def __init__(self, filename, mode="rb"):
if mode not in ("rb", "wb"):
raise IOError("mode not supported:", mode)
self._file = builtin_open(filename, mode)
self.filename = filename
if mode == "rb":
self.header = self._decode_header()
self.size = int(round(math.sqrt(self.header.count)))
self.index = self._decode_index()
def fetch(bibcode, filename=None):
"""
Attempt to fetch a PDF file from ADS. If successful, then
add it into the database. If the fetch succeeds but the bibcode is
not in th database, download file to current folder.
Parameters
----------
bibcode: String
ADS bibcode of entry to update.
filename: String
Filename to assign to the PDF file. If None, get from
guess_name() funcion.
"""
replace, arxiv = True, False
print('Fetching PDF file from Journal website:')
req = request_ads(bibcode, source='journal')
if req is None:
return
if req.status_code != 200:
print('Fetching PDF file from ADS website:')
req = request_ads(bibcode, source='ads')
if req is None:
return
if req.status_code != 200:
print('Fetching PDF file from ArXiv website:')
req = request_ads(bibcode, source='arxiv')
replace, arxiv = False, True
if req is None:
return
if req.status_code == 200:
if bm.find(bibcode=bibcode) is None:
if filename is None:
filename = f'{bibcode}.pdf'
with builtin_open(filename, 'wb') as f:
f.write(req.content)
print(f"Saved PDF to: '{filename}'.\n"
"(Note that BibTex entry is not in the Bibmanager database)")
else:
set_pdf(bibcode,
bin_pdf=req.content,
filename=filename,
arxiv=arxiv,
replace=replace)
return
print('Could not fetch PDF from any source.')
def set_pdf(bib,
pdf=None,
bin_pdf=None,
filename=None,
arxiv=False,
replace=False):
"""
Update the PDF file of the given BibTex entry in database
If pdf is not None, move the file into the database pdf folder.
Parameters
----------
bibcode: String or Bib() instance
Entry to be updated (must exist in the Bibmanager database).
If string, the ADS bibcode of key ID of the entry.
pdf: String
Path to an existing PDF file.
Only one of pdf and bin_pdf must be not None.
bin_pdf: String
PDF content in binary format (e.g., as in req.content).
Only one of pdf and bin_pdf must be not None.
arxiv: Bool
Flag indicating the source of the PDF. If True,
filename: String
Filename to assign to the PDF file. If None, take name from
pdf input argument, or else from guess_name().
replace: Bool
Replace without asking if the entry already has a PDF assigned;
else, ask the user.
"""
if isinstance(bib, str):
e = bm.find(key=bib)
bib = bm.find(bibcode=bib) if e is None else e
if bib is None:
raise ValueError('BibTex entry is not in Bibmanager database')
if (pdf is None) + (bin_pdf is None) != 1:
raise ValueError('Exactly one of pdf or bin_pdf must be not None')
# Let's have a guess, if needed:
guess_filename = guess_name(bib, arxiv=arxiv)
if filename is None:
filename = os.path.basename(pdf) if pdf is not None else guess_filename
if not filename.lower().endswith('.pdf'):
raise ValueError('Invalid filename, must have a .pdf extension')
if os.path.dirname(filename) != '':
raise ValueError('filename must not have a path')
if pdf is not None and bib.pdf is not None:
pdf_is_not_bib_pdf = os.path.abspath(pdf) != f'{u.BM_PDF()}{bib.pdf}'
else:
pdf_is_not_bib_pdf = True
# PDF files in BM_PDF (except for the entry being fetched):
pdf_names = [
file for file in os.listdir(u.BM_PDF())
if os.path.splitext(file)[1].lower() == '.pdf'
]
with u.ignored(ValueError):
pdf_names.remove(bib.pdf)
if pdf == f'{u.BM_PDF()}{filename}':
pdf_names.remove(filename)
if not replace and bib.pdf is not None and pdf_is_not_bib_pdf:
rep = u.req_input(
f"Bibtex entry already has a PDF file: '{bib.pdf}' "
"Replace?\n[]yes, [n]o.\n",
options=['', 'y', 'yes', 'n', 'no'])
if rep in ['n', 'no']:
return
while filename in pdf_names:
overwrite = input(
f"A filename '{filename}' already exists. Overwrite?\n"
f"[]yes, [n]o, or type new file name (e.g., {guess_filename}).\n")
if overwrite in ['n', 'no']:
return
elif overwrite in ['', 'y', 'yes']:
break
elif overwrite.lower().endswith('.pdf'):
filename = overwrite
# Delete pre-existing file only if not merely renaming:
if pdf is None or pdf_is_not_bib_pdf:
with u.ignored(OSError):
os.remove(f"{u.BM_PDF()}{bib.pdf}")
if pdf is not None:
shutil.move(pdf, f"{u.BM_PDF()}{filename}")
else:
with builtin_open(f"{u.BM_PDF()}{filename}", 'wb') as f:
f.write(bin_pdf)
print(f"Saved PDF to: '{u.BM_PDF()}{filename}'.")
# Update entry and database:
bibs = bm.load()
index = bibs.index(bib)
bib.pdf = filename
bibs[index] = bib
bm.save(bibs)
bm.export(bibs, meta=True)
def fetch(bibcode, filename=None, replace=None):
"""
Attempt to fetch a PDF file from ADS. If successful, then
add it into the database. If the fetch succeeds but the bibcode is
not in the database, download file to current folder.
Parameters
----------
bibcode: String
ADS bibcode of entry to update.
filename: String
Filename to assign to the PDF file. If None, get from
guess_name() function.
Replace: Bool
If True, enforce replacing a PDF regardless of a pre-existing one.
If None (default), only ask when fetched PDF comes from arxiv.
Returns
-------
filename: String
If successful, return the full path of the file name.
If not, return None.
"""
arxiv = False
print('Fetching PDF file from Journal website:')
req = request_ads(bibcode, source='journal')
if req is None:
return
if req.status_code != 200:
print('Fetching PDF file from ADS website:')
req = request_ads(bibcode, source='ads')
if req is None:
return
if req.status_code != 200:
print('Fetching PDF file from ArXiv website:')
req = request_ads(bibcode, source='arxiv')
arxiv = True
if replace is None:
replace = False
if req is None:
return
if replace is None:
replace = True
if req.status_code == 200:
if bm.find(bibcode=bibcode) is None:
if filename is None:
filename = f'{bibcode}.pdf'
with builtin_open(filename, 'wb') as f:
f.write(req.content)
print(f"Saved PDF to: '{filename}'.\n"
"(Note that BibTex entry is not in the Bibmanager database)")
else:
filename = set_pdf(bibcode,
bin_pdf=req.content,
filename=filename,
arxiv=arxiv,
replace=replace)
return filename
print('Could not fetch PDF from any source.')