def test_multiple_spaces(self):
field = db.SlugField()
assert field.slugify('a b') == 'a-b'
class SlugTester(db.Document):
title = db.StringField()
slug = db.SlugField(populate_from='title')
meta = {
'allow_inheritance': True,
}
class SlugUpdateTester(db.Document):
title = db.StringField()
slug = db.SlugField(populate_from='title', update=True)
class User(WithMetrics, UserMixin, db.Document):
slug = db.SlugField(max_length=255,
required=True,
populate_from='fullname')
email = db.StringField(max_length=255, required=True, unique=True)
password = db.StringField()
active = db.BooleanField()
roles = db.ListField(db.ReferenceField(Role), default=[])
first_name = db.StringField(max_length=255, required=True)
last_name = db.StringField(max_length=255, required=True)
avatar_url = db.URLField()
avatar = db.ImageField(fs=avatars,
basename=default_image_basename,
thumbnails=AVATAR_SIZES)
website = db.URLField()
about = db.StringField()
prefered_language = db.StringField()
apikey = db.StringField()
created_at = db.DateTimeField(default=datetime.now, required=True)
# The field below is required for Flask-security
# when SECURITY_CONFIRMABLE is True
confirmed_at = db.DateTimeField()
# The 5 fields below are required for Flask-security
# when SECURITY_TRACKABLE is True
last_login_at = db.DateTimeField()
current_login_at = db.DateTimeField()
last_login_ip = db.StringField()
current_login_ip = db.StringField()
login_count = db.IntField()
deleted = db.DateTimeField()
ext = db.MapField(db.GenericEmbeddedDocumentField())
extras = db.ExtrasField()
before_save = Signal()
after_save = Signal()
on_create = Signal()
on_update = Signal()
before_delete = Signal()
after_delete = Signal()
on_delete = Signal()
meta = {
'indexes': ['-created_at', 'slug', 'apikey'],
'ordering': ['-created_at']
}
__search_metrics__ = Object(
properties={
'datasets': Integer(),
'reuses': Integer(),
'followers': Integer(),
'views': Integer()
})
__metrics_keys__ = [
'datasets',
'reuses',
'following',
'followers',
]
def __str__(self):
return self.fullname
@property
def fullname(self):
return ' '.join((self.first_name or '', self.last_name or '')).strip()
@cached_property
def organizations(self):
from udata.core.organization.models import Organization
return Organization.objects(members__user=self, deleted__exists=False)
@property
def sysadmin(self):
return self.has_role('admin')
def url_for(self, *args, **kwargs):
return url_for('users.show', user=self, *args, **kwargs)
display_url = property(url_for)
@property
def external_url(self):
return self.url_for(_external=True)
@property
def visible(self):
count = self.metrics.get('datasets', 0) + self.metrics.get('reuses', 0)
return count > 0 and self.active
@cached_property
def resources_availability(self):
"""Return the percentage of availability for resources."""
# Flatten the list.
availabilities = list(
chain(*[org.check_availability() for org in self.organizations]))
# Filter out the unknown
availabilities = [a for a in availabilities if type(a) is bool]
if availabilities:
# Trick will work because it's a sum() of booleans.
return round(100. * sum(availabilities) / len(availabilities), 2)
# if nothing is unavailable, everything is considered OK
return 100
@cached_property
def datasets_org_count(self):
"""Return the number of datasets of user's organizations."""
from udata.models import Dataset # Circular imports.
return sum(
Dataset.objects(organization=org).visible().count()
for org in self.organizations)
@cached_property
def followers_org_count(self):
"""Return the number of followers of user's organizations."""
from udata.models import Follow # Circular imports.
return sum(
Follow.objects(following=org).count()
for org in self.organizations)
@property
def datasets_count(self):
"""Return the number of datasets of the user."""
return self.metrics.get('datasets', 0)
@property
def followers_count(self):
"""Return the number of followers of the user."""
return self.metrics.get('followers', 0)
def generate_api_key(self):
s = JSONWebSignatureSerializer(current_app.config['SECRET_KEY'])
byte_str = s.dumps({
'user': str(self.id),
'time': time(),
})
self.apikey = byte_str.decode()
def clear_api_key(self):
self.apikey = None
@classmethod
def get(cls, id_or_slug):
obj = cls.objects(slug=id_or_slug).first()
return obj or cls.objects.get_or_404(id=id_or_slug)
@classmethod
def pre_save(cls, sender, document, **kwargs):
cls.before_save.send(document)
@classmethod
def post_save(cls, sender, document, **kwargs):
cls.after_save.send(document)
if kwargs.get('created'):
cls.on_create.send(document)
else:
cls.on_update.send(document)
@cached_property
def json_ld(self):
result = {
'@type': 'Person',
'@context': 'http://schema.org',
'name': self.fullname,
}
if self.about:
result['description'] = mdstrip(self.about)
if self.avatar_url:
result['image'] = self.avatar_url
if self.website:
result['url'] = self.website
return result
def mark_as_deleted(self):
copied_user = copy(self)
self.email = '{}@deleted'.format(self.id)
self.slug = 'deleted'
self.password = None
self.active = False
self.first_name = 'DELETED'
self.last_name = 'DELETED'
self.avatar = None
self.avatar_url = None
self.website = None
self.about = None
self.extras = None
self.apikey = None
self.deleted = datetime.now()
self.save()
for organization in self.organizations:
organization.members = [
member for member in organization.members
if member.user != self
]
organization.save()
for discussion in Discussion.objects(discussion__posted_by=self):
for message in discussion.discussion:
if message.posted_by == self:
message.content = 'DELETED'
discussion.save()
Follow.objects(follower=self).delete()
Follow.objects(following=self).delete()
mail.send(_('Account deletion'), copied_user, 'account_deleted')
def count_datasets(self):
from udata.models import Dataset
self.metrics['datasets'] = Dataset.objects(
owner=self).visible().count()
self.save()
def count_reuses(self):
from udata.models import Reuse
self.metrics['reuses'] = Reuse.objects(owner=self).visible().count()
self.save()
def count_followers(self):
from udata.models import Follow
self.metrics['followers'] = Follow.objects(
until=None).followers(self).count()
self.save()
def count_following(self):
from udata.models import Follow
self.metrics['following'] = Follow.objects.following(self).count()
self.save()
class Organization(WithMetrics, BadgeMixin, db.Datetimed, db.Document):
name = db.StringField(max_length=255, required=True)
acronym = db.StringField(max_length=128)
slug = db.SlugField(max_length=255,
required=True,
populate_from='name',
update=True)
description = db.StringField(required=True)
url = db.StringField()
image_url = db.StringField()
logo = db.ImageField(fs=avatars,
basename=default_image_basename,
thumbnails=LOGO_SIZES)
members = db.ListField(db.EmbeddedDocumentField(Member))
teams = db.ListField(db.EmbeddedDocumentField(Team))
requests = db.ListField(db.EmbeddedDocumentField(MembershipRequest))
badges = db.ListField(db.EmbeddedDocumentField(OrganizationBadge))
ext = db.MapField(db.GenericEmbeddedDocumentField())
extras = db.ExtrasField()
deleted = db.DateTimeField()
meta = {
'allow_inheritance': True,
'indexes': ['-created_at', 'slug'],
'ordering': ['-created_at'],
'queryset_class': OrganizationQuerySet,
}
def __str__(self):
return self.name or ''
__unicode__ = __str__
before_save = Signal()
after_save = Signal()
on_create = Signal()
on_update = Signal()
before_delete = Signal()
after_delete = Signal()
@classmethod
def pre_save(cls, sender, document, **kwargs):
cls.before_save.send(document)
@classmethod
def post_save(cls, sender, document, **kwargs):
cls.after_save.send(document)
if kwargs.get('created'):
cls.on_create.send(document)
else:
cls.on_update.send(document)
@property
def display_url(self):
return url_for('organizations.show', org=self)
@property
def external_url(self):
return url_for('organizations.show', org=self, _external=True)
@property
def pending_requests(self):
return [r for r in self.requests if r.status == 'pending']
@property
def refused_requests(self):
return [r for r in self.requests if r.status == 'refused']
@property
def accepted_requests(self):
return [r for r in self.requests if r.status == 'accepted']
@property
def public_service(self):
badges_kind = [badge.kind for badge in self.badges]
return PUBLIC_SERVICE in badges_kind and CERTIFIED in badges_kind
def member(self, user):
for member in self.members:
if member.user == user:
return member
return None
def is_member(self, user):
return self.member(user) is not None
def is_admin(self, user):
member = self.member(user)
return member is not None and member.role == 'admin'
def pending_request(self, user):
for request in self.requests:
if request.user == user and request.status == 'pending':
return request
return None
@classmethod
def get(cls, id_or_slug):
obj = cls.objects(slug=id_or_slug).first()
return obj or cls.objects.get_or_404(id=id_or_slug)
def by_role(self, role):
return filter(lambda m: m.role == role, self.members)
class Dataset(WithMetrics, BadgeMixin, db.Owned, db.Document):
created_at = DateTimeField(verbose_name=_('Creation date'),
default=datetime.now,
required=True)
last_modified = DateTimeField(verbose_name=_('Last modification date'),
default=datetime.now,
required=True)
title = db.StringField(required=True)
acronym = db.StringField(max_length=128)
# /!\ do not set directly the slug when creating or updating a dataset
# this will break the search indexation
slug = db.SlugField(max_length=255,
required=True,
populate_from='title',
update=True,
follow=True)
description = db.StringField(required=True, default='')
license = db.ReferenceField('License')
tags = db.TagListField()
resources = db.ListField(db.EmbeddedDocumentField(Resource))
private = db.BooleanField(default=False)
frequency = db.StringField(choices=list(UPDATE_FREQUENCIES.keys()))
frequency_date = db.DateTimeField(verbose_name=_('Future date of update'))
temporal_coverage = db.EmbeddedDocumentField(db.DateRange)
spatial = db.EmbeddedDocumentField(SpatialCoverage)
ext = db.MapField(db.GenericEmbeddedDocumentField())
extras = db.ExtrasField()
featured = db.BooleanField(required=True, default=False)
deleted = db.DateTimeField()
archived = db.DateTimeField()
def __str__(self):
return self.title or ''
__badges__ = {
PIVOTAL_DATA: _('Pivotal data'),
}
__search_metrics__ = Object(properties={
'reuses': Integer(),
'followers': Integer(),
'views': Integer(),
})
__metrics_keys__ = [
'discussions',
'issues',
'reuses',
'followers',
'views',
]
meta = {
'indexes': [
'-created_at',
'slug',
'resources.id',
'resources.urlhash',
] + db.Owned.meta['indexes'],
'ordering': ['-created_at'],
'queryset_class':
DatasetQuerySet,
}
before_save = signal('Dataset.before_save')
after_save = signal('Dataset.after_save')
on_create = signal('Dataset.on_create')
on_update = signal('Dataset.on_update')
before_delete = signal('Dataset.before_delete')
after_delete = signal('Dataset.after_delete')
on_delete = signal('Dataset.on_delete')
on_archive = signal('Dataset.on_archive')
on_resource_added = signal('Dataset.on_resource_added')
verbose_name = _('dataset')
@classmethod
def pre_save(cls, sender, document, **kwargs):
cls.before_save.send(document)
@classmethod
def post_save(cls, sender, document, **kwargs):
if 'post_save' in kwargs.get('ignores', []):
return
cls.after_save.send(document)
if kwargs.get('created'):
cls.on_create.send(document)
else:
cls.on_update.send(document)
if document.deleted:
cls.on_delete.send(document)
if document.archived:
cls.on_archive.send(document)
if kwargs.get('resource_added'):
cls.on_resource_added.send(document,
resource_id=kwargs['resource_added'])
def clean(self):
if self.frequency in LEGACY_FREQUENCIES:
self.frequency = LEGACY_FREQUENCIES[self.frequency]
def url_for(self, *args, **kwargs):
return url_for('datasets.show', dataset=self, *args, **kwargs)
display_url = property(url_for)
@property
def is_visible(self):
return not self.is_hidden
@property
def is_hidden(self):
return (len(self.resources) == 0 or self.private or self.deleted
or self.archived)
@property
def full_title(self):
if not self.acronym:
return self.title
return '{title} ({acronym})'.format(**self._data)
@property
def external_url(self):
return self.url_for(_external=True)
@property
def image_url(self):
if self.organization:
return self.organization.logo.url
elif self.owner:
return self.owner.avatar.url
@property
def frequency_label(self):
return UPDATE_FREQUENCIES.get(self.frequency or 'unknown',
UPDATE_FREQUENCIES['unknown'])
def check_availability(self):
"""Check if resources from that dataset are available.
Return a list of (boolean or 'unknown')
"""
# Only check remote resources.
remote_resources = [
resource for resource in self.resources
if resource.filetype == 'remote'
]
if not remote_resources:
return []
return [resource.check_availability() for resource in remote_resources]
@property
def last_update(self):
if self.resources:
return max(resource.published for resource in self.resources)
else:
return self.last_modified
@property
def next_update(self):
"""Compute the next expected update date,
given the frequency and last_update.
Return None if the frequency is not handled.
"""
delta = None
if self.frequency == 'daily':
delta = timedelta(days=1)
elif self.frequency == 'weekly':
delta = timedelta(weeks=1)
elif self.frequency == 'fortnighly':
delta = timedelta(weeks=2)
elif self.frequency == 'monthly':
delta = timedelta(weeks=4)
elif self.frequency == 'bimonthly':
delta = timedelta(weeks=4 * 2)
elif self.frequency == 'quarterly':
delta = timedelta(weeks=52 / 4)
elif self.frequency == 'biannual':
delta = timedelta(weeks=52 / 2)
elif self.frequency == 'annual':
delta = timedelta(weeks=52)
elif self.frequency == 'biennial':
delta = timedelta(weeks=52 * 2)
elif self.frequency == 'triennial':
delta = timedelta(weeks=52 * 3)
elif self.frequency == 'quinquennial':
delta = timedelta(weeks=52 * 5)
if delta is None:
return
else:
return self.last_update + delta
@cached_property
def quality(self):
"""Return a dict filled with metrics related to the inner
quality of the dataset:
* number of tags
* description length
* and so on
"""
from udata.models import Discussion # noqa: Prevent circular imports
result = {}
if not self.id:
# Quality is only relevant on saved Datasets
return result
if self.frequency != 'unknown':
result['frequency'] = self.frequency
if self.next_update:
result['update_in'] = -(self.next_update - datetime.now()).days
if self.tags:
result['tags_count'] = len(self.tags)
if self.description:
result['description_length'] = len(self.description)
if self.resources:
result['has_resources'] = True
result['has_only_closed_or_no_formats'] = all(
resource.closed_or_no_format for resource in self.resources)
result['has_unavailable_resources'] = not all(
self.check_availability())
discussions = Discussion.objects(subject=self)
if discussions:
result['discussions'] = len(discussions)
result['has_untreated_discussions'] = not all(
discussion.person_involved(self.owner)
for discussion in discussions)
result['score'] = self.compute_quality_score(result)
return result
def compute_quality_score(self, quality):
"""Compute the score related to the quality of that dataset."""
score = 0
UNIT = 2
if 'update_in' in quality:
# TODO: should be related to frequency.
if quality['update_in'] < 0:
score += UNIT
else:
score -= UNIT
if 'tags_count' in quality:
if quality['tags_count'] > 3:
score += UNIT
if 'description_length' in quality:
if quality['description_length'] > 100:
score += UNIT
if 'has_resources' in quality:
if quality['has_only_closed_or_no_formats']:
score -= UNIT
else:
score += UNIT
if quality['has_unavailable_resources']:
score -= UNIT
else:
score += UNIT
if 'discussions' in quality:
if quality['has_untreated_discussions']:
score -= UNIT
else:
score += UNIT
if score < 0:
return 0
return score
@classmethod
def get(cls, id_or_slug):
obj = cls.objects(slug=id_or_slug).first()
return obj or cls.objects.get_or_404(id=id_or_slug)
def add_resource(self, resource):
'''Perform an atomic prepend for a new resource'''
resource.validate()
self.update(
__raw__={
'$push': {
'resources': {
'$each': [resource.to_mongo()],
'$position': 0
}
}
})
self.reload()
post_save.send(self.__class__,
document=self,
resource_added=resource.id)
def update_resource(self, resource):
'''Perform an atomic update for an existing resource'''
index = self.resources.index(resource)
data = {'resources__{index}'.format(index=index): resource}
self.update(**data)
self.reload()
post_save.send(self.__class__, document=self)
@property
def community_resources(self):
return self.id and CommunityResource.objects.filter(dataset=self) or []
@cached_property
def json_ld(self):
result = {
'@context':
'http://schema.org',
'@type':
'Dataset',
'@id':
str(self.id),
'alternateName':
self.slug,
'dateCreated':
self.created_at.isoformat(),
'dateModified':
self.last_modified.isoformat(),
'url':
url_for('datasets.show', dataset=self, _external=True),
'name':
self.title,
'keywords':
','.join(self.tags),
'distribution': [resource.json_ld for resource in self.resources],
# Theses values are not standard
'contributedDistribution':
[resource.json_ld for resource in self.community_resources],
'extras':
[get_json_ld_extra(*item) for item in self.extras.items()],
}
if self.description:
result['description'] = mdstrip(self.description)
if self.license and self.license.url:
result['license'] = self.license.url
if self.organization:
author = self.organization.json_ld
elif self.owner:
author = self.owner.json_ld
else:
author = None
if author:
result['author'] = author
return result
@property
def views_count(self):
return self.metrics.get('views', 0)
def count_discussions(self):
from udata.models import Discussion
self.metrics['discussions'] = Discussion.objects(subject=self,
closed=None).count()
self.save()
def count_issues(self):
from udata.models import Issue
self.metrics['issues'] = Issue.objects(subject=self,
closed=None).count()
self.save()
def count_reuses(self):
from udata.models import Reuse
self.metrics['reuses'] = Reuse.objects(datasets=self).visible().count()
self.save()
def count_followers(self):
from udata.models import Follow
self.metrics['followers'] = Follow.objects(
until=None).followers(self).count()
self.save()
class Team(db.EmbeddedDocument):
name = db.StringField(required=True)
slug = db.SlugField(max_length=255, required=True, populate_from='name', update=True, unique=False)
description = db.StringField()
members = db.ListField(db.ReferenceField('User'))
class Dataset(WithMetrics, BadgeMixin, db.Document):
created_at = DateTimeField(verbose_name=_('Creation date'),
default=datetime.now,
required=True)
last_modified = DateTimeField(verbose_name=_('Last modification date'),
default=datetime.now,
required=True)
title = db.StringField(max_length=255, required=True)
slug = db.SlugField(max_length=255,
required=True,
populate_from='title',
update=True)
description = db.StringField(required=True, default='')
license = db.ReferenceField('License')
tags = db.TagListField()
resources = db.ListField(db.EmbeddedDocumentField(Resource))
private = db.BooleanField()
owner = db.ReferenceField('User', reverse_delete_rule=db.NULLIFY)
organization = db.ReferenceField('Organization',
reverse_delete_rule=db.NULLIFY)
frequency = db.StringField(choices=UPDATE_FREQUENCIES.keys())
frequency_date = db.DateTimeField(verbose_name=_('Future date of update'))
temporal_coverage = db.EmbeddedDocumentField(db.DateRange)
spatial = db.EmbeddedDocumentField(SpatialCoverage)
ext = db.MapField(db.GenericEmbeddedDocumentField())
extras = db.ExtrasField()
featured = db.BooleanField(required=True, default=False)
deleted = db.DateTimeField()
def __str__(self):
return self.title or ''
__unicode__ = __str__
__badges__ = {
PIVOTAL_DATA: _('Pivotal data'),
}
meta = {
'allow_inheritance':
True,
'indexes': [
'-created_at',
'slug',
'organization',
'resources.id',
'resources.urlhash',
],
'ordering': ['-created_at'],
'queryset_class':
DatasetQuerySet,
}
before_save = signal('Dataset.before_save')
after_save = signal('Dataset.after_save')
on_create = signal('Dataset.on_create')
on_update = signal('Dataset.on_update')
before_delete = signal('Dataset.before_delete')
after_delete = signal('Dataset.after_delete')
on_delete = signal('Dataset.on_delete')
verbose_name = _('dataset')
@classmethod
def pre_save(cls, sender, document, **kwargs):
cls.before_save.send(document)
@classmethod
def post_save(cls, sender, document, **kwargs):
cls.after_save.send(document)
if kwargs.get('created'):
cls.on_create.send(document)
else:
cls.on_update.send(document)
def url_for(self, *args, **kwargs):
return url_for('datasets.show', dataset=self, *args, **kwargs)
display_url = property(url_for)
@property
def external_url(self):
return self.url_for(_external=True)
@property
def image_url(self):
if self.organization:
return self.organization.logo.url
elif self.owner:
return self.owner.avatar.url
@property
def frequency_label(self):
return UPDATE_FREQUENCIES.get(self.frequency or 'unknown',
UPDATE_FREQUENCIES['unknown'])
def check_availability(self):
"""Check if resources from that dataset are available.
Return a list of booleans.
"""
# Only check remote resources.
remote_resources = [
resource for resource in self.resources
if resource.filetype == 'remote'
]
if not remote_resources:
return []
# First, we try to retrieve all data from the group (slug).
error, response = check_url_from_group(self.slug)
if error:
# The group is unknown, the check will be performed by resource.
return [
resource.check_availability(self.slug)
for resource in remote_resources
]
else:
return [
int(url_infos['status']) == 200
for url_infos in response['urls']
]
@property
def last_update(self):
if self.resources:
return max(resource.published for resource in self.resources)
else:
return self.last_modified
@property
def next_update(self):
"""Compute the next expected update date,
given the frequency and last_update.
Return None if the frequency is not handled.
"""
delta = None
if self.frequency == 'daily':
delta = timedelta(days=1)
elif self.frequency == 'weekly':
delta = timedelta(weeks=1)
elif self.frequency == 'fortnighly':
delta = timedelta(weeks=2)
elif self.frequency == 'monthly':
delta = timedelta(weeks=4)
elif self.frequency == 'bimonthly':
delta = timedelta(weeks=4 * 2)
elif self.frequency == 'quarterly':
delta = timedelta(weeks=52 / 4)
elif self.frequency == 'biannual':
delta = timedelta(weeks=52 / 2)
elif self.frequency == 'annual':
delta = timedelta(weeks=52)
elif self.frequency == 'biennial':
delta = timedelta(weeks=52 * 2)
elif self.frequency == 'triennial':
delta = timedelta(weeks=52 * 3)
elif self.frequency == 'quinquennial':
delta = timedelta(weeks=52 * 5)
if delta is None:
return
else:
return self.last_update + delta
@cached_property
def quality(self):
"""Return a dict filled with metrics related to the inner
quality of the dataset:
* number of tags
* description length
* and so on
"""
result = {}
if self.next_update:
result['frequency'] = self.frequency
result['update_in'] = -(self.next_update - datetime.now()).days
if self.tags:
result['tags_count'] = len(self.tags)
if self.description:
result['description_length'] = len(self.description)
if self.resources:
result['has_resources'] = True
result['has_only_closed_formats'] = all(
resource.closed_format for resource in self.resources)
result['has_unavailable_resources'] = not all(
self.check_availability())
discussions = DatasetDiscussion.objects(subject=self.id)
if discussions:
result['discussions'] = len(discussions)
result['has_untreated_discussions'] = not all(
discussion.person_involved(self.owner)
for discussion in discussions)
result['score'] = self.compute_quality_score(result)
return result
def compute_quality_score(self, quality):
"""Compute the score related to the quality of that dataset."""
score = 0
UNIT = 2
if 'frequency' in quality:
# TODO: should be related to frequency.
if quality['update_in'] < 0:
score += UNIT
else:
score -= UNIT
if 'tags_count' in quality:
if quality['tags_count'] > 3:
score += UNIT
if 'description_length' in quality:
if quality['description_length'] > 100:
score += UNIT
if 'has_resources' in quality:
if quality['has_only_closed_formats']:
score -= UNIT
else:
score += UNIT
if quality['has_unavailable_resources']:
score -= UNIT
else:
score += UNIT
if 'discussions' in quality:
if quality['has_untreated_discussions']:
score -= UNIT
else:
score += UNIT
if score < 0:
return 0
return score
@classmethod
def get(cls, id_or_slug):
obj = cls.objects(slug=id_or_slug).first()
return obj or cls.objects.get_or_404(id=id_or_slug)
def add_resource(self, resource):
'''Perform an atomic prepend for a new resource'''
resource.validate()
self.update(
__raw__={
'$push': {
'resources': {
'$each': [resource.to_mongo()],
'$position': 0
}
}
})
self.reload()
post_save.send(self.__class__, document=self)
def update_resource(self, resource):
'''Perform an atomic update for an existing resource'''
index = self.resources.index(resource)
data = {'resources__{index}'.format(index=index): resource}
self.update(**data)
self.reload()
post_save.send(self.__class__, document=self)
@property
def community_resources(self):
return self.id and CommunityResource.objects.filter(dataset=self) or []
class License(db.Document):
# We need to declare id explicitly since we do not use the default
# value set by Mongo.
id = db.StringField(primary_key=True)
created_at = db.DateTimeField(default=datetime.now, required=True)
title = db.StringField(required=True)
alternate_titles = db.ListField(db.StringField())
slug = db.SlugField(required=True, populate_from='title')
url = db.URLField()
alternate_urls = db.ListField(db.URLField())
maintainer = db.StringField()
flags = db.ListField(db.StringField())
active = db.BooleanField()
def __str__(self):
return self.title
@classmethod
def guess(cls, *strings, **kwargs):
'''
Try to guess a license from a list of strings.
Accept a `default` keyword argument which will be
the default fallback license.
'''
license = None
for string in strings:
license = cls.guess_one(string)
if license:
break
return license or kwargs.get('default')
@classmethod
def guess_one(cls, text):
'''
Try to guess license from a string.
Try to exact match on identifier then slugified title
and fallback on edit distance ranking (after slugification)
'''
if not text:
return
qs = cls.objects
text = text.strip().lower() # Stored identifiers are lower case
slug = cls.slug.slugify(text) # Use slug as it normalize string
license = qs(
db.Q(id__iexact=text) | db.Q(slug=slug) | db.Q(url__iexact=text)
| db.Q(alternate_urls__iexact=text)).first()
if license is None:
# Try to single match with a low Damerau-Levenshtein distance
computed = ((l, rdlevenshtein(l.slug, slug)) for l in cls.objects)
candidates = [l for l, d in computed if d <= MAX_DISTANCE]
# If there is more that one match, we cannot determinate
# which one is closer to safely choose between candidates
if len(candidates) == 1:
license = candidates[0]
if license is None:
# Try to single match with a low Damerau-Levenshtein distance
computed = ((l, rdlevenshtein(cls.slug.slugify(t), slug))
for l in cls.objects for t in l.alternate_titles)
candidates = [l for l, d in computed if d <= MAX_DISTANCE]
# If there is more that one match, we cannot determinate
# which one is closer to safely choose between candidates
if len(candidates) == 1:
license = candidates[0]
return license
@classmethod
def default(cls):
return cls.objects(id=DEFAULT_LICENSE['id']).first()
class Reuse(db.Datetimed, WithMetrics, BadgeMixin, db.Owned, db.Document):
title = db.StringField(required=True)
slug = db.SlugField(
max_length=255, required=True, populate_from='title', update=True)
description = db.StringField(required=True)
type = db.StringField(required=True, choices=REUSE_TYPES.keys())
url = db.StringField(required=True)
urlhash = db.StringField(required=True, unique=True)
image_url = db.StringField()
image = db.ImageField(
fs=images, basename=default_image_basename, max_size=IMAGE_MAX_SIZE,
thumbnails=IMAGE_SIZES)
datasets = db.ListField(
db.ReferenceField('Dataset', reverse_delete_rule=db.PULL))
tags = db.TagListField()
# badges = db.ListField(db.EmbeddedDocumentField(ReuseBadge))
private = db.BooleanField()
ext = db.MapField(db.GenericEmbeddedDocumentField())
extras = db.ExtrasField()
featured = db.BooleanField()
deleted = db.DateTimeField()
def __unicode__(self):
return self.title or ''
__badges__ = {}
meta = {
'indexes': ['-created_at', 'urlhash'] + db.Owned.meta['indexes'],
'ordering': ['-created_at'],
'queryset_class': ReuseQuerySet,
}
before_save = Signal()
after_save = Signal()
on_create = Signal()
on_update = Signal()
before_delete = Signal()
after_delete = Signal()
on_delete = Signal()
verbose_name = _('reuse')
@classmethod
def pre_save(cls, sender, document, **kwargs):
# Emit before_save
cls.before_save.send(document)
@classmethod
def post_save(cls, sender, document, **kwargs):
cls.after_save.send(document)
if kwargs.get('created'):
cls.on_create.send(document)
else:
cls.on_update.send(document)
if document.deleted:
cls.on_delete.send(document)
def url_for(self, *args, **kwargs):
return url_for('reuses.show', reuse=self, *args, **kwargs)
display_url = property(url_for)
@property
def external_url(self):
return self.url_for(_external=True)
@property
def type_label(self):
return REUSE_TYPES[self.type]
def clean(self):
'''Auto populate urlhash from url'''
if not self.urlhash or 'url' in self._get_changed_fields():
self.urlhash = hash_url(self.url)
super(Reuse, self).clean()
@classmethod
def get(cls, id_or_slug):
obj = cls.objects(slug=id_or_slug).first()
return obj or cls.objects.get_or_404(id=id_or_slug)
@classmethod
def url_exists(cls, url):
urlhash = hash_url(url)
return cls.objects(urlhash=urlhash).count() > 0
@cached_property
def json_ld(self):
result = {
'@context': 'http://schema.org',
'@type': 'CreativeWork',
'alternateName': self.slug,
'dateCreated': self.created_at.isoformat(),
'dateModified': self.last_modified.isoformat(),
'url': url_for('reuses.show', reuse=self, _external=True),
'name': self.title,
'isBasedOnUrl': self.url,
}
if self.description:
result['description'] = mdstrip(self.description)
if self.organization:
author = self.organization.json_ld
elif self.owner:
author = self.owner.json_ld
else:
author = None
if author:
result['author'] = author
return result
class Organization(WithMetrics, BadgeMixin, db.Datetimed, db.Document):
name = db.StringField(required=True)
acronym = db.StringField(max_length=128)
slug = db.SlugField(max_length=255,
required=True,
populate_from='name',
update=True)
description = db.StringField(required=True)
url = db.StringField()
image_url = db.StringField()
logo = db.ImageField(fs=avatars,
basename=default_image_basename,
max_size=LOGO_MAX_SIZE,
thumbnails=LOGO_SIZES)
members = db.ListField(db.EmbeddedDocumentField(Member))
teams = db.ListField(db.EmbeddedDocumentField(Team))
requests = db.ListField(db.EmbeddedDocumentField(MembershipRequest))
ext = db.MapField(db.GenericEmbeddedDocumentField())
zone = db.StringField()
extras = db.ExtrasField()
deleted = db.DateTimeField()
meta = {
'indexes': ['-created_at', 'slug'],
'ordering': ['-created_at'],
'queryset_class': OrganizationQuerySet,
}
def __unicode__(self):
return self.name or ''
__badges__ = {
PUBLIC_SERVICE: _('Public Service'),
CERTIFIED: _('Certified'),
}
before_save = Signal()
after_save = Signal()
on_create = Signal()
on_update = Signal()
before_delete = Signal()
after_delete = Signal()
@classmethod
def pre_save(cls, sender, document, **kwargs):
cls.before_save.send(document)
@classmethod
def post_save(cls, sender, document, **kwargs):
cls.after_save.send(document)
if kwargs.get('created'):
cls.on_create.send(document)
else:
cls.on_update.send(document)
def url_for(self, *args, **kwargs):
return url_for('organizations.show', org=self, *args, **kwargs)
display_url = property(url_for)
@property
def external_url(self):
return self.url_for(_external=True)
@property
def pending_requests(self):
return [r for r in self.requests if r.status == 'pending']
@property
def refused_requests(self):
return [r for r in self.requests if r.status == 'refused']
@property
def accepted_requests(self):
return [r for r in self.requests if r.status == 'accepted']
@property
def certified(self):
return any(b.kind == CERTIFIED for b in self.badges)
@property
def public_service(self):
is_public_service = any(b.kind == PUBLIC_SERVICE for b in self.badges)
return self.certified and is_public_service
def member(self, user):
for member in self.members:
if member.user == user:
return member
return None
def is_member(self, user):
return self.member(user) is not None
def is_admin(self, user):
member = self.member(user)
return member is not None and member.role == 'admin'
def pending_request(self, user):
for request in self.requests:
if request.user == user and request.status == 'pending':
return request
return None
@classmethod
def get(cls, id_or_slug):
obj = cls.objects(slug=id_or_slug).first()
return obj or cls.objects.get_or_404(id=id_or_slug)
def by_role(self, role):
return filter(lambda m: m.role == role, self.members)
def check_availability(self):
from udata.models import Dataset # Circular imports.
# Performances: only check the first 20 datasets for now.
return chain(*[
dataset.check_availability()
for dataset in Dataset.objects(organization=self).visible()[:20]
])
@cached_property
def json_ld(self):
type_ = 'GovernmentOrganization' if self.public_service \
else 'Organization'
result = {
'@context': 'http://schema.org',
'@type': type_,
'alternateName': self.slug,
'url': url_for('organizations.show', org=self, _external=True),
'name': self.name,
}
if self.description:
result['description'] = mdstrip(self.description)
logo = self.logo(external=True)
if logo:
result['logo'] = logo
return result
def test_is_stripped(self):
field = db.SlugField()
self.assertEqual(field.slugify(' ab '), 'ab')
def test_custom_separator(self):
field = db.SlugField(separator='+')
self.assertEqual(field.slugify('a b'), 'a+b')
def test_lower_case_false(self):
field = db.SlugField(lower_case=False)
self.assertEqual(field.slugify('AbC'), 'AbC')
def test_lower_case_default(self):
field = db.SlugField()
self.assertEqual(field.slugify('ABC'), 'abc')
def test_multiple_spaces(self):
field = db.SlugField()
self.assertEqual(field.slugify('a b'), 'a-b')
