def test_measurements(self):
edition = self._edition()
measurement = MeasurementData(quantity_measured=Measurement.POPULARITY,
value=100)
metadata = Metadata(measurements=[measurement],
data_source=edition.data_source)
metadata.apply(edition)
[m] = edition.primary_identifier.measurements
eq_(Measurement.POPULARITY, m.quantity_measured)
eq_(100, m.value)
def extract_measurement(cls, rating_tag):
type = rating_tag.get('{http://schema.org/}additionalType')
value = rating_tag.get('{http://schema.org/}ratingValue')
if not value:
value = rating_tag.attrib.get('{http://schema.org}ratingValue')
if not type:
type = Measurement.RATING
try:
value = float(value)
return MeasurementData(
quantity_measured=type,
value=value,
)
except ValueError:
return None
def test_metadata_can_be_deepcopied(self):
# Check that we didn't put something in the metadata that
# will prevent it from being copied. (e.g., self.log)
subject = SubjectData(Subject.TAG, "subject")
contributor = ContributorData()
identifier = IdentifierData(Identifier.GUTENBERG_ID, "1")
link = LinkData(Hyperlink.OPEN_ACCESS_DOWNLOAD, "example.epub")
measurement = MeasurementData(Measurement.RATING, 5)
circulation = CirculationData(data_source=DataSource.GUTENBERG,
primary_identifier=identifier,
licenses_owned=0,
licenses_available=0,
licenses_reserved=0,
patrons_in_hold_queue=0)
primary_as_data = IdentifierData(type=identifier.type,
identifier=identifier.identifier)
other_data = IdentifierData(type=u"abc", identifier=u"def")
m = Metadata(
DataSource.GUTENBERG,
subjects=[subject],
contributors=[contributor],
primary_identifier=identifier,
links=[link],
measurements=[measurement],
circulation=circulation,
title="Hello Title",
subtitle="Subtle Hello",
sort_title="Sorting Howdy",
language="US English",
medium=Edition.BOOK_MEDIUM,
series="1",
series_position=1,
publisher="Hello World Publishing House",
imprint=u"Follywood",
issued=datetime.datetime.utcnow(),
published=datetime.datetime.utcnow(),
identifiers=[primary_as_data, other_data],
data_source_last_updated=datetime.datetime.utcnow(),
)
m_copy = deepcopy(m)
# If deepcopy didn't throw an exception we're ok.
assert m_copy is not None
def book_info_to_metadata(cls,
book,
include_bibliographic=True,
include_formats=True):
"""Turn Overdrive's JSON representation of a book into a Metadata
object.
Note: The json data passed into this method is from a different file/stream
from the json data that goes into the book_info_to_circulation() method.
"""
if not 'id' in book:
return None
overdrive_id = book['id']
primary_identifier = IdentifierData(Identifier.OVERDRIVE_ID,
overdrive_id)
if include_bibliographic:
title = book.get('title', None)
sort_title = book.get('sortTitle')
subtitle = book.get('subtitle', None)
series = book.get('series', None)
publisher = book.get('publisher', None)
imprint = book.get('imprint', None)
if 'publishDate' in book:
published = datetime.datetime.strptime(
book['publishDate'][:10], cls.DATE_FORMAT)
else:
published = None
languages = [l['code'] for l in book.get('languages', [])]
if 'eng' in languages or not languages:
language = 'eng'
else:
language = sorted(languages)[0]
contributors = []
for creator in book.get('creators', []):
sort_name = creator['fileAs']
display_name = creator['name']
role = creator['role']
roles = cls.parse_roles(overdrive_id,
role) or [Contributor.UNKNOWN_ROLE]
contributor = ContributorData(sort_name=sort_name,
display_name=display_name,
roles=roles,
biography=creator.get(
'bioText', None))
contributors.append(contributor)
subjects = []
for sub in book.get('subjects', []):
subject = SubjectData(type=Subject.OVERDRIVE,
identifier=sub['value'],
weight=100)
subjects.append(subject)
for sub in book.get('keywords', []):
subject = SubjectData(type=Subject.TAG,
identifier=sub['value'],
weight=1)
subjects.append(subject)
extra = dict()
if 'grade_levels' in book:
# n.b. Grade levels are measurements of reading level, not
# age appropriateness. We can use them as a measure of age
# appropriateness in a pinch, but we weight them less
# heavily than other information from Overdrive.
for i in book['grade_levels']:
subject = SubjectData(type=Subject.GRADE_LEVEL,
identifier=i['value'],
weight=10)
subjects.append(subject)
overdrive_medium = book.get('mediaType', None)
if overdrive_medium and overdrive_medium not in cls.overdrive_medium_to_simplified_medium:
cls.log.error("Could not process medium %s for %s",
overdrive_medium, overdrive_id)
medium = cls.overdrive_medium_to_simplified_medium.get(
overdrive_medium, Edition.BOOK_MEDIUM)
measurements = []
if 'awards' in book:
extra['awards'] = book.get('awards', [])
num_awards = len(extra['awards'])
measurements.append(
MeasurementData(Measurement.AWARDS, str(num_awards)))
for name, subject_type in (('ATOS', Subject.ATOS_SCORE),
('lexileScore', Subject.LEXILE_SCORE),
('interestLevel',
Subject.INTEREST_LEVEL)):
if not name in book:
continue
identifier = str(book[name])
subjects.append(
SubjectData(type=subject_type,
identifier=identifier,
weight=100))
for grade_level_info in book.get('gradeLevels', []):
grade_level = grade_level_info.get('value')
subjects.append(
SubjectData(type=Subject.GRADE_LEVEL,
identifier=grade_level,
weight=100))
identifiers = []
links = []
for format in book.get('formats', []):
for new_id in format.get('identifiers', []):
t = new_id['type']
v = new_id['value']
orig_v = v
type_key = None
if t == 'ASIN':
type_key = Identifier.ASIN
elif t == 'ISBN':
type_key = Identifier.ISBN
if len(v) == 10:
v = isbnlib.to_isbn13(v)
if v is None or not isbnlib.is_isbn13(v):
# Overdrive sometimes uses invalid values
# like "n/a" as placeholders. Ignore such
# values to avoid a situation where hundreds of
# books appear to have the same ISBN. ISBNs
# which fail check digit checks or are invalid
# also can occur. Log them for review.
cls.log.info("Bad ISBN value provided: %s", orig_v)
continue
elif t == 'DOI':
type_key = Identifier.DOI
elif t == 'UPC':
type_key = Identifier.UPC
elif t == 'PublisherCatalogNumber':
continue
if type_key and v:
identifiers.append(IdentifierData(type_key, v, 1))
# Samples become links.
if 'samples' in format:
if not format['id'] in cls.format_data_for_overdrive_format:
# Useless to us.
continue
content_type, drm_scheme = cls.format_data_for_overdrive_format.get(
format['id'])
if Representation.is_media_type(content_type):
for sample_info in format['samples']:
href = sample_info['url']
links.append(
LinkData(rel=Hyperlink.SAMPLE,
href=href,
media_type=content_type))
# A cover and its thumbnail become a single LinkData.
if 'images' in book:
images = book['images']
image_data = cls.image_link_to_linkdata(
images.get('cover'), Hyperlink.IMAGE)
for name in ['cover300Wide', 'cover150Wide', 'thumbnail']:
# Try to get a thumbnail that's as close as possible
# to the size we use.
image = images.get(name)
thumbnail_data = cls.image_link_to_linkdata(
image, Hyperlink.THUMBNAIL_IMAGE)
if not image_data:
image_data = cls.image_link_to_linkdata(
image, Hyperlink.IMAGE)
if thumbnail_data:
break
if image_data:
if thumbnail_data:
image_data.thumbnail = thumbnail_data
links.append(image_data)
# Descriptions become links.
short = book.get('shortDescription')
full = book.get('fullDescription')
if full:
links.append(
LinkData(
rel=Hyperlink.DESCRIPTION,
content=full,
media_type="text/html",
))
if short and (not full or not full.startswith(short)):
links.append(
LinkData(
rel=Hyperlink.SHORT_DESCRIPTION,
content=short,
media_type="text/html",
))
# Add measurements: rating and popularity
if book.get('starRating') is not None and book['starRating'] > 0:
measurements.append(
MeasurementData(quantity_measured=Measurement.RATING,
value=book['starRating']))
if book.get('popularity'):
measurements.append(
MeasurementData(quantity_measured=Measurement.POPULARITY,
value=book['popularity']))
metadata = Metadata(
data_source=DataSource.OVERDRIVE,
title=title,
subtitle=subtitle,
sort_title=sort_title,
language=language,
medium=medium,
series=series,
publisher=publisher,
imprint=imprint,
published=published,
primary_identifier=primary_identifier,
identifiers=identifiers,
subjects=subjects,
contributors=contributors,
measurements=measurements,
links=links,
)
else:
metadata = Metadata(
data_source=DataSource.OVERDRIVE,
primary_identifier=primary_identifier,
)
if include_formats:
formats = []
for format in book.get('formats', []):
format_id = format['id']
if format_id in cls.format_data_for_overdrive_format:
content_type, drm_scheme = cls.format_data_for_overdrive_format.get(
format_id)
formats.append(FormatData(content_type, drm_scheme))
elif format_id not in cls.ignorable_overdrive_formats:
cls.log.error(
"Could not process Overdrive format %s for %s",
format_id, overdrive_id)
# Also make a CirculationData so we can write the formats,
circulationdata = CirculationData(
data_source=DataSource.OVERDRIVE,
primary_identifier=primary_identifier,
formats=formats,
)
metadata.circulation = circulationdata
return metadata
class ItemListParser(XMLParser):
DATE_FORMAT = "%Y-%m-%d"
YEAR_FORMAT = "%Y"
NAMESPACES = {}
def parse(self, xml):
for i in self.process_all(xml, "//Item"):
yield i
parenthetical = re.compile(" \([^)]+\)$")
@classmethod
def contributors_from_string(cls, string):
contributors = []
if not string:
return contributors
for sort_name in string.split(';'):
sort_name = cls.parenthetical.sub("", sort_name.strip())
contributors.append(
ContributorData(
sort_name=sort_name.strip(),
roles=[Contributor.AUTHOR_ROLE]
)
)
return contributors
@classmethod
def parse_genre_string(self, s):
genres = []
if not s:
return genres
for i in s.split(","):
i = i.strip()
if not i:
continue
i = i.replace("&", "&").replace("&", "&").replace("'", "'")
genres.append(SubjectData(Subject.THREEM, i, weight=15))
return genres
def process_one(self, tag, namespaces):
"""Turn an <item> tag into a Metadata and an encompassed CirculationData
objects, and return the Metadata."""
def value(threem_key):
return self.text_of_optional_subtag(tag, threem_key)
links = dict()
identifiers = dict()
subjects = []
primary_identifier = IdentifierData(
Identifier.THREEM_ID, value("ItemId")
)
identifiers = []
for key in ('ISBN13', 'PhysicalISBN'):
v = value(key)
if v:
identifiers.append(
IdentifierData(Identifier.ISBN, v)
)
subjects = self.parse_genre_string(value("Genre"))
title = value("Title")
subtitle = value("SubTitle")
publisher = value("Publisher")
language = value("Language")
contributors = list(self.contributors_from_string(value('Authors')))
published_date = None
published = value("PubDate")
if published:
formats = [self.DATE_FORMAT, self.YEAR_FORMAT]
else:
published = value("PubYear")
formats = [self.YEAR_FORMAT]
for format in formats:
try:
published_date = datetime.strptime(published, format)
except ValueError, e:
pass
links = []
description = value("Description")
if description:
links.append(
LinkData(rel=Hyperlink.DESCRIPTION, content=description)
)
cover_url = value("CoverLinkURL").replace("&", "&")
links.append(LinkData(rel=Hyperlink.IMAGE, href=cover_url))
alternate_url = value("BookLinkURL").replace("&", "&")
links.append(LinkData(rel='alternate', href=alternate_url))
measurements = []
pages = value("NumberOfPages")
if pages:
pages = int(pages)
measurements.append(
MeasurementData(quantity_measured=Measurement.PAGE_COUNT,
value=pages)
)
medium = Edition.BOOK_MEDIUM
book_format = value("BookFormat")
format = None
if book_format == 'EPUB':
format = FormatData(
content_type=Representation.EPUB_MEDIA_TYPE,
drm_scheme=DeliveryMechanism.ADOBE_DRM
)
elif book_format == 'PDF':
format = FormatData(
content_type=Representation.PDF_MEDIA_TYPE,
drm_scheme=DeliveryMechanism.ADOBE_DRM
)
elif book_format == 'MP3':
format = FormatData(
content_type=Representation.MP3_MEDIA_TYPE,
drm_scheme=DeliveryMechanism.ADOBE_DRM
)
medium = Edition.AUDIO_MEDIUM
formats = [format]
metadata = Metadata(
data_source=DataSource.THREEM,
title=title,
subtitle=subtitle,
language=language,
medium=medium,
publisher=publisher,
published=published_date,
primary_identifier=primary_identifier,
identifiers=identifiers,
subjects=subjects,
contributors=contributors,
measurements=measurements,
links=links,
)
# Also make a CirculationData so we can write the formats,
circulationdata = CirculationData(
data_source=DataSource.THREEM,
primary_identifier=primary_identifier,
formats=formats,
links=links,
)
metadata.circulation = circulationdata
return metadata