def check_arxiv_published(value,
field="id",
get_first=True,
keep_eprint=False):
found = False
published = False
bib = ""
value = re.sub("arxiv\:", "", value, flags=re.I)
found, items = get_arxiv_info(value, field)
if found:
if get_first is False and field == "ti" and len(items) > 1:
found, item = ask_which_is(value, items)
else:
item = items[0]
if found:
if "arxiv_doi" in item:
doi = item["arxiv_doi"]
published, bib = get_bib_from_doi(doi)
if keep_eprint:
eprint = re.split('/|v', item["id"])[-2]
bib = add_eprint_to_bib(bib, eprint)
else:
bib = generate_bib_from_arxiv(item, value, field)
else:
print("\t\nArxiv not found.")
return found, published, bib
def update_bib(bib):
bib_id = bib["ID"]
if "doi" in bib:
found, bib_string = get_bib_from_doi(bib["doi"])
if found:
bib = bibtexparser.loads(bib_string).entries[0]
bib["ID"] = bib_id
return bib
def get_bib_from_title(title, get_first=False, abbrev_journal=False):
found = False
bib = ""
found, item = get_from_title(title, get_first)
if found:
if item["is_crossref"]:
if "DOI" in item:
doi = item["DOI"]
found, bib = get_bib_from_doi(doi)
else:
if "arxiv_doi" in item:
doi = item["arxiv_doi"]
published, bib = get_bib_from_doi(doi, abbrev_journal)
else:
bib = generate_bib_from_arxiv(item, title, field="ti")
# if "short-container-title" in item:
# abbreviated_journal = item["short-container-title"][0]##aqui pode acontecer de realizar a chamada para
return found, bib
def pdf2bib(pdf_file):
"""
Given a PDF file, tries to extract the paper's DOI and fetch the BibTex entry
:param pdf_file: the path to the PDF file
:return: The bibtex entry as a string
"""
found_a_doi = False
bib_string = ''
pdf_text = get_pdf_page_text(pdf_file)
# Try each of the regexes in sequence. Hopefully one will work.
for doi_re in all_doi_res:
doi_match = doi_re.search(pdf_text)
# If we did not find a match, try the next one. If we did, make a note of that, because that
# will affect the error we give if this fails.
if doi_match is None:
continue
else:
found_a_doi = True
doi_string = doi_match.group(0)
# Assume (for now) that the DOI suffix cannot include Unicode. This will stop the DOI at the first non-ASCII
# character. Which corrects an issue with e.g. doi:10.5194/acp-11-8543-2011 where in the PDF the (c) symbol comes
# right after the DOI and gets included
last_idx = 0
for idx, char in enumerate(doi_string):
if ord(char) > 127:
break
else:
last_idx = idx + 1
doi_string = doi_string[:last_idx]
root_logger.debug('Looking up DOI "{}"'.format(doi_string))
# Try to retrieve the bib string based on the doi. If we do so successfully, go ahead and return.
# If not, then try the next regex. If there are none left, then we'll leave bib_string as an empty
# string and raise the appropriate error.
success, bib_string = get_bib_from_doi(doi_string)
if success:
break
else:
bib_string = ''
if not found_a_doi:
raise DoiNotFoundError('DOI search failed on {}'.format(pdf_file))
elif len(bib_string) == 0:
raise BibRetrievalError(
'Bib string lookup failed on {}'.format(pdf_file))
return bib_string
def main(section):
output = ""
out = reportinator.cache + "/output.bib"
inp = reportinator.cache + "/dois.txt"
f = open(out, "w+")
fp = open(inp, "r")
for line in fp:
doi = line.split(" ")[0]
found, bib = get_bib_from_doi(doi)
if found:
f.write(bib + "\n")
output += "\n\\section{References}\n"
output += "\\nocite{" + "*}\n"
output += "\\printbibliography[heading=none]\n\n"
return output
def _to_bibtex(doi, template, idx):
try:
from doi2bib.crossref import get_bib_from_doi
except ImportError:
print(
"Cannot generate BibTeX citation, missing doi2bib dependency",
file=sys.stderr,
)
return doi
if "doi.org" not in doi:
return doi
bib = get_bib_from_doi(doi)[1]
# replace identifier with template name
m = re.search(r"([A-Z])\w+", bib)
return bib.replace(m.group(), "%s%s" % (template.lower(), idx))
def check_arxiv_published(value, field="id", get_first=True):
found = False
published = False
bib = ""
value = re.sub("arxiv\:", "", value, flags=re.I)
found, items = get_arxiv_info(value, field)
if found:
if get_first is False and field == "ti" and len(items) > 1:
found, item = ask_which_is(value, items)
else:
item = items[0]
if found:
if item["doi"] != None:
doi = item["doi"]
published, bib = get_bib_from_doi(doi)
else:
bib = generate_bib_from_arxiv(item, value, field)
else:
print("\t\nArxiv not found.")
return found, published, bib
dois = set()
with open(f) as h:
for line in h:
for x in re.findall('{doi:([^}]*)}',line):
for y in x.split(','):
y = re.sub(' ','',y)
if not y.startswith('doi:'):
y = 'doi:'+y
dois.add(y)
for x in re.findall('{https://doi.org/([^}]*)}',line):
for y in x.split(','):
y = re.sub(' ','',y)
if not y.startswith('https://doi.org/'):
y = 'https://doi.org/'+y
dois.add(y)
with open(args.bib, 'w') as upv:
if args.b is not None:
with open(args.b) as h:
for line in h:
upv.write(line)
for y in dois - labels:
out = get_bib_from_doi(y)
if out[0]:
x = out[1].split(',')
x[0] = re.sub('{[^,]*','{'+y,x[0])
bib = ','.join(x)
try:
upv.write(bib.encode('UTF-8')+'\n')
except:
import pdb; pdb.set_trace()
def check_arxiv_published(value,
field="id",
get_first=True,
keep_eprint=False):
"""
Parameters
----------
value: str
value of the field
field: str
field used for the arxiv search API
get_first: bool
keep_eprint: bool
If True keep the arxiv number if the paper
has already been published
Returns
-------
found: bool
True if found the arxiv item
published: bool
True if the arxiv has already been
published
bib: str
bibtext string
"""
found = False
published = False
bib = ""
value = re.sub("arxiv\:", "", value, flags=re.I)
found, items = get_arxiv_info(value, field)
if found:
if field == "ti":
title = value.lower().replace(" ", "")
for item_arxiv in items:
title_arxiv = item_arxiv["title"].lower().replace(" ",
"").replace(
"\n", "")
if title_arxiv == title:
items = [item_arxiv]
break
if get_first is False and field == "ti" and len(items) > 1:
found, item = ask_which_is(value, items)
else:
item = items[0]
if found:
if "arxiv_doi" in item:
doi = item["arxiv_doi"]
published, bib = get_bib_from_doi(doi)
if keep_eprint:
eprint = re.split('/|v', item["id"])[-2]
bib = add_eprint_to_bib(bib, eprint)
else:
bib = generate_bib_from_arxiv(item, value, field)
else:
print("\t\nArxiv not found.")
return found, published, bib
