class ResourceApiAttrs(ObjectAttrs, HighlightObjectMixin):
title = TranslatedStr()
description = TranslatedStr()
category = fields.Str()
format = fields.Str()
media_type = fields.Str(
attribute='type'
) # https://jsonapi.org/format/#document-resource-object-fields
visualization_types = ListWithoutNoneStrElement(fields.Str())
downloads_count =\
fields.Function(
lambda obj: obj.computed_downloads_count if is_enabled('S16_new_date_counters.be') else obj.downloads_count)
openness_score = fields.Integer()
views_count =\
fields.Function(
lambda obj: obj.computed_views_count if is_enabled('S16_new_date_counters.be') else obj.views_count)
modified = fields.DateTime()
created = fields.DateTime()
verified = fields.DateTime()
data_date = fields.Date()
file_url = fields.Str()
file_size = fields.Integer()
csv_file_url = fields.Str()
csv_file_size = fields.Integer()
jsonld_file_url = fields.Str()
jsonld_file_size = fields.Integer()
jsonld_download_url = fields.Str()
download_url = fields.Str()
csv_download_url = fields.Str()
link = fields.Str()
data_special_signs = fields.Nested(SpecialSignSchema,
data_key='special_signs',
many=True)
is_chart_creation_blocked = fields.Bool()
if is_enabled('S35_high_value_data.be'):
has_high_value_data = fields.Boolean()
if is_enabled('S37_resources_admin_region_data.be'):
regions = fields.Method('get_regions')
if is_enabled('S40_new_file_model.be'):
files = fields.Method('get_files')
class Meta:
relationships_schema = ResourceApiRelationships
object_type = 'resource'
api_path = 'resources'
url_template = '{api_url}/resources/{ident}'
model = 'resources.Resource'
def get_regions(self, res):
return RegionSchema(many=True).dump(
getattr(res, 'all_regions', res.regions))
def get_files(self, res):
return ResourceFileSchema(many=True).dump(
getattr(res, 'all_files', res.files))
class ResourceCSVMetadataSerializer(schemas.ExtSchema):
frontend_absolute_url = fields.Url(data_key=_('Resource URL'))
title = TranslatedStr(data_key=_('Resource title'), default='')
description = TranslatedStr(data_key=_('Resource description'))
created = fields.DateTime(data_key=_('Resource created'), format='iso8601')
data_date = fields.Date(data_key=_('Data date'))
openness_score = fields.Int(data_key=_('Openness score'))
resource_type = fields.Function(lambda obj: obj.get_type_display(),
data_key=_('Type'))
format = fields.Str(data_key=_('File format'), default='')
file_size = fields.Function(lambda obj: sizeof_fmt(obj.file_size)
if obj.file_size else '',
data_key=_('File size'))
views_count = fields.Int(attribute='computed_views_count',
data_key=_("Resource views count"))
downloads_count = fields.Int(attribute='computed_downloads_count',
data_key=_("Resource downloads count"))
has_table = fields.Function(lambda obj: _('YES')
if obj.has_table else _('NO'),
data_key=_('Table'))
has_chart = fields.Function(lambda obj: _('YES')
if obj.has_chart else _('NO'),
data_key=_('Map'))
has_map = fields.Function(lambda obj: _('YES') if obj.has_map else _('NO'),
data_key=_('Chart'))
download_url = fields.Url(data_key=_('Download URL'))
data_special_signs = fields.Nested(SpecialSignSchema,
data_key=_('special signs'),
many=True)
@ma.post_dump(pass_many=False)
def prepare_nested_data(self, data, **kwargs):
special_signs = data.get(_('special signs'))
signs_str = '\n'.join([
'{name_label}: {name}, {symbol_label}: "{symbol}", {desc_label}: {desc}'
.format(name=sign['name'],
name_label=_('name'),
symbol=sign['symbol'],
symbol_label=_('symbol'),
desc=sign['description'],
desc_label=_('description')) for sign in special_signs
])
data[_('special signs')] = signs_str
values_with_html = [_('Resource title'), _('Resource description')]
for attribute in values_with_html:
data[attribute] = strip_tags(data[attribute])
return data
class Meta:
ordered = True
class InstitutionXMLSerializer(ExtSchema):
id = fields.Integer()
url = fields.Url(attribute='frontend_absolute_url')
type = fields.Function(
lambda organization: organization.get_institution_type_display())
title = TranslatedStr()
abbreviation = fields.Str()
epuap = fields.Str()
website = fields.Url()
created = fields.DateTime(format='iso8601')
modified = fields.DateTime(format='iso8601')
postal_code = fields.Str()
city = fields.Str()
street = fields.Str()
street_number = fields.Str()
street_type = fields.Str()
flat_number = fields.Str()
email = fields.Str()
tel = fields.Str(data_key='phone_number')
regon = fields.Str()
published_datasets_count = fields.Method('get_published_datasets_count')
published_resources_count = fields.Method('get_published_resources_count')
def get_published_datasets_count(self, organization):
return self.context['published_datasets_count']
def get_published_resources_count(self, organization):
return self.context['published_resources_count']
class DatasetApiAttrs(ObjectAttrs, HighlightObjectMixin):
title = TranslatedStr()
slug = TranslatedStr()
notes = TranslatedStr()
categories = fields.Nested(DatasetCategoryAttr, many=True)
category = fields.Nested(DatasetCategoryAttr, many=False)
formats = fields.List(fields.String())
types = fields.List(fields.String())
keywords = KeywordsList(TranslatedStr())
openness_scores = fields.List(fields.Int())
license_chosen = fields.Integer()
license_condition_db_or_copyrighted = fields.String()
license_condition_personal_data = fields.String()
license_condition_modification = fields.Boolean()
license_condition_original = fields.Boolean()
license_condition_responsibilities = fields.String()
license_condition_source = fields.Boolean()
license_condition_timestamp = fields.Boolean()
license_name = fields.String()
license_description = fields.String()
update_frequency = TransUpdateFreqField()
views_count =\
fields.Function(
lambda obj: obj.computed_views_count if is_enabled('S16_new_date_counters.be') else obj.views_count)
downloads_count =\
fields.Function(
lambda obj: obj.computed_downloads_count if is_enabled('S16_new_date_counters.be') else obj.downloads_count)
url = fields.String()
followed = fields.Boolean()
modified = fields.DateTime()
resource_modified = fields.DateTime()
created = fields.DateTime()
verified = fields.DateTime()
visualization_types = ListWithoutNoneStrElement(fields.Str())
source = fields.Nested(SourceSchema)
image_url = fields.Str()
image_alt = TranslatedStr()
if is_enabled('S35_high_value_data.be'):
has_high_value_data = fields.Boolean()
if is_enabled('S37_resources_admin_region_data.be'):
regions = fields.Nested(RegionSchema, many=True)
class Meta:
relationships_schema = DatasetApiRelationships
object_type = 'dataset'
url_template = '{api_url}/datasets/{ident}'
model = 'datasets.Dataset'
class ResourceXMLSerializer(schemas.ExtSchema):
id = fields.Integer()
access_url = fields.Url(attribute='frontend_absolute_url')
title = TranslatedStr()
description = TranslatedStr()
openness_score = fields.Integer()
format = fields.Str()
views_count = fields.Int(attribute='computed_views_count')
downloads_count = fields.Int(attribute='computed_downloads_count')
created = fields.DateTime(format='iso8601')
data_date = fields.Date()
type = fields.Function(lambda resource: resource.get_type_display())
file_size = fields.Function(lambda obj: sizeof_fmt(obj.file_size)
if obj.file_size else '')
visualization_types = ListWithoutNoneStrElement(fields.Str())
download_url = fields.Str()
data_special_signs = fields.Nested(SpecialSignSchema,
data_key='special_signs',
many=True)
class DatasetXMLSerializer(ExtSchema):
id = fields.Integer()
url = fields.Url(attribute='frontend_absolute_url')
title = TranslatedStr()
notes = TranslatedStr()
keywords = fields.Function(lambda dataset: (
tag.name for tag in getattr(dataset, f'tags_{get_language()}')))
categories = fields.Nested(DatasetCategoryAttr, many=True)
update_frequency = TransUpdateFreqField()
created = fields.DateTime()
verified = fields.DateTime()
views_count = fields.Int(attribute='computed_views_count')
downloads_count = fields.Int(attribute='computed_downloads_count')
published_resources_count = fields.Int(
attribute='published_resources__count')
license = fields.Str(attribute='license_name')
conditions = fields.Method('get_conditions')
organization = fields.Method('get_organization')
resources = fields.Method('get_resources')
source = fields.Nested(SourceXMLSchema)
def get_conditions(self, dataset):
conditions = _(
'This dataset is public information, it can be reused under the following conditions: '
)
terms = [
str(
dataset._meta.get_field(
'license_condition_modification').verbose_name)
if dataset.license_condition_modification else '',
str(
dataset._meta.get_field('license_condition_source').
verbose_name) if dataset.license_condition_source else '',
dataset.license_condition_db_or_copyrighted,
dataset.license_condition_personal_data
]
return conditions + '\n'.join([term for term in terms if term])
def get_organization(self, dataset):
context = {
'published_datasets_count':
dataset.organization_published_datasets__count,
'published_resources_count':
dataset.organization_published_resources__count,
}
return InstitutionXMLSerializer(many=False, context=context).dump(
dataset.organization)
def get_resources(self, dataset):
return ResourceXMLSerializer(many=True).dump(
dataset.published_resources)
class InstitutionCSVMetadataSerializer(ExtSchema):
organization_url = fields.Url(attribute='frontend_absolute_url',
data_key=_('Organization URL'))
organization_type = fields.Function(
lambda obj: obj.get_institution_type_display(),
data_key=_('Institution type'))
organization_title = TranslatedStr(attribute='title', data_key=_('Name'))
organization_abbr_title = TranslatedStr(attribute='abbreviation',
data_key=_('Abbreviation'),
default='')
organization_regon = fields.Str(data_key=_('REGON'), attribute='regon')
organization_epuap = fields.Str(attribute='epuap',
data_key=_('EPUAP'),
default='')
organization_website = fields.Url(attribute='website',
data_key=_('Website'))
organization_created = fields.DateTime(attribute='created',
data_key=_('Organization created'),
format='iso8601')
organization_modified = fields.DateTime(
attribute='modified',
data_key=_('Organization modified'),
format='iso8601')
organization_datasets_count = fields.Method(
'get_published_datasets_count', data_key=_('Number of datasets'))
organization_resources_count = fields.Method(
'get_published_resources_count',
data_key=_('Number of organization resources'))
organization_postal_code = fields.Str(attribute='postal_code',
data_key=_('Postal code'))
organization_city = fields.Str(attribute='city', data_key=_('City'))
organization_street_type = fields.Str(attribute='street_type',
data_key=_('Street type'))
organization_street = fields.Str(attribute='street', data_key=_('Street'))
organization_street_number = fields.Str(attribute='street_number',
data_key=_('Street number'))
organization_flat_number = fields.Str(attribute='flat_number',
data_key=_('Flat number'))
organization_email = fields.Email(attribute='email', data_key=_('Email'))
organization_phone_number = fields.Str(attribute='tel',
data_key=_('Phone'))
class Meta:
ordered = True
def get_published_datasets_count(self, organization):
return self.context['published_datasets_count']
def get_published_resources_count(self, organization):
return self.context['published_resources_count']
class DatasetResourcesCSVSerializer(CSVSerializer):
dataset_url = fields.Url(attribute='frontend_absolute_url',
data_key=_('Dataset URL'))
dataset_title = TranslatedStr(attribute='title', data_key=_('Title'))
dataset_description = TranslatedStr(attribute='notes', data_key=_('Notes'))
dataset_keywords = fields.Function(lambda obj: ', '.join(
(tag.name for tag in getattr(obj, f'tags_{get_language()}'))),
data_key=_('Tag'))
dataset_categories = fields.Function(lambda obj: ', '.join(
(category.title_i18n for category in obj.categories.all())),
data_key=_('Category'))
dataset_update_frequency = fields.Str(attribute='frequency_display',
data_key=_('Update frequency'))
dataset_created = fields.DateTime(attribute='created',
data_key=_('Dataset created'),
format='iso8601')
dataset_verified = fields.DateTime(attribute='verified',
data_key=_('Dataset verified'),
format='iso8601')
views_count = fields.Int(attribute='computed_views_count',
data_key=_("Dataset views count"))
downloads_count = fields.Int(attribute='computed_downloads_count',
data_key=_('Dataset downloads count'))
dataset_resources_count = fields.Int(
attribute='published_resources__count', data_key=_('Number of data'))
dataset_conditions = fields.Method('get_dataset_conditions',
data_key=_('Terms of use'))
dataset_license = fields.Str(attribute='license_name',
data_key=_('License'))
dataset_source = fields.Nested(SourceXMLSchema,
attribute='source',
data_key=_('source'))
organization = fields.Method('get_organization')
resources = fields.Nested(ResourceCSVMetadataSerializer,
many=True,
attribute='published_resources')
@ma.post_dump(pass_many=True)
def unpack_nested_data(self, data, many, **kwargs):
new_result_data = []
for record in data:
resources = record.pop('resources')
organization = record.pop('organization')
record.update(**organization)
for resource in resources:
tmp_record = record.copy()
tmp_record.update(**resource)
new_result_data.append(tmp_record)
return new_result_data
def get_dataset_conditions(self, dataset):
conditions = _(
'This dataset is public information, it can be reused under the following conditions: '
)
terms = [
str(
dataset._meta.get_field(
'license_condition_modification').verbose_name)
if dataset.license_condition_modification else '',
str(
dataset._meta.get_field('license_condition_source').
verbose_name) if dataset.license_condition_source else '',
dataset.license_condition_db_or_copyrighted,
dataset.license_condition_personal_data
]
return conditions + '\n'.join([term for term in terms if term])
@ma.post_dump(pass_many=False)
def prepare_nested_data(self, data, **kwargs):
source = data.get(_('source'))
if source:
source_str =\
'{title_label}: {title}, {url_label}: {url},' \
' {last_import_label}: {last_import}, {frequency_label}: {frequency}'.format(
title=source['title'], title_label=_('name'),
url=source['url'], url_label=_('url'),
last_import=source['last_import_timestamp'], last_import_label=_('last import timestamp'),
frequency=source['update_frequency'], frequency_label=_('Update frequency')
)
data[_('source')] = source_str
data[_('Notes')] = strip_tags(data[_('Notes')])
return data
def get_organization(self, dataset):
context = {
'published_datasets_count':
dataset.organization_published_datasets__count,
'published_resources_count':
dataset.organization_published_resources__count,
}
return InstitutionCSVMetadataSerializer(
many=False, context=context).dump(dataset.organization)
def get_csv_headers(self):
result = []
for field_name, field in self.fields.items():
if field_name == 'organization':
org_headers = [
org_field.data_key for org_field_name, org_field in
InstitutionCSVMetadataSerializer().fields.items()
]
result.extend(org_headers)
elif field_name == 'resources':
res_headers = [
res_field.data_key for res_field_name, res_field in
field.schema.fields.items()
]
result.extend(res_headers)
else:
header = field.data_key or field_name
result.append(header)
return result
class Meta:
ordered = True
class DatasetRDFResponseSchema(ProfilesMixin, RDFResponseSchema):
identifier = ma.fields.Function(lambda ds: ds.frontend_absolute_url)
id = ma.fields.Str()
frontend_absolute_url = ma.fields.Str()
title_pl = ma.fields.Str(attribute='title_translated.pl')
title_en = ma.fields.Str(attribute='title_translated.en')
notes_pl = ma.fields.Str(attribute='notes_translated.pl')
notes_en = ma.fields.Str(attribute='notes_translated.en')
status = ma.fields.Str()
created = ma.fields.DateTime()
modified = ma.fields.DateTime()
landing_page = fields.Function(lambda ds: ds.frontend_absolute_url)
version = ma.fields.Str()
tags = rdf_fields.Tags(ma.fields.Str())
resources = ma.fields.Function(resources_dump)
organization = ma.fields.Function(organization_dump)
categories = ma.fields.Function(categories_dump)
update_frequency = DcatUpdateFrequencyField()
license = ma.fields.Function(lambda ds: ds.license_link)
if is_enabled('S38_dcat_spatial_data.be'):
spatial = ma.fields.Nested(RDFRegionSchema,
many=True,
attribute='regions')
@staticmethod
def _from_path(es_resp, path):
try:
obj = es_resp
for step in path.split('.'):
obj = getattr(obj, step)
return obj
except AttributeError:
return None
@ma.pre_dump(pass_many=True)
def extract_pagination(self, data, many, **kwargs):
request = self.context['request'] if 'request' in self.context else None
cleaned_data = dict(getattr(request.context, 'cleaned_data',
{})) if request else {}
def _get_page_link(page_number):
cleaned_data['page'] = page_number
return '{}{}?{}'.format(settings.API_URL, request.path,
builder.build(cleaned_data))
if self.many:
page, per_page = cleaned_data.get('page', 1), cleaned_data.get(
'per_page', 20)
self.context['count'] = self._from_path(data, 'hits.total')
self.context['per_page'] = per_page
items_count = self._from_path(data, 'hits.total')
if page > 1:
self.context['first_page'] = _get_page_link(1)
self.context['prev_page'] = _get_page_link(page - 1)
if items_count:
max_count = min(items_count, 10000)
off = 1 if max_count % per_page else 0
last_page = max_count // per_page + off
if last_page > 1:
self.context['last_page'] = _get_page_link(last_page)
if page * per_page < max_count:
self.context['next_page'] = _get_page_link(page + 1)
return data
@ma.pre_dump(pass_many=True)
def prepare_datasets(self, data, many, **kwargs):
self.context['dataset_refs'] = []
if self.many:
self.context['catalog_modified'] = self._from_path(
data, 'aggregations.catalog_modified.value_as_string')
dataset_ids = [x.id for x in data]
data = Dataset.objects.filter(pk__in=dataset_ids)
return data
@ma.post_dump(pass_many=False)
def prepare_graph_triples(self, data, **kwargs):
self.context['dataset_refs'].append(data['frontend_absolute_url'])
dataset = self.get_rdf_class_for_model(model=Dataset)()
return dataset.to_triples(data, self.include_nested_triples)
@ma.post_dump(pass_many=True)
def prepare_graph(self, data, many, **kwargs):
graph = ExtendedGraph(ordered=True)
self.add_bindings(graph=graph)
# Jeżeli many == True, to serializujemy katalog.
if many:
triples = []
# Dla katalogu, w data, mamy listę list, trzeba to spłaszczyć.
for _triples in data:
triples.extend(_triples)
self.add_pagination_bindings(graph=graph)
paged_collection = HYDRAPagedCollection()
triples.extend(paged_collection.to_triples(self.context))
catalog = self.get_rdf_class_for_catalog()()
triples.extend(catalog.to_triples(self.context))
else:
triples = data
for triple in triples:
graph.add(triple)
return graph
class Meta:
model = 'datasets.Dataset'