             return self.record_for_book(book)


@@ -145,15 +152,25 @@ Returns (header, metadata, about) for given record.
                 deleted_book = Deleted.objects.get(content_type=book_type,
                                                   slug=slug)
             except:
                 deleted_book = Deleted.objects.get(content_type=book_type,
                                                   slug=slug)
             except:

-                raise error.NoRecordsMatchError()
+                raise error.IdDoesNotExistError("No item for this identifier")

             return self.record_for_book(deleted_book)
 
             return self.record_for_book(deleted_book)
 

+    def validate_kw(self, kw):
+        if 'resumptionToken' in kw:
+            raise error.BadResumptionTokenError("No resumption token support at this point")
+        if 'metadataPrefix' in kw and not self.metadata_registry.hasWriter(kw['metadataPrefix']):
+            raise error.CannotDisseminateFormatError("This format is not supported")
+
+    def identifier_to_slug(self, ident):
+        return ident.split(':')[-1]
+
+    def slug_to_identifier(self, slug):
+        return self.oai_id % slug
+

     def listIdentifiers(self, **kw):
     def listIdentifiers(self, **kw):

-        print "list identifiers %s" % (kw, )
+        self.validate_kw(kw)

         records = [self.record_for_book(book, headers_only=True) for
         records = [self.record_for_book(book, headers_only=True) for

-                   book in self.books(
-                       self.setspec_to_tag(
-                           kw.get('set', None)),
+                   book in self.books(None,

                            kw.get('from_', None),
                            kw.get('until', None))]
         return records, None
                            kw.get('from_', None),
                            kw.get('until', None))]
         return records, None


@@ -163,27 +180,40 @@ Returns (header, metadata, about) for given record.
 can get a resumptionToken kw.
 returns result, token
         """
 can get a resumptionToken kw.
 returns result, token
         """

+        self.validate_kw(kw)

         records = [self.record_for_book(book) for
         records = [self.record_for_book(book) for

-                   book in self.books(
-                       self.setspec_to_tag(
-                           kw.get('set', None)),
+                   book in self.books(None,

                            kw.get('from_', None),
                            kw.get('until', None))]
 
         return records, None
 
     def listMetadataFormats(self, **kw):
                            kw.get('from_', None),
                            kw.get('until', None))]
 
         return records, None
 
     def listMetadataFormats(self, **kw):

-        return [('oaidc',
+        formats = [('oai_dc',

                  'http://www.openarchives.org/OAI/2.0/oai_dc.xsd',
                  server.NS_OAIDC)]
                  'http://www.openarchives.org/OAI/2.0/oai_dc.xsd',
                  server.NS_OAIDC)]

+        if 'identifier' in kw:
+            slug = self.identifier_to_slug(kw['identifier'])
+            try:
+                b = Book.objects.get(slug=slug)
+                return formats
+            except:
+                try:
+                    d = Deleted.objects.get(slug=slug)
+                    return []
+                except:
+                    raise error.IdDoesNotExistError("This id does not exist")
+        else:
+            return formats

 
     def listSets(self, **kw):
 
     def listSets(self, **kw):

-        tags = []
-        for category in Catalogue.TAG_CATEGORIES:
-            for tag in Tag.objects.filter(category=category):
-                tags.append(("%s:%s" % (tag.category, tag.slug),
-                             tag.name,
-                             tag.description))
-        return tags, None
+        raise error.NoSetHierarchyError("Wolne Lektury does not support sets.")
+        # tags = []
+        # for category in Catalogue.TAG_CATEGORIES:
+        #     for tag in Tag.objects.filter(category=category):
+        #         tags.append(("%s:%s" % (tag.category, tag.slug),
+        #                      tag.name,
+        #                      tag.description))
+        # return tags, None