def submission_fulltext_download(obj, eng):
submission_pdf = obj.extra_data.get('submission_pdf')
if submission_pdf and is_pdf_link(submission_pdf):
filename = secure_filename('fulltext.pdf')
pdf = download_file_to_workflow(
workflow=obj,
name=filename,
url=submission_pdf,
)
if pdf:
obj.data['documents'] = [
document for document in obj.data.get('documents', ())
if document.get('key') != filename
]
lb = LiteratureBuilder(source=obj.data['acquisition_source']['source'], record=obj.data)
lb.add_document(
filename,
fulltext=True,
original_url=submission_pdf,
url='/api/files/{bucket}/{key}'.format(bucket=obj.files[filename].bucket_id, key=filename)
)
obj.data = lb.record
obj.log.info('PDF provided by user from %s', submission_pdf)
return obj.files[filename].file.uri
else:
obj.log.info('Cannot fetch PDF provided by user from %s', submission_pdf)
def test_download_file_to_workflow_retries_on_protocol_error():
with requests_mock.Mocker() as requests_mocker:
filename = pkg_resources.resource_filename(
__name__, os.path.join('fixtures', '1605.03844.pdf'))
requests_mocker.register_uri(
'GET', 'http://export.arxiv.org/pdf/1605.03844', [
{
'exc': requests.packages.urllib3.exceptions.ProtocolError
},
{
'body': filename,
'status_code': 200
},
])
data = {}
extra_data = {}
files = MockFiles({})
obj = MockObj(data, extra_data, files=files)
expected = MockFileObject(key='1605.03844.pdf')
result = download_file_to_workflow(
obj, '1605.03844.pdf', 'http://export.arxiv.org/pdf/1605.03844')
assert expected == result
def download_documents(obj, eng):
documents = obj.data.get('documents', [])
for document in documents:
filename = document['key']
url = document['url']
downloaded = download_file_to_workflow(
workflow=obj,
name=filename,
url=url,
)
if downloaded:
document['url'] = '/api/files/{bucket}/{key}'.format(
bucket=obj.files[filename].bucket_id, key=filename)
obj.log.info('Document downloaded from %s', url)
else:
obj.log.error('Cannot download document from %s', url)
def download_documents(obj, eng):
documents = obj.data.get('documents', [])
for document in documents:
filename = document['key']
url = document['url']
downloaded = download_file_to_workflow(
workflow=obj,
name=filename,
url=url,
)
if downloaded:
document['url'] = '/api/files/{bucket}/{key}'.format(
bucket=obj.files[filename].bucket_id, key=filename)
obj.log.info('Document downloaded from %s', url)
else:
obj.log.error(
'Cannot download document from %s', url)
def arxiv_package_download(obj, eng):
"""Perform the package download step for arXiv records.
:param obj: Workflow Object to process
:param eng: Workflow Engine processing the object
"""
arxiv_id = get_arxiv_id(obj.data)
filename = secure_filename('{0}.tar.gz'.format(arxiv_id))
tarball = download_file_to_workflow(
workflow=obj,
name=filename,
url=current_app.config['ARXIV_TARBALL_URL'].format(arxiv_id=arxiv_id),
)
if tarball:
obj.log.info('Tarball retrieved from arXiv for %s', arxiv_id)
else:
obj.log.error('Cannot retrieve tarball from arXiv for %s', arxiv_id)
def arxiv_package_download(obj, eng):
"""Perform the package download step for arXiv records.
:param obj: Workflow Object to process
:param eng: Workflow Engine processing the object
"""
arxiv_id = get_arxiv_id(obj.data)
filename = secure_filename('{0}.tar.gz'.format(arxiv_id))
tarball = download_file_to_workflow(
workflow=obj,
name=filename,
url=current_app.config['ARXIV_TARBALL_URL'].format(arxiv_id=arxiv_id),
)
if tarball:
obj.log.info('Tarball retrieved from arXiv for %s', arxiv_id)
else:
obj.log.error('Cannot retrieve tarball from arXiv for %s', arxiv_id)
def arxiv_fulltext_download(obj, eng):
"""Perform the fulltext download step for arXiv records.
:param obj: Workflow Object to process
:param eng: Workflow Engine processing the object
"""
arxiv_id = get_arxiv_id(obj.data)
filename = secure_filename('{0}.pdf'.format(arxiv_id))
url = current_app.config['ARXIV_PDF_URL'].format(arxiv_id=arxiv_id)
if not is_pdf_link(url):
if NO_PDF_ON_ARXIV in requests.get(url).content:
obj.log.info('No PDF is available for %s', arxiv_id)
return
raise DownloadError("{url} is not serving a PDF file.".format(url=url))
pdf = download_file_to_workflow(
workflow=obj,
name=filename,
url=url,
)
if pdf:
obj.data['documents'] = [
document for document in obj.data.get('documents', ())
if document.get('key') != filename
]
lb = LiteratureBuilder(source='arxiv', record=obj.data)
lb.add_document(filename,
fulltext=True,
hidden=True,
material='preprint',
original_url=url,
url='/api/files/{bucket}/{key}'.format(
bucket=obj.files[filename].bucket_id,
key=filename))
obj.data = lb.record
obj.log.info('PDF retrieved from arXiv for %s', arxiv_id)
else:
obj.log.error('Cannot retrieve PDF from arXiv for %s', arxiv_id)
def test_download_file_to_workflow_retries_on_protocol_error():
with requests_mock.Mocker() as requests_mocker:
filename = pkg_resources.resource_filename(
__name__, os.path.join('fixtures', '1605.03844.pdf'))
requests_mocker.register_uri(
'GET', 'http://export.arxiv.org/pdf/1605.03844', [
{'exc': requests.packages.urllib3.exceptions.ProtocolError},
{'body': filename, 'status_code': 200},
])
data = {}
extra_data = {}
files = MockFiles({})
obj = MockObj(data, extra_data, files=files)
expected = MockFileObject(key='1605.03844.pdf')
result = download_file_to_workflow(
obj, '1605.03844.pdf', 'http://export.arxiv.org/pdf/1605.03844')
assert expected == result
