def embed(self):
"""Get the html to embed into the crowdsource"""
try:
# first try to get embed code from oEmbed
return mark_safe(
PyEmbed(
# we don't use the default discoverer because it contains a bug
# that makes it always match spotify
discoverer=ChainingDiscoverer([
FileDiscoverer(
resource_filename(
__name__, 'oembed_providers.json'
)
),
AutoDiscoverer(),
])
).embed(self.url, max_height=400)
)
except PyEmbedConsumerError:
# if this is a private document cloud document, it will not have
# an oEmbed, create the embed manually
doc_match = DOCUMENT_URL_RE.match(self.url)
if doc_match:
return mark_safe(
DOCCLOUD_EMBED.format(doc_id=doc_match.group('doc_id'))
)
else:
# fall back to a simple iframe
return format_html(
'<iframe src="{}" width="100%" height="400px"></iframe>',
self.url,
)
def process_data_csv(self, crowdsource):
"""Create the crowdsource data from the uploaded CSV"""
url_validator = URLValidator()
data_csv = self.cleaned_data["data_csv"]
doccloud_each_page = self.cleaned_data["doccloud_each_page"]
if data_csv:
reader = csv.reader(codecs.iterdecode(data_csv, "utf-8"))
headers = [h.lower() for h in next(reader)]
for line in reader:
data = dict(list(zip(headers, line)))
url = data.pop("url", "")
doc_match = DOCUMENT_URL_RE.match(url)
proj_match = PROJECT_URL_RE.match(url)
if doccloud_each_page and doc_match:
datum_per_page.delay(crowdsource.pk,
doc_match.group("doc_id"), data)
elif proj_match:
import_doccloud_proj.delay(
crowdsource.pk,
proj_match.group("proj_id"),
data,
doccloud_each_page,
)
elif url:
# skip invalid URLs
try:
url_validator(url)
except forms.ValidationError:
pass
else:
crowdsource.data.create(url=url, metadata=data)
else:
crowdsource.data.create(metadata=data)
def save(self, commit=True, doccloud_each_page=False):
"""Apply special cases to Document Cloud URLs"""
instances = super(CrowdsourceDataFormset, self).save(commit=False)
return_instances = []
for instance in instances:
doc_match = DOCUMENT_URL_RE.match(instance.url)
proj_match = PROJECT_URL_RE.match(instance.url)
if doccloud_each_page and doc_match:
datum_per_page.delay(
self.instance.pk,
doc_match.group('doc_id'),
{},
)
elif proj_match:
import_doccloud_proj.delay(
self.instance.pk,
proj_match.group('proj_id'),
{},
doccloud_each_page,
)
else:
return_instances.append(instance)
if commit:
instance.save()
return return_instances
def embed(self):
"""Get the html to embed into the crowdsource"""
try:
# first try to get embed code from oEmbed
return mark_safe(PyEmbed().embed(self.url, max_height=400))
except PyEmbedConsumerError:
# if this is a private document cloud document, it will not have
# an oEmbed, create the embed manually
doc_match = DOCUMENT_URL_RE.match(self.url)
if doc_match:
return mark_safe(
DOCCLOUD_EMBED.format(doc_id=doc_match.group('doc_id')))
else:
# fall back to a simple iframe
return format_html(
'<iframe src="{}" width="100%" height="400px"></iframe>',
self.url,
)