def _author_list(obj, eng):
from inspirehep.modules.converter.xslt import convert
model = eng.workflow_definition.model(obj)
record = get_record_from_model(model)
arxiv_id = get_arxiv_id_from_record(record)
existing_file = get_file_by_name(model, "{0}.tar.gz".format(arxiv_id))
if not existing_file:
# We download it
tarball = get_tarball_for_model(eng, arxiv_id)
if tarball is None:
obj.log.error("No tarball found")
return
add_file_by_name(model, tarball)
else:
tarball = existing_file.get_syspath()
sub_dir = os.path.abspath("{0}_files".format(tarball))
try:
file_list = untar(tarball, sub_dir)
except InvalidTarball:
obj.log.error("Invalid tarball {0}".format(tarball))
return
obj.log.info("Extracted tarball to: {0}".format(sub_dir))
xml_files_list = [filename for filename in file_list
if filename.endswith(".xml")]
obj.log.info("Found xmlfiles: {0}".format(xml_files_list))
for xml_file in xml_files_list:
xml_file_fd = open(xml_file, "r")
xml_content = xml_file_fd.read()
xml_file_fd.close()
match = REGEXP_AUTHLIST.findall(xml_content)
if match:
obj.log.info("Found a match for author extraction")
authors_xml = convert(xml_content, stylesheet)
authorlist_record = get_json_from_marcxml(authors_xml)[0]
record.update(authorlist_record)
obj.update_task_results(
"authors",
[{
"name": "authors",
"results": authorlist_record["authors"]
}]
)
obj.update_task_results(
"number_of_authors",
[{
"name": "number_of_authors",
"results": authorlist_record["number_of_authors"]
}]
)
break
model.update()
def test_payload_file_creation(self):
"""Can add a file to a Payload."""
from invenio_workflows.models import BibWorkflowObject
from inspirehep.modules.workflows.models import Payload
from inspirehep.utils.helpers import (
get_file_by_name,
add_file_by_name,
)
obj = BibWorkflowObject.create_object()
obj.save()
obj.data = obj.get_data() # FIXME hack until workflow 2.0
payload = Payload.create(workflow_object=obj, type="payload_fixture")
payload.save()
fd, filename = tempfile.mkstemp()
os.close(fd)
newpath = add_file_by_name(payload, filename)
self.assertTrue(newpath)
self.assertTrue(get_file_by_name(payload,
os.path.basename(filename)))
BibWorkflowObject.delete(obj)
def arxiv_refextract(obj, eng):
"""Perform the reference extraction step.
:param obj: Bibworkflow Object to process
:param eng: BibWorkflowEngine processing the object
"""
model = eng.workflow_definition.model(obj)
record = get_record_from_model(model)
arxiv_id = get_arxiv_id_from_record(record)
existing_file = get_file_by_name(model, "{0}.pdf".format(arxiv_id))
if not existing_file:
# We download it
pdf = get_pdf_for_model(eng, arxiv_id)
if pdf is None:
obj.log.error("No pdf found")
return
add_file_by_name(model, pdf)
else:
pdf = existing_file.get_syspath()
if pdf and os.path.isfile(pdf):
mapped_references = extract_references(pdf)
if mapped_references:
record["references"] = mapped_references
obj.log.info("Extracted {0} references".format(
len(mapped_references)
))
obj.update_task_results(
"References",
[{"name": "References",
"result": mapped_references,
"template": "workflows/results/refextract.html"}]
)
model.update()
else:
obj.log.info("No references extracted")
else:
obj.log.error("Not able to download and process the PDF")
def arxiv_plot_extract(obj, eng):
"""Extract plots from an arXiv archive."""
model = eng.workflow_definition.model(obj)
record = get_record_from_model(model)
arxiv_id = get_arxiv_id_from_record(record)
existing_file = get_file_by_name(model, "{0}.tar.gz".format(arxiv_id))
if not existing_file:
# We download it
tarball = get_tarball_for_model(eng, arxiv_id)
if tarball is None:
obj.log.error("No tarball found")
return
add_file_by_name(model, tarball)
else:
tarball = existing_file.get_syspath()
try:
plots = process_tarball(tarball)
except InvalidTarball:
eng.log.error(
'Invalid tarball {0}'.format(tarball)
)
return
if plots:
# We store the path to the directory the tarball contents lives
new_dict = get_json_for_plots(plots)
record.update(new_dict)
obj.update_task_results(
"Plots",
[{
"name": "Plots",
"result": new_dict["fft"],
"template": "workflows/results/plots.html"
}]
)
obj.log.info("Added {0} plots.".format(len(new_dict["fft"])))
model.update()
def _arxiv_fulltext_download(obj, eng):
model = eng.workflow_definition.model(obj)
record = get_record_from_model(model)
arxiv_id = get_arxiv_id_from_record(record)
existing_file = get_file_by_name(model, "{0}.pdf".format(arxiv_id))
if not existing_file:
# We download it
pdf = get_pdf_for_model(eng, arxiv_id)
if pdf is None:
obj.log.error("No pdf found")
return
pdf = add_file_by_name(model, pdf)
obj.extra_data["pdf"] = pdf
else:
pdf = existing_file.get_syspath()
if pdf:
if "fft" in record:
record["fft"].append({
"url": pdf,
"docfile_type": doctype
})
else:
new_dict_representation = {
"fft": [
{
"url": pdf,
"docfile_type": doctype
}
]
}
record.update(new_dict_representation)
fileinfo = {
"type": "fulltext",
"filename": os.path.basename(pdf),
"full_path": pdf,
}
obj.update_task_results(
os.path.basename(pdf),
[{
"name": "PDF",
"result": fileinfo,
"template": "workflows/results/files.html"
}]
)
model.update()
else:
obj.log.info("No PDF found.")
