def get_pdf_for_model(eng, arxiv_id):
"""We download it."""
extract_path = os.path.join(
cfg.get('OAIHARVESTER_STORAGEDIR', cfg.get('CFG_TMPSHAREDDIR')),
str(eng.uuid)
)
return get_pdf_from_arxiv(
arxiv_id,
extract_path
)
def arxiv_fulltext_download(obj, eng):
"""Perform the fulltext download step for arXiv records.
:param obj: Bibworkflow Object to process
:param eng: BibWorkflowEngine processing the object
"""
from invenio.utils.plotextractor.api import get_pdf_from_arxiv
if "result" not in obj.extra_data:
obj.extra_data["_result"] = {}
if "pdf" not in obj.extra_data["_result"]:
extract_path = os.path.join(
cfg.get('OAIHARVESTER_STORAGEDIR', cfg.get('CFG_TMPSHAREDDIR')),
str(eng.uuid))
pdf = get_pdf_from_arxiv(
obj.data.get(cfg.get('OAIHARVESTER_RECORD_ARXIV_ID_LOOKUP')),
extract_path)
arguments = obj.extra_data["repository"]["arguments"]
try:
if not arguments['t_doctype'] == '':
doctype = arguments['t_doctype']
else:
doctype = 'arXiv'
except KeyError:
eng.log.error("WARNING: HASARDOUS BEHAVIOUR EXPECTED, "
"You didn't specified t_doctype in argument"
" for fulltext_download,"
"try to recover by using the default one!")
doctype = 'arXiv'
if pdf:
obj.extra_data["_result"]["pdf"] = pdf
new_dict_representation = {
"fft": [{
"url": pdf,
"docfile_type": doctype
}]
}
_attach_files_to_obj(obj, new_dict_representation)
fileinfo = {
"type": "fulltext",
"filename": os.path.basename(pdf),
"full_path": pdf,
}
obj.update_task_results(
"PDF", [{
"name": "PDF",
"result": fileinfo,
"template": "workflows/results/fft.html"
}])
else:
obj.log.info("No PDF found.")
else:
eng.log.info("There was already a pdf register for this record,"
"perhaps a duplicate task in you workflow.")
def arxiv_refextract(obj, eng):
"""Perform the reference extraction step.
:param obj: Bibworkflow Object to process
:param eng: BibWorkflowEngine processing the object
"""
from invenio.legacy.refextract.api import extract_references_from_file_xml
from invenio.utils.plotextractor.api import get_pdf_from_arxiv
from invenio.modules.workflows.utils import convert_marcxml_to_bibfield
if "_result" not in obj.extra_data:
obj.extra_data["_result"] = {}
try:
pdf = obj.extra_data["_result"]["pdf"]
except KeyError:
pdf = None
if not pdf:
extract_path = os.path.join(
cfg.get('OAIHARVESTER_STORAGEDIR', cfg.get('CFG_TMPSHAREDDIR')),
str(eng.uuid)
)
pdf = get_pdf_from_arxiv(
obj.data.get(cfg.get('OAIHARVESTER_RECORD_ARXIV_ID_LOOKUP')),
extract_path
)
obj.extra_data["_result"]["pdf"] = pdf
if pdf and os.path.isfile(pdf):
references_xml = extract_references_from_file_xml(
obj.extra_data["_result"]["pdf"]
)
if references_xml:
updated_xml = '<?xml version="1.0" encoding="UTF-8"?>\n' \
'<collection>\n' + references_xml + \
"\n</collection>"
new_dict_representation = convert_marcxml_to_bibfield(updated_xml)
if "reference" in new_dict_representation:
obj.data["reference"] = new_dict_representation["reference"]
obj.log.info("Extracted {0} references".format(len(obj.data["reference"])))
obj.update_task_results(
"References",
[{"name": "References",
"result": new_dict_representation['reference'],
"template": "workflows/results/refextract.html"}]
)
return
else:
obj.log.info("No references extracted")
else:
obj.log.error("Not able to download and process the PDF")
def arxiv_refextract(obj, eng):
"""Perform the reference extraction step.
:param obj: Bibworkflow Object to process
:param eng: BibWorkflowEngine processing the object
"""
from invenio.legacy.refextract.api import extract_references_from_file_xml
from invenio.utils.plotextractor.api import get_pdf_from_arxiv
from invenio.modules.workflows.utils import convert_marcxml_to_bibfield
if "_result" not in obj.extra_data:
obj.extra_data["_result"] = {}
try:
pdf = obj.extra_data["_result"]["pdf"]
except KeyError:
pdf = None
if not pdf:
extract_path = os.path.join(
cfg.get('OAIHARVESTER_STORAGEDIR', cfg.get('CFG_TMPSHAREDDIR')),
str(eng.uuid))
pdf = get_pdf_from_arxiv(
obj.data.get(cfg.get('OAIHARVESTER_RECORD_ARXIV_ID_LOOKUP')),
extract_path)
obj.extra_data["_result"]["pdf"] = pdf
if pdf and os.path.isfile(pdf):
references_xml = extract_references_from_file_xml(
obj.extra_data["_result"]["pdf"])
if references_xml:
updated_xml = '<?xml version="1.0" encoding="UTF-8"?>\n' \
'<collection>\n' + references_xml + \
"\n</collection>"
new_dict_representation = convert_marcxml_to_bibfield(updated_xml)
if "reference" in new_dict_representation:
obj.data["reference"] = new_dict_representation["reference"]
obj.log.info("Extracted {0} references".format(
len(obj.data["reference"])))
obj.update_task_results(
"References",
[{
"name": "References",
"result": new_dict_representation['reference'],
"template": "workflows/results/refextract.html"
}])
return
else:
obj.log.info("No references extracted")
else:
obj.log.error("Not able to download and process the PDF")
def _arxiv_fulltext_download(obj, eng):
"""Perform the fulltext download step for arXiv records.
:param obj: Bibworkflow Object to process
:param eng: BibWorkflowEngine processing the object
"""
from invenio.utils.plotextractor.api import get_pdf_from_arxiv
if "result" not in obj.extra_data:
obj.extra_data["_result"] = {}
if "pdf" not in obj.extra_data["_result"]:
extract_path = os.path.join(
cfg.get('OAIHARVESTER_STORAGEDIR', cfg.get('CFG_TMPSHAREDDIR')),
str(eng.uuid)
)
pdf = get_pdf_from_arxiv(
obj.data.get(cfg.get('OAIHARVESTER_RECORD_ARXIV_ID_LOOKUP')),
extract_path
)
if pdf:
obj.extra_data["_result"]["pdf"] = pdf
new_dict_representation = {
"fft": [
{
"url": pdf,
"docfile_type": doctype
}
]
}
_attach_files_to_obj(obj, 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"
}]
)
else:
obj.log.info("No PDF found.")
else:
eng.log.info("There was already a pdf register for this record,"
"perhaps a duplicate task in you workflow.")
def _arxiv_fulltext_download(obj, eng):
"""Perform the fulltext download step for arXiv records.
:param obj: Bibworkflow Object to process
:param eng: BibWorkflowEngine processing the object
"""
from invenio.utils.plotextractor.api import get_pdf_from_arxiv
if "result" not in obj.extra_data:
obj.extra_data["_result"] = {}
if "pdf" not in obj.extra_data["_result"]:
extract_path = os.path.join(
cfg.get('OAIHARVESTER_STORAGEDIR',
cfg.get('CFG_TMPSHAREDDIR')), str(eng.uuid))
pdf = get_pdf_from_arxiv(
obj.data.get(cfg.get('OAIHARVESTER_RECORD_ARXIV_ID_LOOKUP')),
extract_path)
if pdf:
obj.extra_data["_result"]["pdf"] = pdf
new_dict_representation = {
"fft": [{
"url": pdf,
"docfile_type": doctype
}]
}
_attach_files_to_obj(obj, 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"
}])
else:
obj.log.info("No PDF found.")
else:
eng.log.info("There was already a pdf register for this record,"
"perhaps a duplicate task in you workflow.")
def arxiv_fulltext_download(obj, eng):
"""Perform the fulltext download step for arXiv records.
:param obj: Bibworkflow Object to process
:param eng: BibWorkflowEngine processing the object
"""
from invenio.utils.plotextractor.api import get_pdf_from_arxiv
if "result" not in obj.extra_data:
obj.extra_data["_result"] = {}
if "pdf" not in obj.extra_data["_result"]:
extract_path = os.path.join(
cfg.get('OAIHARVESTER_STORAGEDIR', cfg.get('CFG_TMPSHAREDDIR')),
str(eng.uuid)
)
pdf = get_pdf_from_arxiv(
obj.data.get(cfg.get('OAIHARVESTER_RECORD_ARXIV_ID_LOOKUP')),
extract_path
)
arguments = obj.extra_data["repository"]["arguments"]
try:
if not arguments['t_doctype'] == '':
doctype = arguments['t_doctype']
else:
doctype = 'arXiv'
except KeyError:
eng.log.error("WARNING: HASARDOUS BEHAVIOUR EXPECTED, "
"You didn't specified t_doctype in argument"
" for fulltext_download,"
"try to recover by using the default one!")
doctype = 'arXiv'
if pdf:
obj.extra_data["_result"]["pdf"] = pdf
new_dict_representation = {
"fft": [
{
"url": pdf,
"docfile_type": doctype
}
]
}
_attach_files_to_obj(obj, new_dict_representation)
fileinfo = {
"type": "fulltext",
"filename": os.path.basename(pdf),
"full_path": pdf,
}
obj.update_task_results(
"PDF",
[{
"name": "PDF",
"result": fileinfo,
"template": "workflows/results/fft.html"
}]
)
else:
obj.log.info("No PDF found.")
else:
eng.log.info("There was already a pdf register for this record,"
"perhaps a duplicate task in you workflow.")
