def merge_folder_tree(folder, use_backup):
"""
Merge bib files from the current subtree into a master bib file at the root.
This function updates the 'file' link of each entry with the relative path
to each subfolder that has been processed.
Args:
folder (str): relative or absolute path of the folder to process.
Returns:
Nothing, but creates a file named `master.bib` in the given folder.
"""
db = BibDatabase()
for subdir, _dirs, _files in os.walk(os.path.abspath(folder)):
if os.path.exists(os.path.join(subdir, '.nobib')):
continue # Skip blacklisted folders
reldir = os.path.relpath(subdir, os.path.abspath(folder))
bib_path = os.path.join(subdir, 'biblio.bib')
subdb = utils.read_bib_file(bib_path)
for entry in subdb.entries:
filename = utils.decode_filename_field(entry['file'])
filename = os.path.join(reldir, filename)
entry['file'] = utils.encode_filename_field(filename)
db.entries += subdb.entries
# Remove duplicated entries
entries_dict = db.entries_dict
db.entries = [val for key, val in entries_dict.items()]
# Write result
bib_path = os.path.join(folder, 'master.bib')
utils.write_with_backup(bib_path, utils.write_bib(db, order=True), use_backup)
def merge_folder_tree(folder, use_backup):
"""
Merge bib files from the current subtree into a master bib file at the root.
This function updates the 'file' link of each entry with the relative path
to each subfolder that has been processed.
Args:
folder (str): relative or absolute path of the folder to process.
Returns:
Nothing, but creates a file named `master.bib` in the given folder.
"""
db = BibDatabase()
for subdir, _dirs, _files in os.walk(os.path.abspath(folder)):
if os.path.exists(os.path.join(subdir, '.nobib')):
continue # Skip blacklisted folders
reldir = os.path.relpath(subdir, os.path.abspath(folder))
bib_path = os.path.join(subdir, 'biblio.bib')
subdb = utils.read_bib_file(bib_path)
for entry in subdb.entries:
filename = utils.decode_filename_field(entry['file'])
filename = os.path.join(reldir, filename)
entry['file'] = utils.encode_filename_field(filename)
db.entries += subdb.entries
# Remove duplicated entries
entries_dict = db.entries_dict
db.entries = [val for key, val in entries_dict.items()]
# Write result
bib_path = os.path.join(folder, 'master.bib')
utils.write_with_backup(bib_path, utils.write_bib(db, order=True),
use_backup)
def rename_folder(folder, use_backup):
"""
Rename the pdf files in the given folder according to the information found
in `biblio.bib`. Note that this function will update file entries in
`biblio.bib`, but also in `.queried.bib`.
Args:
folder (str): absolute or relative path to the folder to process.
use_backup (bool): whether to backup previous files before writing.
Returns:
Nothing, but renames the pdfs in the given folder, and update bib files.
"""
# Read input bib files
pretty_bib_path = os.path.join(folder, 'biblio.bib')
pretty_db = utils.read_bib_file(pretty_bib_path)
queried_bib_path = os.path.join(folder, '.queried.bib')
queried_db = utils.read_bib_file(queried_bib_path)
queried_files = utils.create_file_dict(queried_db)
# Iterate over db entries
need_rename = False
for entry in pretty_db.entries:
old_filename = utils.decode_filename_field(entry['file'])
new_filename = nomenclature.gen_filename(entry)
if not os.path.exists(os.path.join(folder, old_filename)):
print(termcolor.colored('file not found: ', 'red') + old_filename)
elif old_filename != new_filename:
need_rename = True
print(termcolor.colored('-', 'red') + old_filename)
print(termcolor.colored('+', 'green') + new_filename)
# Skip if nothing to rename
if not need_rename:
return
# Ask confirmation
cmd = input('(Y/n) ')
if cmd == '' or cmd == 'y' or cmd == 'Y':
for entry in pretty_db.entries:
old_filename = utils.decode_filename_field(entry['file'])
new_filename = nomenclature.gen_filename(entry)
old_path = os.path.join(folder, old_filename)
new_path = os.path.join(folder, new_filename)
if os.path.exists(old_path):
os.rename(old_path, new_path)
new_val = utils.encode_filename_field(new_filename)
if old_filename in queried_files:
idx = queried_files[old_filename]
queried_db.entries[idx]['file'] = new_val
entry['file'] = new_val
# Write output bibtex files
utils.write_with_backup(pretty_bib_path,
utils.write_bib(pretty_db, order=False),
use_backup)
utils.write_with_backup(queried_bib_path,
utils.write_bib(queried_db, order=False),
use_backup)
def rename_folder(folder, use_backup):
"""
Rename the pdf files in the given folder according to the information found
in `biblio.bib`. Note that this function will update file entries in
`biblio.bib`, but also in `.queried.bib`.
Args:
folder (str): absolute or relative path to the folder to process.
use_backup (bool): whether to backup previous files before writing.
Returns:
Nothing, but renames the pdfs in the given folder, and update bib files.
"""
# Read input bib files
pretty_bib_path = os.path.join(folder, 'biblio.bib')
pretty_db = utils.read_bib_file(pretty_bib_path)
queried_bib_path = os.path.join(folder, '.queried.bib')
queried_db = utils.read_bib_file(queried_bib_path)
queried_files = utils.create_file_dict(queried_db)
# Iterate over db entries
need_rename = False
for entry in pretty_db.entries:
old_filename = utils.decode_filename_field(entry['file'])
new_filename = nomenclature.gen_filename(entry)
if not os.path.exists(os.path.join(folder, old_filename)):
print(termcolor.colored('file not found: ', 'red') + old_filename)
elif old_filename != new_filename:
need_rename = True
print(termcolor.colored('-', 'red') + old_filename)
print(termcolor.colored('+', 'green') + new_filename)
# Skip if nothing to rename
if not need_rename:
return
# Ask confirmation
cmd = input('(Y/n) ')
if cmd == '' or cmd == 'y' or cmd == 'Y':
for entry in pretty_db.entries:
old_filename = utils.decode_filename_field(entry['file'])
new_filename = nomenclature.gen_filename(entry)
old_path = os.path.join(folder, old_filename)
new_path = os.path.join(folder, new_filename)
if os.path.exists(old_path):
os.rename(old_path, new_path)
new_val = utils.encode_filename_field(new_filename)
if old_filename in queried_files:
idx = queried_files[old_filename]
queried_db.entries[idx]['file'] = new_val
entry['file'] = new_val
# Write output bibtex files
utils.write_with_backup(pretty_bib_path, utils.write_bib(pretty_db, order=False), use_backup)
utils.write_with_backup(queried_bib_path, utils.write_bib(queried_db, order=False), use_backup)
def query_crossref_folder(folder, use_backup):
"""
Query metadata information for unmatched pdf files in the given folder.
This function only queries Crossref.
Args:
folder (str): absolute or relative path to the folder to process.
use_backup (bool): whether to backup previous files before writing.
Returns:
Nothing, but writes the queried databases in bibtex format in the given
folder (and backup previous database if it differed).
"""
# Create database
db = utils.read_bib_file(os.path.join(folder, '.queried.bib'))
files = utils.guess_manual_files(folder, db, update_queried_db=False)
utils.add_skip_files(folder, files)
json_entries = []
rejected = []
# For each pdf in the folder
import pdb
pdb.set_trace()
for path in utils.get_pdf_list(folder):
file = os.path.basename(path)
parsed = nomenclature.parse_filename(file)
if parsed is None or file in files:
continue
print('Q: ' + os.path.basename(file))
authors, title = parsed
# Crossref
rbib, rjson, score = providers.crossref_query(authors, title)
if score >= config.crossref_accept_threshold:
# Append filename and store entry
rbib['file'] = utils.encode_filename_field(file)
json_entries.append(rjson)
db.entries.append(rbib)
else:
rejected.append(os.path.basename(file))
# Store results
bib_path = os.path.join(folder, '.queried.bib')
utils.write_with_backup(bib_path, utils.write_bib(db, order=False),
use_backup)
json_path = os.path.join(folder, '.queried.json')
json_str = json.dumps(json_entries,
sort_keys=True,
indent=4,
separators=(',', ': '))
utils.write_with_backup(json_path, json_str, use_backup)
rejected_path = os.path.join(folder, '.rejected.txt')
if len(rejected) > 0:
utils.write_with_backup(rejected_path, '\n'.join(rejected), use_backup)
def query_crossref_folder(folder, use_backup):
"""
Query metadata information for unmatched pdf files in the given folder.
This function only queries Crossref.
Args:
folder (str): absolute or relative path to the folder to process.
use_backup (bool): whether to backup previous files before writing.
Returns:
Nothing, but writes the queried databases in bibtex format in the given
folder (and backup previous database if it differed).
"""
# Create database
db = utils.read_bib_file(os.path.join(folder, '.queried.bib'))
files = utils.guess_manual_files(folder, db, update_queried_db=False)
utils.add_skip_files(folder, files)
json_entries = []
rejected = []
# For each pdf in the folder
for path in utils.get_pdf_list(folder):
file = os.path.basename(path)
parsed = nomenclature.parse_filename(file)
if parsed is None or file in files:
continue
print('Q: ' + os.path.basename(file))
authors, title = parsed
# Crossref
rbib, rjson, score = providers.crossref_query(authors, title)
if score >= config.crossref_accept_threshold:
# Append filename and store entry
rbib['file'] = utils.encode_filename_field(file)
json_entries.append(rjson)
db.entries.append(rbib)
else:
rejected.append(os.path.basename(file))
# Store results
bib_path = os.path.join(folder, '.queried.bib')
utils.write_with_backup(bib_path, utils.write_bib(db, order=False), use_backup)
json_path = os.path.join(folder, '.queried.json')
json_str = json.dumps(json_entries, sort_keys=True, indent=4, separators=(',', ': '))
utils.write_with_backup(json_path, json_str, use_backup)
rejected_path = os.path.join(folder, '.rejected.txt')
if len(rejected) > 0:
utils.write_with_backup(rejected_path, '\n'.join(rejected), use_backup)
def sync_folder(folder, use_backup):
"""
Update the file field of bibtex entries for the given folder.
When an entry could not find a good match, it will be removed from the
bibtex, unless the user explicitly prevents it.
Args:
folder (str): absolute or relative path to the folder to process.
use_backup (bool): whether to backup previous files before writing.
Returns:
Nothing, but updates `.queried.bib` and `biblio.bib` files.
"""
for bib_file in ('.queried.bib', 'biblio.bib'):
bib_path = os.path.join(folder, bib_file)
db = utils.read_bib_file(bib_path)
unmatched = set(
[os.path.basename(f) for f in utils.get_pdf_list(folder)])
to_delete = []
for i, entry in enumerate(db.entries):
guess = nomenclature.gen_filename(entry)
if 'file' in entry:
guess = utils.decode_filename_field(entry['file'])
match, score = utils.most_similar_filename(guess, unmatched)
if score >= 0.90:
unmatched.remove(match)
entry['file'] = utils.encode_filename_field(match)
else:
print(
termcolor.colored(bib_file, "magenta") +
": ({1}) will remove '{0}'".format(
guess, termcolor.colored(score, "yellow")))
to_delete.append(i)
# Delete unmatched entries
if to_delete:
cmd = input('(Y/n) ')
if cmd == '' or cmd == 'y' or cmd == 'Y':
for i in sorted(to_delete, reverse=True):
del db.entries[i]
# Write synced database
utils.write_with_backup(bib_path, utils.write_bib(db, order=False),
use_backup)
def sync_folder(folder, use_backup):
"""
Update the file field of bibtex entries for the given folder.
When an entry could not find a good match, it will be removed from the
bibtex, unless the user explicitly prevents it.
Args:
folder (str): absolute or relative path to the folder to process.
use_backup (bool): whether to backup previous files before writing.
Returns:
Nothing, but updates `.queried.bib` and `biblio.bib` files.
"""
for bib_file in ('.queried.bib', 'biblio.bib'):
bib_path = os.path.join(folder, bib_file)
db = utils.read_bib_file(bib_path)
unmatched = set([os.path.basename(f) for f in utils.get_pdf_list(folder)])
to_delete = []
for i, entry in enumerate(db.entries):
guess = nomenclature.gen_filename(entry)
if 'file' in entry:
guess = utils.decode_filename_field(entry['file'])
match, score = utils.most_similar_filename(guess, unmatched)
if score >= 0.90:
unmatched.remove(match)
entry['file'] = utils.encode_filename_field(match)
else:
print(termcolor.colored(bib_file, "magenta") +
": ({1}) will remove '{0}'".format(guess, termcolor.colored(score, "yellow")))
to_delete.append(i)
# Delete unmatched entries
if to_delete:
cmd = input('(Y/n) ')
if cmd == '' or cmd == 'y' or cmd == 'Y':
for i in sorted(to_delete, reverse=True):
del db.entries[i]
# Write synced database
utils.write_with_backup(bib_path, utils.write_bib(db, order=False), use_backup)
def query_google_folder(folder, use_backup):
"""
Query metadata information for unmatched pdf files in the given folder.
This function only queries Google Scholar.
Args:
folder (str): absolute or relative path to the folder to process.
use_backup (bool): whether to backup previous files before writing.
Returns:
Nothing, but writes the queried databases in bibtex format in the given
folder (and backup previous database if it differed).
"""
# Create database
db = utils.read_bib_file(os.path.join(folder, '.queried.bib'))
files = utils.guess_manual_files(folder, db, update_queried_db=False)
utils.add_skip_files(folder, files)
for path in utils.get_pdf_list(folder):
file = os.path.basename(path)
parsed = nomenclature.parse_filename(file)
if parsed is None or file in files:
continue
print('Q: ' + os.path.basename(file))
authors, title = parsed
# Google Scholar
rbib = providers.scholarly_query(authors, title)
if rbib is None:
continue
# Append filename and store entry
rbib['file'] = utils.encode_filename_field(file)
db.entries.append(rbib)
# Store results
bib_path = os.path.join(folder, '.queried.bib')
utils.write_with_backup(bib_path, utils.write_bib(db, order=False),
use_backup)
def query_google_folder(folder, use_backup):
"""
Query metadata information for unmatched pdf files in the given folder.
This function only queries Google Scholar.
Args:
folder (str): absolute or relative path to the folder to process.
use_backup (bool): whether to backup previous files before writing.
Returns:
Nothing, but writes the queried databases in bibtex format in the given
folder (and backup previous database if it differed).
"""
# Create database
db = utils.read_bib_file(os.path.join(folder, '.queried.bib'))
files = utils.guess_manual_files(folder, db, update_queried_db=False)
utils.add_skip_files(folder, files)
for path in utils.get_pdf_list(folder):
file = os.path.basename(path)
parsed = nomenclature.parse_filename(file)
if parsed is None or file in files:
continue
print('Q: ' + os.path.basename(file))
authors, title = parsed
# Google Scholar
rbib = providers.scholarly_query(authors, title)
if rbib is None:
continue
# Append filename and store entry
rbib['file'] = utils.encode_filename_field(file)
db.entries.append(rbib)
# Store results
bib_path = os.path.join(folder, '.queried.bib')
utils.write_with_backup(bib_path, utils.write_bib(db, order=False), use_backup)