def parse_bibfile(bibfilename: str, encoding: str) -> "BibliographyData":
"""Parse *bibfilename* with given *encoding*, and return parsed data."""
parser = Parser(encoding)
logger.info("parsing bibtex file {0}... ".format(bibfilename), nonl=True)
parser.parse_file(bibfilename)
logger.info("parsed {0} entries".format(len(parser.data.entries)))
return parser.data
def references_to_markdown(references):
"""Utility function to convert a BibTeX string containing
references into a Markdown string.
Args:
references: BibTeX string
Returns:
Markdown string
"""
pybtex_style = find_plugin('pybtex.style.formatting', 'plain')()
pybtex_md_backend = find_plugin('pybtex.backends', 'markdown')
pybtex_parser = Parser()
# hack to not print labels (may remove this later)
def write_entry(self, key, label, text):
self.output(u'%s \n' % text)
pybtex_md_backend.write_entry = write_entry
pybtex_md_backend = pybtex_md_backend()
data = pybtex_parser.parse_stream(StringIO(references))
data_formatted = pybtex_style.format_entries(data.entries.itervalues())
output = StringIO()
pybtex_md_backend.write_to_stream(data_formatted, output)
# add blockquote style
references_md = '> {}'.format(output.getvalue())
references_md.replace('\n', '\n> ')
return references_md
def run(self):
style = find_plugin('pybtex.style.formatting', self.options.get('style', 'unsrt'))()
parser = Parser()
# Sort the publication entries by year reversed
data = sorted(parser.parse_file(self.arguments[0]).entries.items(),
key=lambda e: e[1].fields['year'], reverse=True)
_, entries = zip(*data)
html = '<div class = "publication-list">\n'
cur_year = None
for entry in entries:
# print a year title when year changes
if entry.fields['year'] != cur_year:
if cur_year is not None: # not first year group
html += '</ul>'
cur_year = entry.fields['year']
html += '<h3>{}</h3>\n<ul>'.format(cur_year)
html += '<li class = "publication">{}</li>'.format(
list(style.format_entries((entry,)))[0].text.render_as('html'))
if len(entries) != 0: # publication list is nonempty
html += '</ul>'
html += '</div>'
return [nodes.raw('', html, format='html'), ]
def parse_bibtex(app):
# load bibtex path
if not app.config.bibtex_path:
raise SphinxWarning('No BibTeX path specified.')
if os.path.isabs(app.config.bibtex_path):
filepath = app.config.bibtex_path
else:
filepath = os.path.join(app.confdir, app.config.bibtex_path)
if not os.path.exists(filepath):
raise SphinxWarning("BibTeX file (%s) does not exists." % filepath)
app.env.bibtex_path = filepath
# parse bib
parser = BibtexParser()
try:
data = parser.parse_file(filepath)
except PybtexError as ex:
raise SphinxWarning(ex)
app.env.bibtex_entries = data.entries
# latex reconfiguration
filename = os.path.splitext(os.path.basename(filepath))[0]
app.config.latex_elements.setdefault('footer', '')
app.config.latex_elements['footer'] += "\\bibliography{%s}" % filename
def parse_bibtex(app):
# load bibtex path
if not app.config.bibtex_path:
raise SphinxWarning('No BibTeX path specified.')
if os.path.isabs(app.config.bibtex_path):
filepath = app.config.bibtex_path
else:
filepath = os.path.join(app.confdir, app.config.bibtex_path)
if not os.path.exists(filepath):
raise SphinxWarning("BibTeX file (%s) does not exists." % filepath)
app.env.bibtex_path = filepath
# parse bib
parser = BibtexParser()
try:
data = parser.parse_file(filepath)
except PybtexError as ex:
raise SphinxWarning(ex)
app.env.bibtex_entries = data.entries
# latex reconfiguration
filename = os.path.splitext(os.path.basename(filepath))[0]
app.config.latex_elements.setdefault('footer', '')
app.config.latex_elements['footer'] += "\\bibliography{%s}" % filename
def load_citekey(citekey, bibfile=BIBFILE):
""" Get the info corresponding to specific citekey
from bibfile.
"""
parser = Parser()
bib_data = parser.parse_file(bibfile)
entry = bib_data.entries[citekey]
return entry
def load_citekey(citekey, bibfile=BIBFILE):
""" Get the info corresponding to specific citekey
from bibfile.
"""
parser = Parser()
bib_data = parser.parse_file(bibfile)
entry = bib_data.entries[citekey]
return entry
def run(self):
style = find_plugin('pybtex.style.formatting', self.options.get('style', 'unsrt'))()
bibtex_dir = self.options.get('bibtex_dir', 'bibtex')
highlight_author = self.options.get('highlight_author', None)
parser = Parser()
# Sort the publication entries by year reversed
data = sorted(parser.parse_file(self.arguments[0]).entries.items(),
key=lambda e: e[1].fields['year'], reverse=True)
print(type(data))
html = '<div class = "publication-list">\n'
cur_year = None
if bibtex_dir: # create the bibtex dir if the option is set
try:
os.mkdir(os.path.sep.join((self.output_folder, bibtex_dir)))
except OSError: # probably because the dir already exists
pass
for label, entry in data:
# print a year title when year changes
if entry.fields['year'] != cur_year:
if cur_year is not None: # not first year group
html += '</ul>'
cur_year = entry.fields['year']
html += '<h3>{}</h3>\n<ul>'.format(cur_year)
pub_html = list(style.format_entries((entry,)))[0].text.render_as('html')
if highlight_author: # highlight an author (usually oneself)
pub_html = pub_html.replace(highlight_author,
'<strong>{}</strong>'.format(highlight_author), 1)
html += '<li class = "publication">' + pub_html
extra_links = ""
if bibtex_dir: # write bib files to bibtex_dir for downloading
bib_link = '{}/{}.bib'.format(bibtex_dir, label)
bib_data = BibliographyData(dict({label: entry}))
bib_data.to_file('/'.join([self.output_folder, bib_link]), 'bibtex')
extra_links += '[<a href="{}">bibtex</a>] '.format(bib_link)
if 'pdf' in entry.fields: # the link to the pdf file
extra_links += '[<a href="{}">pdf</a>] '.format(entry.fields['pdf'])
if extra_links:
html += '<br/>' + extra_links
html += '</li>'
if len(data) != 0: # publication list is nonempty
html += '</ul>'
html += '</div>'
return [nodes.raw('', html, format='html'), ]
def setUp(self):
unittest.TestCase.setUp(self)
curdir = os.path.dirname(os.path.abspath(__file__))
parser = BibtexParser()
self.entries = \
parser.parse_file(os.path.join(curdir, 'testdata', 'bibtex.bib')).entries
self.style = Style(os.path.join(curdir, 'styles'), 'apa')
def doi2Entry(doi):
"""Returns a pybtex.database.BibliographyData from a doi"""
bib_parser = Parser()
bibtext = doi2bibtex(doi)
if bibtext != None:
bibtext = bibtext.decode('utf8')
bib_data = bib_parser.parse_stream(StringIO(bibtext))
return bib_data
else:
return None
def add_citations(generators):
global global_bib
global content_path
if not pyb_imported:
logger.warn('`pelican-cite` failed to load dependency `pybtex`')
return
if 'PUBLICATIONS_SRC' in generators[0].settings:
refs_file = generators[0].settings['PUBLICATIONS_SRC']
refs_file = os.path.join(content_path, refs_file)
try:
global_bib = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' %
(refs_file, str(e)))
if 'PATH' in generators[0].settings:
content_path = generators[0].settings['PATH']
else:
logger.warn('`pelican-cite` failed to obtain content path')
# Process the articles and pages
for generator in generators:
if isinstance(generator, ArticlesGenerator):
for article in generator.articles:
process_content(article)
elif isinstance(generator, PagesGenerator):
for page in generator.pages:
process_content(page)
def make_bibliography(aux_filename,
bib_format=None,
bib_encoding=None,
output_encoding=None,
bst_encoding=None,
min_crossrefs=2,
**kwargs):
from os import path
import pybtex.io
from pybtex.bibtex import bst
from pybtex.bibtex.interpreter import Interpreter
from pybtex import auxfile
if bib_format is None:
from pybtex.database.input.bibtex import Parser as bib_format
aux_data = auxfile.parse_file(aux_filename, output_encoding)
bst_filename = aux_data.style + path.extsep + 'bst'
bst_script = bst.parse_file(bst_filename, bst_encoding)
base_filename = path.splitext(aux_filename)[0]
bbl_filename = base_filename + path.extsep + 'bbl'
bib_filenames = [
filename + bib_format.get_default_suffix()
for filename in aux_data.data
]
bbl_file = pybtex.io.open_unicode(bbl_filename,
'w',
encoding=output_encoding)
interpreter = Interpreter(bib_format, bib_encoding)
interpreter.run(bst_script,
aux_data.citations,
bib_filenames,
bbl_file,
min_crossrefs=min_crossrefs)
def make_bibliography(aux_filename,
bib_format=None,
bib_encoding=None,
output_encoding=None,
bst_encoding=None,
min_crossrefs=2,
**kwargs
):
from os import path
import pybtex.io
from pybtex.bibtex import bst
from pybtex.bibtex.interpreter import Interpreter
from pybtex import auxfile
if bib_format is None:
from pybtex.database.input.bibtex import Parser as bib_format
aux_data = auxfile.parse_file(aux_filename, output_encoding)
bst_filename = aux_data.style + path.extsep + 'bst'
bst_script = bst.parse_file(bst_filename, bst_encoding)
base_filename = path.splitext(aux_filename)[0]
bbl_filename = base_filename + path.extsep + 'bbl'
bib_filenames = [filename + bib_format.get_default_suffix() for filename in aux_data.data]
bbl_file = pybtex.io.open_unicode(bbl_filename, 'w', encoding=output_encoding)
interpreter = Interpreter(bib_format, bib_encoding)
interpreter.run(bst_script, aux_data.citations, bib_filenames, bbl_file, min_crossrefs=min_crossrefs)
def add_citations(generators):
global global_bib
if not pyb_imported:
logger.warn("`pelican-cite` failed to load dependency `pybtex`")
return
if "PUBLICATIONS_SRC" in generators[0].settings:
refs_file = generators[0].settings["PUBLICATIONS_SRC"]
try:
global_bib = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warn(
"`pelican_bibtex` failed to parse file %s: %s" % (refs_file, str(e))
)
# Process the articles and pages
for generator in generators:
if isinstance(generator, ArticlesGenerator):
for article in (
generator.articles + generator.translations + generator.drafts
):
process_content(article)
elif isinstance(generator, PagesGenerator):
for page in generator.pages:
process_content(page)
def groomBib(bibfile, groomedfile=""):
"""Grooms a bib file according to the present doi or to a doi found in internet. Returns a filename.verified.bib and filename.unverified.bib"""
if groomedfile == "":
groomedfile = bibfile + ".groomed.bib"
with open(bibfile,"r") as f:
bstr = f.read()
# Assume the input is utf8 encoded
bstr = bstr.decode('utf8')
# Parse the bibtex file
bib_parser = Parser()
bib_data = bib_parser.parse_stream(StringIO(bstr))
#first I groom the original bib file:
bib_cleandata=cleanBibliographyData(bib_data)
listofpersons =[]
if bib_cleandata != None:
for (key,entry) in bib_cleandata.entries.items():
if _journal_field in entry.fields:
journal = entry.fields[_journal_field]
journaltitlecase = titlecase.titlecase(journal)
if journal != journaltitlecase:
entry.fields[_journal_field] = changeThisforThat(journal,journaltitlecase ,"Change for %s." % _journal_field)
#check inconsistencies in names
for persontype in entry.persons: #authors for example
for ip in range(0,len( entry.persons[persontype])):
lastname = entry.persons[persontype][ip].get_part_as_text("last")
firstname = entry.persons[persontype][ip].get_part_as_text("first")
middle=entry.persons[persontype][ip].get_part_as_text("middle")
prelast=entry.persons[persontype][ip].get_part_as_text("prelast")
lineage= entry.persons[persontype][ip].get_part_as_text("lineage")
result = [element for element in listofpersons #if they have the same lastname and different names tell
if element[0] == lastname and (firstname !=element[1] or middle !=element[2] or prelast != element[3] or lineage != element[4] )]
if len(result) >0: #warns regarding conflict with names
for possibleresults in result:
oneversion = Person(last =lastname,first=firstname , middle=middle,prelast=prelast,lineage=lineage)
otherversion = Person(last =possibleresults[0],first=possibleresults[1] , middle=possibleresults[2],prelast=possibleresults[3],lineage=possibleresults[4])
warnings.warn("Possible problem with authors: %s vs. %s" % (oneversion,otherversion ))
listofpersons = listofpersons + [[lastname, firstname, middle,prelast,lineage]]
savebib(bib_cleandata, groomedfile)
return "Done"
def remfieldsBib(bibfile, cleanfile=""):
"""Grooms a bib file according to the present doi or to a doi found in internet. Returns a filename.verified.bib and filename.unverified.bib"""
if cleanfile == "":
cleanfile = bibfile + ".clean.bib"
with open(bibfile,"r") as f:
bstr = f.read()
# Assume the input is utf8 encoded
bstr = bstr.decode('utf8')
# Parse the bibtex file
bib_parser = Parser()
bib_data = bib_parser.parse_stream(StringIO(bstr))
for (key,entry) in bib_data.entries.items():
for field_name in _unwanted_fields:
if field_name in entry.fields:
del entry.fields[field_name]
savebib(bib_data, cleanfile)
return "Done"
def run(self):
refs_file = self.arguments[0].strip()
try:
bibdata_all = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' %
(refs_file, str(e)))
return
# format entries
jo_style = jo.Style()
jo_style.strong = 'Razik'
html_backend = html.Backend()
formatted_entries = jo_style.format_entries(
bibdata_all.entries.values())
publications = []
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
year = entry.fields.get('year')
# This shouldn't really stay in the field dict
# but new versions of pybtex don't support pop
pdf = entry.fields.get('pdf', None)
slides = entry.fields.get('slides', None)
poster = entry.fields.get('poster', None)
# render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
text = formatted_entry.text.render(html_backend)
publications.append(
(key, year, text, bib_buf.getvalue(), pdf, slides, poster))
# Load the publications template
if 'template' in self.options:
template_path = self.options['template']
template_dir, template_name = os.path.split(template_path)
env = Environment(loader=FileSystemLoader(template_dir))
template = env.get_template(template_name)
else:
# Use template from the Pelican theme
template = pelican_generator.get_template('publications')
rendered_template = template.render(publications=publications)
return [nodes.raw('', rendered_template, format='html')]
def get_bib_file(article):
"""
If a bibliography file is specified for this article/page, parse
it and return the parsed object.
"""
if 'publications_src' in article.metadata:
refs_file = article.metadata['publications_src']
try:
local_bib = Parser().parse_file(refs_file)
return local_bib
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' %
(refs_file, str(e)))
return global_bib
else:
return global_bib
def init(pelican_instance):
global global_bib, bibliography_start, bibliography_end
if not pyb_imported:
logger.warning('`pelican-cite` failed to load dependency `pybtex`')
return
if 'BIBLIOGRAPHY_START' in pelican_instance.settings:
bibliography_start = pelican_instance.settings['BIBLIOGRAPHY_START']
if 'BIBLIOGRAPHY_END' in pelican_instance.settings:
bibliography_end = pelican_instance.settings['BIBLIOGRAPHY_END']
if 'PUBLICATIONS_SRC' in pelican_instance.settings:
refs_file = pelican_instance.settings['PUBLICATIONS_SRC']
try:
global_bib = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warning('`pelican_bibtex` failed to parse file %s: %s' %
(refs_file, str(e)))
def get_bib_file(article):
"""
If a bibliography file is specified for this article/page, parse
it and return the parsed object.
"""
if 'publications_src' in article.metadata:
refs_file = article.metadata['publications_src']
try:
logger.warning(
f"Looking for bib file at {os.path.dirname(article.source_path) + os.path.sep + refs_file}"
)
local_bib = Parser().parse_file(
os.path.dirname(article.source_path) + os.path.sep + refs_file)
return local_bib
except PybtexError as e:
logger.warning('`pelican_bibtex` failed to parse file %s: %s' %
(refs_file, str(e)))
return global_bib
else:
return global_bib
return fields[x]
else:
return ""
BIBFILE = 'cs.bib'
jfile = open('journal.text', 'w')
cfile = open('conf.text', 'w')
cfile_selected = open('conf_selected.text', 'w')
cfile_selected2 = open('conf_selected2.text', 'w')
cfile_other = open('conf_other.text', 'w')
parser = Parser()
bib_data = parser.parse_file(BIBFILE)
bib_sorted = sorted(bib_data.entries.items(), key=cmp_to_key(sort_list_cmp))
jnumber = 0
jyear = 9999
print_jyear = 0
cnumber = 0
cyear = 9999
print_cyear = 0
cnumber_select = 0
cyear_select = 9999
print_cyear_select = 0
def meta_str_format(val,meta):
return val
META_FORMATS = {
'dir':meta_dir_format,
'jinja':meta_jinja_format,
'csv':meta_csv_format,
'jsonfile':meta_jsonfile_format,
'json':meta_json_format,
'str':meta_str_format
}
try:
from pybtex.database.input.bibtex import Parser as BibParser
P = BibParser()
P.macros['true']=True
CONVERT_KEYS=['title','pages']
def fromBTeX(t):
ret = t.replace('{','').replace('}','').replace("\\'\\i",u'í').replace('\\vc',u'č').replace('---','—').replace('--','–')
return ret.replace('\\textendash','–')
def meta_bib_format(val,meta):
bib = P.parse_file(val)
ret = []
for entry in bib.entries.values():
pub = {'type':entry.type}
pub.update(entry.fields)
authors = entry.persons.get('author',[])
editors = entry.persons.get('editor',[])
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS_SRC']:
Local path to the BibTeX file to read.
generator.settings['PUBLICATIONS_SPLIT_BY']:
The name of the bibtex field used for splitting the publications.
No splitting if title is not provided.
generator.settings['PUBLICATIONS_UNTAGGED_TITLE']:
The title of the header for all untagged entries.
No such list if title is not provided.
Output
------
generator.context['publications_lists']:
A map with keys retrieved from the field named in PUBLICATIONS_SPLIT_TAG.
Values are lists of tuples (key, year, text, bibtex, pdf, slides, poster)
See Readme.md for more details.
generator.context['publications']:
Contains all publications as a list of tuples
(key, year, text, bibtex, pdf, slides, poster).
See Readme.md for more details.
"""
if 'PUBLICATIONS_SRC' not in generator.settings:
return
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
from pybtex.style.formatting import plain
except ImportError:
logger.warn('`pelican_bib` failed to load dependency `pybtex`')
return
refs_file = generator.settings['PUBLICATIONS_SRC']
try:
bibdata_all = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warn('`pelican_bib` failed to parse file %s: %s' %
(refs_file, str(e)))
return
publications = []
publications_lists = {}
publications_untagged = []
split_by = None
untagged_title = None
if 'PUBLICATIONS_SPLIT_BY' in generator.settings:
split_by = generator.settings['PUBLICATIONS_SPLIT_BY']
if 'PUBLICATIONS_UNTAGGED_TITLE' in generator.settings:
untagged_title = generator.settings['PUBLICATIONS_UNTAGGED_TITLE']
# format entries
plain_style = plain.Style()
html_backend = html.Backend()
formatted_entries = plain_style.format_entries(
bibdata_all.entries.values())
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
year = entry.fields.get('year')
# This shouldn't really stay in the field dict
# but new versions of pybtex don't support pop
pdf = entry.fields.get('pdf', None)
slides = entry.fields.get('slides', None)
poster = entry.fields.get('poster', None)
tags = []
if split_by:
tags = entry.fields.get(split_by, [])
# parse to list, and trim each string
if tags:
tags = [tag.strip() for tag in tags.split(',')]
# create keys in publications_lists if at least one
# tag is given
for tag in tags:
publications_lists[tag] = publications_lists.get(tag, [])
#render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
text = formatted_entry.text.render(html_backend)
entry_tuple = {
'key': key,
'year': year,
'text': text,
'bibtex': bib_buf.getvalue(),
'pdf': pdf,
'slides': slides,
'poster': poster
}
publications.append(entry_tuple)
for tag in tags:
publications_lists[tag].append(entry_tuple)
if not tags and untagged_title:
publications_untagged.append(entry_tuple)
# append untagged list if title is given
if untagged_title and publications_untagged:
publications_lists[untagged_title] = publications_untagged
# output
generator.context['publications'] = publications
generator.context['publications_lists'] = publications_lists
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS_SRC']:
local path to the BibTeX file to read.
Output
------
generator.context['publications']:
List of tuples (key, year, text, bibtex, pdf, slides, poster).
See Readme.md for more details.
"""
if 'PUBLICATIONS_SRC' not in generator.settings:
return
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
from pybtex.style.formatting import plain
except ImportError:
logger.warn('`pelican_bibtex` failed to load dependency `pybtex`')
return
refs_file = generator.settings['PUBLICATIONS_SRC']
try:
bibdata_all = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' % (
refs_file,
str(e)))
return
publications = []
# format entries
plain_style = plain.Style()
html_backend = html.Backend()
formatted_entries = plain_style.format_entries(bibdata_all.entries.values())
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
year = entry.fields.get('year')
# This shouldn't really stay in the field dict
# but new versions of pybtex don't support pop
pdf = entry.fields.get('pdf', None)
slides = entry.fields.get('slides', None)
poster = entry.fields.get('poster', None)
#render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
text = formatted_entry.text.render(html_backend)
publications.append((key,
year,
text,
bib_buf.getvalue(),
pdf,
slides,
poster))
generator.context['publications'] = publications
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS_SRC']:
local path to the BibTeX file to read.
Output
------
generator.context['publications']:
List of tuples (key, year, text, bibtex, pdf, slides, poster).
See Readme.md for more details.
"""
if 'PUBLICATIONS_SRC' not in generator.settings:
return
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
from pybtex.style.formatting import plain, toplevel
from pybtex.style.template import (sentence, words,
optional, optional_field, field, tag)
from pybtex.richtext import Symbol
except ImportError:
logger.warn('`pelican_bibtex` failed to load dependency `pybtex`')
return
refs_file = generator.settings['PUBLICATIONS_SRC']
try:
bibdata_all = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warning('`pelican_bibtex` failed to parse file %s: %s' % (
refs_file,
str(e)))
return
class CustomStyle(plain.Style):
def format_bold_title(self, e, which_field, as_sentence=True):
formatted_title = tag('strong')[field(which_field)]
if as_sentence:
return sentence[formatted_title]
else:
return formatted_title
def get_inproceedings_template(self, e):
template = toplevel[
self.format_bold_title(e, 'title'),
Symbol('newline'),
sentence[self.format_names('author')],
Symbol('newline'),
words[
'In',
sentence[
optional[self.format_editor(e, as_sentence=False)],
self.format_btitle(e, 'booktitle', as_sentence=False),
self.format_volume_and_series(e, as_sentence=False),
],
self.format_address_organization_publisher_date(e),
],
sentence[optional_field('note')],
self.format_web_refs(e),
]
return template
def get_article_template(self, e):
volume_and_pages = first_of[
# volume and pages, with optional issue number
optional[
join[
field('volume'),
optional['(', field('number'), ')'],
':', pages
],
],
# pages only
words['pages', pages],
]
template = toplevel[
self.format_bold_title(e, 'title'),
Symbol('newline'),
self.format_names('author'),
Symbol('newline'),
sentence[
tag('em')[field('journal')],
optional[volume_and_pages],
date],
sentence[optional_field('note')],
self.format_web_refs(e),
]
return template
def get_techreport_template(self, e):
template = toplevel[
self.format_bold_title(e, 'title'),
Symbol('newline'),
sentence[self.format_names('author')],
Symbol('newline'),
sentence[
words[
first_of[
optional_field('type'),
'Technical Report',
],
optional_field('number'),
],
field('institution'),
optional_field('address'),
date,
],
sentence[optional_field('note')],
self.format_web_refs(e),
]
return template
def format_entry(self, label, entry, bib_data=None):
return super().format_entry(label, entry, bib_data)
publications = []
# format entries
my_style = CustomStyle()
html_backend = html.Backend()
html_backend.symbols.update({'newline': '<br>'})
formatted_entries = my_style.format_entries(bibdata_all.entries.values())
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
year = entry.fields.get('year')
# This shouldn't really stay in the field dict
# but new versions of pybtex don't support pop
pdf = entry.fields.get('pdf', None)
slides = entry.fields.get('slides', None)
poster = entry.fields.get('poster', None)
entrytype = entry.fields.get('type', None)
# render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
text = formatted_entry.text.render(html_backend)
publications.append((key,
year,
text,
bib_buf.getvalue(),
pdf,
slides,
poster,
entrytype))
publications.sort(key=itemgetter(1), reverse=True)
generator.context['publications'] = publications
def run(self):
style = find_plugin('pybtex.style.formatting', self.options.get('style', 'unsrt'))()
bibtex_dir = self.options.get('bibtex_dir', 'bibtex')
detail_page_dir = self.options.get('detail_page_dir', 'papers')
highlight_author = self.options.get('highlight_author', None)
self.state.document.settings.record_dependencies.add(self.arguments[0])
parser = Parser()
# Sort the publication entries by year reversed
data = sorted(parser.parse_file(self.arguments[0]).entries.items(),
key=lambda e: e[1].fields['year'], reverse=True)
print(type(data))
html = '<div class = "publication-list">\n'
cur_year = None
if bibtex_dir: # create the bibtex dir if the option is set
try:
os.mkdir(os.path.sep.join((self.output_folder, bibtex_dir)))
except OSError: # probably because the dir already exists
pass
if detail_page_dir: # create the detail page dir if the option is set
try:
os.mkdir(os.path.sep.join((self.output_folder, detail_page_dir)))
except OSError: # probably because the dir already exists
pass
for label, entry in data:
# print a year title when year changes
if entry.fields['year'] != cur_year:
if cur_year is not None: # not first year group
html += '</ul>'
cur_year = entry.fields['year']
html += '<h3>{}</h3>\n<ul>'.format(cur_year)
pub_html = list(style.format_entries((entry,)))[0].text.render_as('html')
if highlight_author: # highlight an author (usually oneself)
pub_html = pub_html.replace(highlight_author,
'<strong>{}</strong>'.format(highlight_author), 1)
html += '<li class = "publication">' + pub_html
extra_links = ""
bib_data = BibliographyData(dict({label: entry})) # detail_page_dir may need it later
if bibtex_dir: # write bib files to bibtex_dir for downloading
bib_link = '{}/{}.bib'.format(bibtex_dir, label)
bib_data.to_file('/'.join([self.output_folder, bib_link]), 'bibtex')
extra_links += '[<a href="{}">BibTeX</a>] '.format(
self.site.config['BASE_URL'] + bib_link)
if 'fulltext' in entry.fields: # the link to the full text, usually a link to the pdf file
extra_links += '[<a href="{}">full text</a>] '.format(entry.fields['fulltext'])
if extra_links or detail_page_dir:
html += '<br>'
html += extra_links
if detail_page_dir: # render the details page of a paper
page_url = '/'.join((detail_page_dir, label + '.html'))
html += ' [<a href="{}">abstract and details</a>]'.format(
self.site.config['BASE_URL'] + page_url)
context = {
'title': process_bibtex_string(entry.fields['title']),
'abstract': process_bibtex_string(entry.fields['abstract']) if 'abstract' in entry.fields else '',
'bibtex': bib_data.to_string('bibtex'),
'bibtex_link': '/' + bib_link if bibtex_dir else '',
'default_lang': self.site.config['DEFAULT_LANG'],
'label': label,
'lang': self.site.config['DEFAULT_LANG'],
'permalink': self.site.config['SITE_URL'] + page_url,
'reference': pub_html,
'extra_links': extra_links
}
if 'fulltext' in entry.fields and entry.fields['fulltext'].endswith('.pdf'):
context['pdf'] = entry.fields['fulltext']
self.site.render_template(
'publication.tmpl',
os.path.sep.join((self.output_folder, detail_page_dir, label + '.html')),
context,
)
html += '</li>'
if len(data) != 0: # publication list is nonempty
html += '</ul>'
html += '</div>'
return [nodes.raw('', html, format='html'), ]
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS_SRC']:
local path to the BibTeX file to read.
Output
------
generator.context['publications']:
List of tuples (key, year, text, bibtex, pdf, slides, poster).
See Readme.md for more details.
"""
# check if settings are provided via pelicanconf.py
settings_present = False
for s in ['PUBLICATIONS_SRC', 'PRESENTATIONS_SRC', 'POSTERS_SRC']:
if s in generator.settings:
settings_present = True
if not settings_present:
return
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
from pybtex.style.formatting import plain
except ImportError:
logger.warn('`pelican_bibtex` failed to load dependency `pybtex`')
return
for s, c in zip(['PUBLICATIONS_SRC', 'PRESENTATIONS_SRC', 'POSTERS_SRC'],
['publications', 'presentations', 'posters']):
if s not in generator.settings:
continue
refs_file = generator.settings[s]
try:
bibdata_all = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' %
(refs_file, str(e)))
continue
publications = []
# format entries
plain_style = plain.Style()
html_backend = html.Backend()
all_entries = bibdata_all.entries.values()
# remove URL field if DOI is present
for entry in all_entries:
if "doi" in entry.fields.keys():
entry.fields._dict["url"] = ""
formatted_entries = plain_style.format_entries(all_entries)
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
year = entry.fields.get('year')
slides = entry.fields.pop('slides', None)
poster = entry.fields.pop('poster', None)
# add PDF link if file is present
# Zotero exports a 'file' field, which contains the 'Zotero' and
# 'Filesystem' filenames, seperated by ':'
try:
filename = entry.fields['file'].split(':')[0]
if os.access(os.path.join('content', 'download', filename),
os.R_OK):
pdf = os.path.join('download', filename)
else:
pdf = None
except KeyError:
pdf = None
#render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
text = formatted_entry.text.render(html_backend)
doi = (entry.fields.get('doi')
if 'doi' in entry.fields.keys() else "")
url = (entry.fields.get('url')
if 'url' in entry.fields.keys() else "")
# prettify entries
# remove BibTeX's {}
text = text.replace("\{", "")
text = text.replace("{", "")
text = text.replace("\}", "")
text = text.replace("}", "")
# subscript 2 in NO2, CO2, SO2
text = text.replace("NO2", "NO<sub>2</sub>")
text = text.replace("CO2", "CO<sub>2</sub>")
text = text.replace("CO2", "CO<sub>2</sub>")
# for posters and presentations, make for nicer printing
text = text.replace("In <em>", "Presented at <em>")
# remove empty URL link
text = text.replace("<a href=\"\">URL:</a>, ", "")
publications.append((key, year, text, bib_buf.getvalue(), doi, url,
pdf, slides, poster))
# store the list of artifacts in the generator context
generator.context[c] = publications
def run(self):
bibtex_dir = self.options.get('bibtex_dir', 'bibtex')
detail_page_dir = self.options.get('detail_page_dir', 'papers')
highlight_authors = self.options.get('highlight_author', None)
if highlight_authors:
highlight_authors = highlight_authors.split(';')
style = Style(self.site.config['BASE_URL'] +
detail_page_dir if detail_page_dir else None)
self.state.document.settings.record_dependencies.add(self.arguments[0])
parser = Parser()
# Sort the publication entries by year reversed
data = sorted(parser.parse_file(self.arguments[0]).entries.items(),
key=lambda e: e[1].fields['year'],
reverse=True)
html = '<div class="publication-list">\n'
cur_year = None
if bibtex_dir: # create the bibtex dir if the option is set
try:
os.makedirs(os.path.sep.join((self.output_folder, bibtex_dir)))
except OSError: # probably because the dir already exists
pass
if detail_page_dir: # create the detail page dir if the option is set
try:
os.makedirs(
os.path.sep.join((self.output_folder, detail_page_dir)))
except OSError: # probably because the dir already exists
pass
for label, entry in data:
# print a year title when year changes
if entry.fields['year'] != cur_year:
if cur_year is not None: # not first year group
html += '</ul>'
cur_year = entry.fields['year']
html += '<h3>{}</h3>\n<ul>'.format(cur_year)
entry.label = label # Pass label to the style.
pub_html = list(style.format_entries(
(entry, )))[0].text.render_as('html')
if highlight_authors: # highlight one of several authors (usually oneself)
for highlight_author in highlight_authors:
pub_html = pub_html.replace(
highlight_author.strip(),
'<strong>{}</strong>'.format(highlight_author), 1)
html += '<li class="publication" style="padding-bottom: 1em;">' + pub_html
extra_links = ""
if 'fulltext' in entry.fields: # the link to the full text, usually a link to the pdf file
extra_links += '[<a href="{}">full text</a>] '.format(
entry.fields['fulltext'])
bibtex_fields = dict(entry.fields)
# Collect and remove custom links (fields starting with "customlink")
custom_links = dict()
for key, value in bibtex_fields.items():
if key.startswith('customlink'):
custom_links[key[len('customlink'):]] = value
# custom fields (custom links)
for key, value in custom_links.items():
extra_links += '[<a href="{}">{}</a>] '.format(value, key)
# Remove some fields for the publicly available BibTeX file since they are mostly only
# used by this plugin.
for field_to_remove in ('abstract', 'fulltext'):
if field_to_remove in bibtex_fields:
del bibtex_fields[field_to_remove]
# Prepare for the bib file. Note detail_page_dir may need bib_data later.
bibtex_entry = Entry(entry.type, bibtex_fields, entry.persons)
bib_data = BibliographyData(dict({label: bibtex_entry}))
bib_string = bib_data.to_string('bibtex')
extra_links += '''
[<a href="javascript:void(0)" onclick="
(function(target, id) {{
if ($('#' + id).css('display') == 'block')
{{
$('#' + id).hide('fast');
$(target).text('BibTeX▼')
}}
else
{{
$('#' + id).show('fast');
$(target).text('BibTeX▲')
}}
}})(this, '{}');">BibTeX▼</a>]
'''.format('bibtex-' + label)
if bibtex_dir: # write bib files to bibtex_dir for downloading
bib_link = '{}/{}.bib'.format(bibtex_dir, label)
bib_data.to_file('/'.join([self.output_folder, bib_link]),
'bibtex')
if extra_links or detail_page_dir or 'abstract' in entry.fields:
html += '<br>'
# Add the abstract link.
if 'abstract' in entry.fields:
html += '''
[<a href="javascript:void(0)" onclick="
(function(target, id) {{
if ($('#' + id).css('display') == 'block')
{{
$('#' + id).hide('fast');
$(target).text('abstract▼')
}}
else
{{
$('#' + id).show('fast');
$(target).text('abstract▲')
}}
}})(this, '{}');">abstract▼</a>] '''.format(
'abstract-' + label)
display_none = '<div id="{}" style="display:none"><pre>{}</pre></div>'
bibtex_display = display_none.format('bibtex-' + label, bib_string)
abstract_text = str(LaTeXParser(entry.fields['abstract']).parse()
) if 'abstract' in entry.fields else ''
if detail_page_dir: # render the details page of a paper
page_url = '/'.join((detail_page_dir, label + '.html'))
html += '[<a href="{}">details</a>] '.format(
self.site.config['BASE_URL'] + page_url)
context = {
'title': str(LaTeXParser(entry.fields['title']).parse()),
'abstract': abstract_text,
'bibtex': bib_data.to_string('bibtex'),
'bibtex_link': '/' + bib_link if bibtex_dir else '',
'default_lang': self.site.config['DEFAULT_LANG'],
'label': label,
'lang': self.site.config['DEFAULT_LANG'],
'permalink': self.site.config['SITE_URL'] + page_url,
'reference': pub_html,
'extra_links': extra_links + bibtex_display
}
if 'fulltext' in entry.fields and entry.fields[
'fulltext'].endswith('.pdf'):
context['pdf'] = entry.fields['fulltext']
self.site.render_template(
'publication.tmpl',
os.path.sep.join((self.output_folder, detail_page_dir,
label + '.html')),
context,
)
html += extra_links
# Add the hidden abstract and bibtex.
if 'abstract' in entry.fields:
html += '''
<div id="{}" class="publication-abstract" style="display:none">
<blockquote>{}</blockquote></div>
'''.format('abstract-' + label, abstract_text)
html += bibtex_display
html += '</li>'
if len(data) != 0: # publication list is nonempty
html += '</ul>'
html += '</div>'
return [
nodes.raw('', html, format='html'),
]
def update_contents(self, new_store_contents):
"""
Structure -> mpid -> BibTeX references from MP -> (optional doi lookup
via Crossref) -> formatting.
Formatting is very messy right now.
DOI lookup and (possibly) formatting should be cached in a builder.
"""
struct = self.from_data(new_store_contents)
if not isinstance(struct, Structure):
raise PreventUpdate(
"Literature mentions can only be retrieved for crystallographic "
"structures at present and not molecules. Please make a feature "
"request if this would be useful for you, and it will be "
"prioritized."
)
with MPRester() as mpr:
mpids = mpr.find_structure(struct)
if len(mpids) == 0:
raise PreventUpdate(
"No structures in the Materials Project database match this "
"crystal structure, so literature mentions cannot be retrieved. "
"Please submit this structure to Materials Project if you'd "
"like it to be added to the Materials Project database."
)
all_references = []
for mpid in mpids:
all_references.append(mpr.get_materials_id_references(mpid))
self.logger.debug(f"Retrieved references for {mpid}.")
if self.use_crossref:
cr = Crossref(mailto=CROSSREF_MAILTO)
individual_references = set()
for references in all_references:
individual_references.update(set(references.split("\n\n")))
# exclude Materials Proect references (these are intended to be
# references for the structure specifically)
refs_to_remove = set()
for ref in individual_references:
if "Jain2013" in ref:
refs_to_remove.add(ref)
individual_references -= refs_to_remove
works = [cr.works(query=ref, limit=1) for ref in individual_references]
self.logger.debug(f"Retrieved {len(works)} works from Crossref.")
items = [
work["message"]["items"][0]
for work in works
if len(work["message"]["items"]) > 0
]
dois_to_item = {
item["DOI"]: {
"cited-by": item.get("is-referenced-by-count", 0),
"score": item["score"],
"title": item.get("title", None),
"authors": item.get("author", []),
"journal": item.get("container-title", [None])[0],
"issue": item.get("issue", None),
"volume": item.get("volume", None),
"pages": item.get("page", None),
"date-parts": item.get("issued", {}).get("date-parts", [[None]]),
}
for item in items
if item["score"] > 40
}
num_refs = len(dois_to_item)
sorted_dois = sorted(
list(dois_to_item.keys()),
key=lambda doi: -dois_to_item[doi]["cited-by"],
)
if self.use_crossref_formatting:
# use Crossref to retrieve pre-formatted text
# remove leading "1. " from Science CSL style
refs = {
doi: content_negotiation(ids=doi, format="text", style="science")[
3:
]
for doi in dois_to_item.keys()
}
self.logger.debug(
f"Retrieved {len(refs)} formatted references from Crossref."
)
md = " \n\n".join(
f"> [{refs[doi]}](https://dx.doi.org/{doi}) "
f"Cited by {dois_to_item[doi]['cited-by']}."
for doi in sorted_dois
)
formatted_references = dcc.Markdown(
md, className="mpc-markdown"
)
else:
# else retrieve BibTeX entries to extract a nice author list
# and perform our own formatting
entries = {
doi: content_negotiation(ids=doi, format="bibtex")
for doi in sorted_dois
}
formatted_entries = []
for doi, entry in entries.items():
author_string = self._bibtex_entry_to_author_text(entry)
journal_div = self._item_to_journal_div(dois_to_item[doi])
formatted_entries.append(
html.Blockquote(
[
html.A(
[
html.Div(
[
html.I(
# necessary since titles can contain HTML for superscripts etc.
dcc.Markdown(
dois_to_item[doi]["title"],
dangerously_allow_html=True
)
)
]
),
html.Div([author_string]),
html.Div(
[
journal_div,
html.Span(
f" Cited by {dois_to_item[doi]['cited-by']}."
),
]
),
],
href=f"https://dx.doi.org/{doi}",
)
],
className="mpc",
style={"padding-left": "1rem", "margin-bottom": "1rem"}
)
)
formatted_references = html.Div(formatted_entries)
else:
# this uses pybtex directly on stored BibTeX entries from MP
# most-accurate references and faster since no Crossref lookup
# is required but no dois/hyperlinks available
all_entries = {}
for references in all_references:
all_entries.update(Parser().parse_string(references).entries)
md = self._pybtex_entries_to_markdown(all_entries)
formatted_references = dcc.Markdown(md, className="mpc-markdown")
num_refs = len(all_entries)
return html.Div(
[
Label(f"{num_refs} references found{':' if num_refs>0 else '.'}"),
formatted_references,
],
style={"max-height": "20rem", "overflow-y": "scroll"},
)
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS']:
Dictionary that contains bibliographies:
The key denotes the bibliographies name to use in headers
The values describe the BibTeX files to read
Mandatory for this plugin.
generator.settings['PUBLICATIONS_NAVBAR']:
Bool denoting whether a navigation bar containing links to each bibliography should be produced.
Defaults to 'True'.
generator.settings['PUBLICATIONS_HEADER']:
Bool denoting whether a header (h2) should be produced for each bibliography.
Defaults to 'True'.
generator.settings['PUBLICATIONS_SPLIT']:
Bool denoting whether bibliographies should be split by year (h3).
Defaults to 'True'.
generator.settings['PUBLICATIONS_HIGHLIGHTs']:
String, e.g., a name, that will be entailed in a <strong> tag to highlight.
Default: empty
Output
------
generator.context['publications']:
Dictionary containing the name of the publication list a a key, bibliography entries as a value.
A bibliography entry contains of a list of tuples (key, year, text, bibtex, pdf, slides, poster).
See Readme.md for more details.
"""
if 'PUBLICATIONS' not in generator.settings:
return
if 'PUBLICATIONS_NAVBAR' not in generator.settings:
generator.context['PUBLICATIONS_NAVBAR'] = True
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
from pybtex.style.formatting import plain
except ImportError:
logger.warn('`pelican_bibtex` failed to load dependency `pybtex`')
return
refs = generator.settings['PUBLICATIONS']
generator.context['publications'] = collections.OrderedDict()
for rid in refs:
ref = refs[rid]
bibfile = os.path.join(generator.settings['PATH'], ref['file'])
try:
bibdata_all = Parser().parse_file(bibfile)
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' %
(bibfile, str(e)))
return
if 'title' in ref:
title = ref['title']
else:
title = rid
if 'header' in ref:
header = ref['header']
else:
header = True
if 'split' in ref:
split = ref['split']
else:
split = True
if 'split_link' in ref:
split_link = ref['split_link']
else:
split_link = True
if 'bottom_link' in ref:
bottom_link = ref['bottom_link']
else:
bottom_link = True
if 'all_bibtex' in ref:
all_bibtex = ref['all_bibtex']
else:
all_bibtex = False
if 'highlight' in ref:
highlights = ref['highlight']
else:
highlights = []
if 'group_type' in ref:
group_type = ref['group_type']
else:
group_type = False
publications = []
# format entries
plain_style = plain.Style()
html_backend = html.Backend()
formatted_entries = plain_style.format_entries(
bibdata_all.entries.values())
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
year = entry.fields.get('year')
typee = entry.type
if entry.fields.get('tags'):
tags = [
tag.strip() for tag in entry.fields.get('tags').split(';')
]
else:
tags = []
display_tags = [
x for x in tags if x != "doi-open" and x != "url-open"
]
# This shouldn't really stay in the field dict
# but new versions of pybtex don't support pop
pdf = entry.fields.get('pdf', None)
slides = entry.fields.get('slides', None)
poster = entry.fields.get('poster', None)
doi = entry.fields.get('doi', None)
url = entry.fields.get('url', None)
#clean fields from appearing in bibtex and on website
entry_tmp = entry
for to_del in ['pdf', 'slides', 'poster', 'tags']:
entry_tmp.fields.pop(to_del, None)
#render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry_tmp})
Writer().write_stream(bibdata_this, bib_buf)
#clean more fields from appearing on website
for to_del in ['doi', 'url']:
entry_tmp.fields.pop(to_del, None)
entry_clean = next(
plain_style.format_entries(bibdata_this.entries.values()),
None)
# apply highlight (strong)
text = entry_clean.text.render(html_backend)
for replace in highlights:
text = text.replace(replace,
'<strong>' + replace + '</strong>')
publications.append(
(key, typee, year, text, tags, display_tags,
bib_buf.getvalue(), pdf, slides, poster, doi, url))
generator.context['publications'][rid] = {}
generator.context['publications'][rid]['title'] = title
generator.context['publications'][rid]['path'] = os.path.basename(
bibfile)
generator.context['publications'][rid]['header'] = header
generator.context['publications'][rid]['split'] = split
generator.context['publications'][rid]['bottom_link'] = bottom_link
generator.context['publications'][rid]['split_link'] = split_link
generator.context['publications'][rid]['all_bibtex'] = all_bibtex
generator.context['publications'][rid][
'data'] = collections.OrderedDict()
if group_type:
generator.context['publications'][rid]['data'] = sorted(
publications,
key=lambda pub:
(-int(pub[2].replace("in press", "9999")), pub[1]))
else:
generator.context['publications'][rid]['data'] = sorted(
publications,
key=lambda pub: -int(pub[2].replace("in press", "9999")))
def run(self):
bibtex_dir = self.options.get('bibtex_dir', 'bibtex')
detail_page_dir = self.options.get('detail_page_dir', 'papers')
highlight_authors = self.options.get('highlight_author', None)
if highlight_authors:
highlight_authors = highlight_authors.split(';')
style = Style(self.site.config['BASE_URL'] + detail_page_dir if detail_page_dir else None)
self.state.document.settings.record_dependencies.add(self.arguments[0])
all_entries = []
labels = set()
for a in self.arguments:
parser = Parser()
for item in parser.parse_file(a).entries.items():
if item[0] in labels: # duplicated entries
LOGGER.warning(
("publication_list: BibTeX entries with duplicated labels are found. "
"Only the first occurrence will be used."))
continue
labels.add(item[0])
all_entries.append(item)
# Sort the publication entries by year reversed
data = sorted(all_entries, key=lambda e: e[1].fields['year'], reverse=True)
html = '<div class="publication-list">\n'
cur_year = None
if bibtex_dir: # create the bibtex dir if the option is set
try:
os.makedirs(os.path.sep.join((self.output_folder, bibtex_dir)))
except OSError: # probably because the dir already exists
pass
if detail_page_dir: # create the detail page dir if the option is set
try:
os.makedirs(os.path.sep.join((self.output_folder, detail_page_dir)))
except OSError: # probably because the dir already exists
pass
for label, entry in data:
# print a year title when year changes
if entry.fields['year'] != cur_year:
if cur_year is not None: # not first year group
html += '</ul>'
cur_year = entry.fields['year']
html += '<h3>{}</h3>\n<ul>'.format(cur_year)
entry.label = label # Pass label to the style.
pub_html = list(style.format_entries((entry,)))[0].text.render_as('html')
if highlight_authors: # highlight one of several authors (usually oneself)
for highlight_author in highlight_authors:
# We need to replace all occurrence of space except for the last one with
# , since pybtex does it for all authors
count = highlight_author.count(' ') - 1
pub_html = pub_html.replace(
highlight_author.strip().replace(' ', ' ', count),
'<strong>{}</strong>'.format(highlight_author), 1)
html += '<li class="publication" style="padding-bottom: 1em;">' + pub_html
extra_links = ""
if 'fulltext' in entry.fields: # the link to the full text, usually a link to the pdf file
extra_links += '[<a href="{}">full text</a>] '.format(entry.fields['fulltext'])
bibtex_fields = dict(entry.fields)
# Collect and remove custom links (fields starting with "customlink")
custom_links = dict()
for key, value in bibtex_fields.items():
if key.startswith('customlink'):
custom_links[key[len('customlink'):]] = value
# custom fields (custom links)
for key, value in custom_links.items():
extra_links += '[<a href="{}">{}</a>] '.format(value, key)
# Remove some fields for the publicly available BibTeX file since they are mostly only
# used by this plugin.
for field_to_remove in ('abstract', 'fulltext'):
if field_to_remove in bibtex_fields:
del bibtex_fields[field_to_remove]
# Prepare for the bib file. Note detail_page_dir may need bib_data later.
bibtex_entry = Entry(entry.type, bibtex_fields, entry.persons)
bib_data = BibliographyData(dict({label: bibtex_entry}))
bib_string = bib_data.to_string('bibtex')
extra_links += '''
[<a href="javascript:void(0)" onclick="
(function(target, id) {{
if ($('#' + id).css('display') == 'block')
{{
$('#' + id).hide('fast');
$(target).text('BibTeX▼')
}}
else
{{
$('#' + id).show('fast');
$(target).text('BibTeX▲')
}}
}})(this, '{}');">BibTeX▼</a>]
'''.format('bibtex-' + label)
if bibtex_dir: # write bib files to bibtex_dir for downloading
bib_link = '{}/{}.bib'.format(bibtex_dir, label)
bib_data.to_file('/'.join([self.output_folder, bib_link]), 'bibtex')
if extra_links or detail_page_dir or 'abstract' in entry.fields:
html += '<br>'
# Add the abstract link.
if 'abstract' in entry.fields:
html += '''
[<a href="javascript:void(0)" onclick="
(function(target, id) {{
if ($('#' + id).css('display') == 'block')
{{
$('#' + id).hide('fast');
$(target).text('abstract▼')
}}
else
{{
$('#' + id).show('fast');
$(target).text('abstract▲')
}}
}})(this, '{}');">abstract▼</a>] '''.format('abstract-' + label)
display_none = '<div id="{}" style="display:none"><pre>{}</pre></div>'
bibtex_display = display_none.format(
'bibtex-' + label, bib_string)
abstract_text = str(
LaTeXParser(entry.fields['abstract']).parse()) if 'abstract' in entry.fields else ''
if detail_page_dir: # render the details page of a paper
page_url = '/'.join((detail_page_dir, label + '.html'))
html += '[<a href="{}">details</a>] '.format(
self.site.config['BASE_URL'] + page_url)
context = {
'title': str(LaTeXParser(entry.fields['title']).parse()),
'abstract': abstract_text,
'bibtex': bib_data.to_string('bibtex'),
'bibtex_link': '/' + bib_link if bibtex_dir else '',
'default_lang': self.site.config['DEFAULT_LANG'],
'label': label,
'lang': self.site.config['DEFAULT_LANG'],
'permalink': self.site.config['SITE_URL'] + page_url,
'reference': pub_html,
'extra_links': extra_links + bibtex_display
}
if 'fulltext' in entry.fields:
context['pdf'] = entry.fields['fulltext']
self.site.render_template(
'publication.tmpl',
os.path.sep.join((self.output_folder, detail_page_dir, label + '.html')),
context,
)
html += extra_links
# Add the hidden abstract and bibtex.
if 'abstract' in entry.fields:
html += '''
<div id="{}" class="publication-abstract" style="display:none">
<blockquote>{}</blockquote></div>
'''.format('abstract-' + label, abstract_text)
html += bibtex_display
html += '</li>'
if len(data) != 0: # publication list is nonempty
html += '</ul>'
html += '</div>'
return [nodes.raw('', html, format='html'), ]
def verifyBib(bibfile, verifiedfile="", unverifiedfile=""):
"""Verifies a bib file according to the present doi or to a doi found in internet. Returns a filename.verified.bib and filename.unverified.bib"""
if verifiedfile == "":
verifiedfile = bibfile + ".verified.bib"
if unverifiedfile == "":
unverifiedfile = bibfile + ".unverified.bib"
with open(bibfile,"r") as f:
bstr = f.read()
# Assume the input is utf8 encoded
bstr = bstr.decode('utf8')
# Parse the bibtex file
bib_parser = Parser()
bib_data = bib_parser.parse_stream(StringIO(bstr))
#first I groom the original bib file:
bib_data=cleanBibliographyData(bib_data)
newbib_parser = Parser()
unverif_parser = Parser()
#bib_verifdata = bib_parser.parse_stream(StringIO(""))
entrynumber=0
check =0
for (key,entry) in bib_data.entries.items():
check = check+1
print "-------------------------------------"
print "Checking entry %d of %d" % (check,len(bib_data.entries))
print "Entry: %s..." % key
title = entry.fields['title'] if entry.fields.has_key('title') else ""
author = entry.fields['author'] if entry.persons.has_key('author') else ""
journal = entry.fields[_journal_field] if entry.fields.has_key(_journal_field) else ""
volume = entry.fields['volume'] if entry.fields.has_key('volume') else ""
pages = entry.fields['pages'] if entry.fields.has_key('pages') else ""
print "('%s' by %s)" % (title,author)
#fix from here
"""first look for doi in entry if not use doi_finder,
if doi is found,
then download entryfrom internet and
hten compare with the actual entry, ask for changes
"""
doi=None
if 'doi' in entry.fields:
doi = entry.fields['doi']
print "doi found in bib file..."
verificationmode="from given doi"
else: #if 'doi' is not in entry.fields, we try to find it
print "Looking for doi in crossref..."
doi = doi_finder.crossref_auth_title_to_doi(doi_finder.detex(author), doi_finder.detex(title))
verificationmode="from searched doi in crossref"
if doi != None:
print "doi found in crossref"
else:
print "Looking for doi in google..."
doi = doi_finder.google_doi(journal, volume, pages, doi_finder.detex(title))
verificationmode="from searched doi in google"
if doi != None:
print "doi found in google scholars"
if doi != None:
print "Retrieving fields from internet..."
entriesfrominternet = doi2biblatex(doi)
##check if it's the right entry:
if entriesfrominternet != None and len(entriesfrominternet.entries.keys())>0 and entriesfrominternet.entries[entriesfrominternet.entries.keys()[0]].fields.has_key("title"):
titlefrominternet= entriesfrominternet.entries[entriesfrominternet.entries.keys()[0]].fields["title"]
if doi_finder.fuzzy_match(title.lower(), titlefrominternet.lower()) < .9:
usedoi = raw_input("maybe wrong entry... Title from internet is %s. Is it right? y/[N]:" % titlefrominternet)
if usedoi=="" or usedoi[0].capitalize()=="N":
print "Avoiding entry..."
entriesfrominternet = None
else:
print "Using entry from internet..."
else:
print "incomplete data from internet, avoiding entry..."
entriesfrominternet = None
else:
print "doi not found, avoiding entry..."
entriesfrominternet = None
if entriesfrominternet !=None:
fieldsfrominternet = entriesfrominternet.entries[entriesfrominternet.entries.keys()[0]].fields #It will contain only one entry, so I take the fields from the first one
personsfrominternet = entriesfrominternet.entries[entriesfrominternet.entries.keys()[0]].persons #authors, editors, etc
verifiedfields=""
#example
"""
Entry(u'article', fields={u'doi': u'10.1016/S0364-0213(99)80005-6', u'title': u'{A probabilistic model of lexical and syntactic access and disambiguation}',
u'url': u'http://doi.wiley.com/10.1016/S0364-0213(99)80005-6', u'journaltitle': u'Cognitive Science', u'issn': u'03640213',
u'mendeley-tags': u'Phd1,expectations,predictions', u'number': u'2', u'month': u'06', u'volume': u'2000',
u'file': u':home/bruno/Documents/Papers//Jurafsky - 1996 - A probabilistic model of lexical and syntactic access and disambiguation.pdf:pdf', u'year': u'1996',
u'keywords': u'Phd1,expectations,predictions', u'pages': u'7--194'},
persons={u'author': [Person(u'Jurafsky, D')]})
"""
for persons in personsfrominternet:
verifiedfields = persons +"; "+verifiedfields
if entry.persons.has_key(persons): #check if there's a person field like author, maybe editor
# unicode(entry.persons[persons][0]) == "apellido, nombre"
if entry.persons[persons] != personsfrominternet[persons]: #then compare with the old one and ask
#First checks for badly bwritten entries:
for i in range(0,min(len(entry.persons[persons]),len(personsfrominternet[persons]))): #checks the common persons
lastbib = entry.persons[persons][i].get_part_as_text("last")
lastinet = personsfrominternet[persons][i].get_part_as_text("last")
firstbib = entry.persons[persons][i].get_part_as_text("first")
firstinet = personsfrominternet[persons][i].get_part_as_text("first")
middlebib = entry.persons[persons][i].get_part_as_text("middle")
middleinet = personsfrominternet[persons][i].get_part_as_text("middle")
lineagebib = entry.persons[persons][i].get_part_as_text("lineage")
lineageinet = personsfrominternet[persons][i].get_part_as_text("lineage")
prelastbib = entry.persons[persons][i].get_part_as_text("prelast")
prelastinet = personsfrominternet[persons][i].get_part_as_text("prelast")
#check if the name in internet has less info that the one stored
if lastbib==lastinet and (firstbib == firstinet[0] or firstbib == firstinet[0]+"." ):
print "incomplete last name in internet for %s - Skipping name %s ..." % (unicode(entry.persons[persons][i]),unicode(personsfrominternet[persons][i]))
else: #if there's the same amount of info then check the names
if unicode(entry.persons[persons][i]).strip() != unicode(personsfrominternet[persons][i]).strip():
entry.persons[persons][i] = changeThisforThat(entry.persons[persons][i],personsfrominternet[persons][i],"Change for %s." % persons)
#check for missing or extra authors
if len(entry.persons[persons]) < len(personsfrominternet[persons]): # missing authors in bib file
for j in range(i+1,len(personsfrominternet[persons]) ):
missing = changeThisforThat(None,personsfrominternet[persons][j],"Missing person for %s." % persons)
if missing != None:
entry.persons[persons].append(missing)
elif len(entry.persons[persons]) > len(personsfrominternet[persons]): # extra authors in bib file
for j in range(i+1,len(entry.persons[persons]) ):
extra = changeThisforThat(entry.persons[persons][j],None,"Extra person for %s." % persons)
if extra == None:
entry.persons[persons].remove(entry.persons[persons][j])
else: #if there's new field online, the field has to be added
entry.fields[persons] = fieldsfrominternet[persons]
for fieldname in fieldsfrominternet:
verifiedfields = fieldname +"; "+verifiedfields
if entry.fields.has_key(fieldname): #i check if there are new fields first
if entry.fields[fieldname] != fieldsfrominternet[fieldname]: #then compare with the old ones and ask
#don't try to change the title if it's in title case:
if not (fieldname== "title" and fieldsfrominternet[fieldname] == titlecase.titlecase(fieldsfrominternet[fieldname])):
entry.fields[fieldname] = changeThisforThat(entry.fields[fieldname],fieldsfrominternet[fieldname],"Change for %s." % fieldname)
else:
print "avoiding title '%s' because of titlecase" % fieldsfrominternet[fieldname]
else: #if there's new field online, the field has to be added
entry.fields[fieldname] = fieldsfrominternet[fieldname]
textverif = "%s were verified %s" % (verifiedfields,verificationmode)
entry.fields["citation-verif"] = textverif
#add to new bib
newbib_parser.data.add_entry(key, entry) #these are the verified entries
entrynumber=entrynumber+1
else: # if cit.citation() =={}: #if we failed to find it in internet
unverif_parser.data.add_entry(key, entry)
savebib(unverif_parser.data,unverifiedfile)
savebib(newbib_parser.data,verifiedfile)
print "%s/%s entries were checked. %s were verified" % (check, len(bib_data.entries), entrynumber)
return "done"
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS_SRC']:
local path to the BibTeX file to read.
Output
------
generator.context['publications']:
List of tuples (key, year, text, bibtex, pdf, slides, poster).
See Readme.md for more details.
"""
if 'PUBLICATIONS_SRC' not in generator.settings:
return
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
from pybtex.style.formatting import plain
except ImportError:
LOGGER.warn('`pelican_bibtex` failed to load dependency `pybtex`')
return
try:
bib_items = Parser().parse_file(generator.settings['PUBLICATIONS_SRC'])
except PybtexError as err:
LOGGER.warn('`pelican_bibtex` failed to parse file %s: %s',
generator.settings['PUBLICATIONS_SRC'],
str(err))
return
publications = []
for fmt_entry in plain.Style().format_entries(bib_items.entries.values()):
key = fmt_entry.key
entry = bib_items.entries[key]
# Render the bibtex string for the entry
buf = StringIO()
Writer().write_stream(BibliographyData(entries={key: entry}), buf)
# Prettify BibTeX entries
text = fmt_entry.text.render(html.Backend())
text = text.replace(r"\{", "").replace(r"\}", "")
text = text.replace("{", "").replace("}", "")
publications.append({'bibtex' : buf.getvalue(),
'doi' : get_field(entry, 'doi'),
'entry' : entrytype(entry.type),
'key' : key,
'pdf' : get_field(entry, 'pdf'),
'poster' : get_field(entry, 'poster'),
'slides' : get_field(entry, 'slides'),
'text' : text,
'url' : get_field(entry, 'url'),
'note' : get_field(entry, 'note'),
'year' : entry.fields.get('year'),
'authorizer': get_field(entry, 'authorizer'),
'acceptance': get_field(entry, 'acceptance'),
'stats': get_field(entry, 'stats')
})
generator.context['publications'] = publications
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS_SRC']:
local path to the BibTeX file to read.
Output
------
generator.context['publications']:
List of tuples (key, year, text, bibtex, pdf, slides, poster).
See Readme.md for more details.
"""
if 'PUBLICATIONS_SRC' not in generator.settings:
return
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
#from pybtex.style.formatting import plain
from rahul_style import Style as RahulStyle
except ImportError:
logger.warn('`pelican_bibtex` failed to load dependency `pybtex`')
return
refs_file = generator.settings['PUBLICATIONS_SRC']
try:
bibdata_all = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' %
(refs_file, str(e)))
return
publications = []
# format entries
plain_style = RahulStyle()
#plain_style = plain.Style()
html_backend = html.Backend()
html_backend.symbols['br'] = u'<BR/>'
all_entries = bibdata_all.entries.values()
# remove URL field if DOI is present
for entry in all_entries:
if "doi" in entry.fields.keys():
entry.fields._dict["url"] = ""
formatted_entries = plain_style.format_entries(all_entries)
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
pub_type = entry.type
year = entry.fields.get('year')
# This shouldn't really stay in the field dict
# but new versions of pybtex don't support pop
pdf = entry.fields.get('pdf', None)
#slides = entry.fields.get('slides', None)
#poster = entry.fields.get('poster', None)
doi = entry.fields.get('doi', None)
url = entry.fields.get('url', None)
arxiv = entry.fields.get('arxiv', None)
#render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
text = formatted_entry.text.render(html_backend)
# prettify entries
# remove BibTeX's {}
text = text.replace("\{", "")
text = text.replace("{", "")
text = text.replace("\}", "")
text = text.replace("}", "")
# remove textbf used for cv
text = text.replace("\\textbf ", "")
# remove \ that comes after Proc.
text = text.replace("\\", "")
publications.append((pub_type, key, year, text, bib_buf.getvalue(),
pdf, doi, url, arxiv))
generator.context['publications'] = publications
def _test(bibtex_input, correct_result):
parser = Parser(encoding='UTF-8')
parser.parse_stream(StringIO(bibtex_input))
result = parser.data
assert result == correct_result
import logging
logger = logging.getLogger(__name__)
import os
import codecs
from pybtex.database.input.bibtex import Parser
PUBLICATION_DIR = 'content/publications'
# Iterate over all the files in the PUBLICATION_DIR
for file in os.listdir(PUBLICATION_DIR):
logger.warn( '[BIB] Trying to parse {}...'.format(file))
# Try parsing it, should not crash
with codecs.open(PUBLICATION_DIR+os.sep+file, 'r', encoding="utf8") as stream:
assert len(stream.read()) > 0
bibdata = Parser().parse_stream(stream)
def run(self):
style = find_plugin('pybtex.style.formatting',
self.options.get('style', 'unsrt'))()
bibtex_dir = self.options.get('bibtex_dir', 'bibtex')
detail_page_dir = self.options.get('detail_page_dir', 'papers')
highlight_authors = self.options.get('highlight_author', None)
if highlight_authors:
highlight_authors = highlight_authors.split(';')
self.state.document.settings.record_dependencies.add(self.arguments[0])
parser = Parser()
# Sort the publication entries by year reversed
data = sorted(parser.parse_file(self.arguments[0]).entries.items(),
key=lambda e: e[1].fields['year'],
reverse=True)
html = '<div class="publication-list">\n'
cur_year = None
if bibtex_dir: # create the bibtex dir if the option is set
try:
os.mkdir(os.path.sep.join((self.output_folder, bibtex_dir)))
except OSError: # probably because the dir already exists
pass
if detail_page_dir: # create the detail page dir if the option is set
try:
os.mkdir(
os.path.sep.join((self.output_folder, detail_page_dir)))
except OSError: # probably because the dir already exists
pass
for label, entry in data:
# print a year title when year changes
if entry.fields['year'] != cur_year:
if cur_year is not None: # not first year group
html += '</ul>'
cur_year = entry.fields['year']
html += '<h3>{}</h3>\n<ul>'.format(cur_year)
pub_html = list(style.format_entries(
(entry, )))[0].text.render_as('html')
if highlight_authors: # highlight one of several authors (usually oneself)
for highlight_author in highlight_authors:
pub_html = pub_html.replace(
highlight_author.strip(),
'<strong>{}</strong>'.format(highlight_author), 1)
html += '<li class="publication" style="padding-bottom: 1em;">' + pub_html
extra_links = ""
bibtex_fields = dict(entry.fields)
# Remove some fields for the publicly available BibTeX file since they are mostly only
# used by this plugin.
for field_to_remove in ('abstract', 'fulltext'):
if field_to_remove in bibtex_fields:
del bibtex_fields[field_to_remove]
bibtex_entry = Entry(entry.type, bibtex_fields, entry.persons)
# detail_page_dir may need bib_data later
bib_data = BibliographyData(dict({label: bibtex_entry}))
if bibtex_dir: # write bib files to bibtex_dir for downloading
bib_link = '{}/{}.bib'.format(bibtex_dir, label)
bib_data.to_file('/'.join([self.output_folder, bib_link]),
'bibtex')
extra_links += '[<a href="{}">BibTeX</a>] '.format(
self.site.config['BASE_URL'] + bib_link)
if 'fulltext' in entry.fields: # the link to the full text, usually a link to the pdf file
extra_links += '[<a href="{}">full text</a>] '.format(
entry.fields['fulltext'])
if extra_links or detail_page_dir:
html += '<br>'
html += extra_links
if detail_page_dir: # render the details page of a paper
page_url = '/'.join((detail_page_dir, label + '.html'))
html += ' [<a href="{}">abstract and details</a>]'.format(
self.site.config['BASE_URL'] + page_url)
context = {
'title':
str(LaTeXParser(entry.fields['title']).parse()),
'abstract':
str(LaTeXParser(entry.fields['abstract']).parse())
if 'abstract' in entry.fields else '',
'bibtex':
bib_data.to_string('bibtex'),
'bibtex_link':
'/' + bib_link if bibtex_dir else '',
'default_lang':
self.site.config['DEFAULT_LANG'],
'label':
label,
'lang':
self.site.config['DEFAULT_LANG'],
'permalink':
self.site.config['SITE_URL'] + page_url,
'reference':
pub_html,
'extra_links':
extra_links
}
if 'fulltext' in entry.fields and entry.fields[
'fulltext'].endswith('.pdf'):
context['pdf'] = entry.fields['fulltext']
self.site.render_template(
'publication.tmpl',
os.path.sep.join((self.output_folder, detail_page_dir,
label + '.html')),
context,
)
html += '</li>'
if len(data) != 0: # publication list is nonempty
html += '</ul>'
html += '</div>'
return [
nodes.raw('', html, format='html'),
]
def add_publications(self):
# Check if PUBLICATIONS_SRC is set
if 'PUBLICATIONS_SRC' not in self.settings:
logger.warn('PUBLICATIONS_SRC not set')
return
# Try to parse the bibtex files
pub_dir = self.settings['PUBLICATIONS_SRC']
try:
bibdata_all = BibliographyData()
for file in os.listdir(pub_dir):
with codecs.open(pub_dir + os.sep + file, 'r',
encoding="utf8") as stream:
bibdata = Parser().parse_stream(stream)
key, entry = bibdata.entries.items()[0]
bibdata_all.entries[key] = entry
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' %
(file, str(e)))
return
# Create Publication objects and add them to a list
publications = []
# format entries
plain_style = plain.Style()
formatted_entries = list(
plain_style.format_entries(bibdata_all.entries.values()))
decoder = latexcodec.lexer.LatexIncrementalDecoder()
for entry in bibdata_all.entries:
raw_tex = BibliographyData(entries={
entry: bibdata_all.entries[entry]
}).to_string('bibtex')
#raw_tex += '\n}'
formatted_entry = list(
plain_style.format_entries([bibdata_all.entries[entry]]))[0]
key = formatted_entry.key
entry = bibdata_all.entries[key]
year = entry.fields.get('year', 2018)
authors = entry.fields.get('author', '').split(' and ')
print(authors)
parsed_authors = []
for author in authors:
if ',' in author:
parsed_authors.append(LatexNodes2Text().latex_to_text(
re.sub(r'[\{\}]', '', (author.split(',')[1] + ' ' +
author.split(',')[0]).strip())))
else:
parsed_authors.append(
LatexNodes2Text().latex_to_text(author))
authors = parsed_authors
title = LatexNodes2Text().latex_to_text(
entry.fields.get('title', ''))
pdf = entry.fields.get('pdf', None)
slides = entry.fields.get('slides', None)
poster = entry.fields.get('poster', None)
where = ''
if 'booktitle' in entry.fields:
where = LatexNodes2Text().latex_to_text(
entry.fields.get('booktitle'))
elif 'journal' in entry.fields:
where = LatexNodes2Text().latex_to_text(
entry.fields.get('journal'))
abstract = entry.fields.get('abstract', '')
pub = Publication(key,
authors,
title,
year,
where,
abstract=abstract,
pdf_url=pdf,
resource_urls=[('slides', slides),
('poster', poster)])
pub.citations['bib'] = raw_tex.rstrip('\r\n')
publications.append(pub)
self.publications_per_year[pub.year].append(pub)
for author in authors:
if author in self.context['MEDIUS_AUTHORS'].keys():
self.publications_per_author[author].append(pub)
self.publications_per_type[BIBTEX_TYPE_TO_TEXT[entry.type]].append(
pub)
self.publications_per_type_rev[pub] = BIBTEX_TYPE_TO_TEXT[
entry.type]
return publications
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS_SRC']:
local path to the BibTeX file to read.
Output
------
generator.context['publications']:
List of tuples (key, year, text, bibtex, pdf, slides, poster).
See Readme.md for more details.
"""
if 'PUBLICATIONS_SRC' not in generator.settings:
return
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
from pybtex.style.formatting import plain
except ImportError:
logger.warn('`pelican_bibtex` failed to load dependency `pybtex`')
return
refs_file = generator.settings['PUBLICATIONS_SRC']
try:
bibdata_all = Parser().parse_file(refs_file)
except PybtexError as e:
logger.error('`pelican_bibtex` failed to parse file %s: %s' % (
refs_file,
str(e)))
exit(1)
return
publications = []
# format entries
plain_style = plain.Style()
html_backend = html.Backend()
formatted_entries = plain_style.format_entries(bibdata_all.entries.values())
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
year = entry.fields.get('year')
XEcategory = entry.fields.get('XEcategory')
XEmember = entry.fields.get('XEmember')
XEProject = entry.fields.get('XEProject')
url = entry.fields.get('XEurl')
#render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
text = formatted_entry.text.render(html_backend)
# publications.append((key,
# year,
# text,
# url,
# XEmember,
# XEcategory,
# XEProject
# ))
publications.append({'key' : key,
'year' : year,
'text' : text,
'url' : url,
'XEmember' : XEmember,
'XEcategory' : XEcategory,
'XEProject' : XEProject})
generator.context['publications'] = publications
def add_publications(generator):
"""
Populates context with a list of BibTeX publications.
Configuration
-------------
generator.settings['PUBLICATIONS_SRC']:
local path to the BibTeX file to read.
Output
------
generator.context['publications']:
List of tuples (key, year, text, bibtex, url, slides, poster).
See Readme.md for more details.
"""
if 'PUBLICATIONS_SRC' not in generator.settings:
return
refs_file = generator.settings['PUBLICATIONS_SRC']
try:
bibdata_all = Parser().parse_file(refs_file)
except PybtexError as e:
logger.warn('`pelican_bibtex` failed to parse file %s: %s' %
(refs_file, str(e)))
return
# format entries
plain_style = MyStyle()
html_backend = html.Backend()
formatted_entries = plain_style.format_entries(
bibdata_all.entries.values())
publications = []
reports = []
unpublished = []
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_all.entries[key]
try:
year = int(entry.fields.get('year', None))
except TypeError:
year = None
journal = entry.fields.get('journal', "")
sort_key = (year, journal)
# Render the bibtex string for the entry.
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
text = formatted_entry.text.render(html_backend)
entry_res = (key, text, bib_buf.getvalue(), sort_key)
if entry.type == 'article':
publications.append(entry_res)
elif entry.type == 'unpublished':
unpublished.append(entry_res)
else:
reports.append(entry_res)
generator.context['publications'] = sorted(publications,
key=itemgetter(-1),
reverse=True)
generator.context['reports'] = sorted(reports,
key=itemgetter(-1),
reverse=True)
generator.context['unpublished'] = sorted(unpublished,
key=itemgetter(-1),
reverse=True)
def add_publications_to_context(generator, refs_files, refs_string=None, pybtex_style_args={}):
""" Populates context with a list of BibTeX publications. """
try:
from StringIO import StringIO
except ImportError:
from io import StringIO
try:
from pybtex.database.input.bibtex import Parser
from pybtex.database.output.bibtex import Writer
from pybtex.database import BibliographyData, PybtexError
from pybtex.backends import html
from pybtex.style.formatting import BaseStyle, plain
except ImportError:
logger.warn('`pelican_bib` failed to load dependency `pybtex`')
return
decorate_html = generator.settings.get('PUBLICATIONS_DECORATE_HTML', False)
plugin_path = generator.settings.get('PUBLICATIONS_PLUGIN_PATH', 'plugins')
import sys
sys.path.append(plugin_path)
kwargs = generator.settings.get('PUBLICATIONS_STYLE_ARGS', {})
kwargs.update(pybtex_style_args)
style_type = get_style_type(plain.Style, decorate_html)
style = style_type(**kwargs)
if generator.settings.get('PUBLICATIONS_CUSTOM_STYLE', False):
try:
from pybtex_plugins import PelicanStyle
if not isinstance(PelicanStyle, type) or not issubclass(PelicanStyle, BaseStyle):
raise TypeError()
style_type = get_style_type(PelicanStyle, decorate_html)
style = style_type(**kwargs)
except ImportError as e:
logger.warn(str(e))
logger.warn('pybtex_plugins.PelicanStyle not found, using Pybtex plain style')
except TypeError:
logger.warn('PelicanStyle must be a subclass of pybtex.style.formatting.BaseStyle')
# collect entries
bibdata_entries = {}
for file in refs_files:
bibdata_entries.update(Parser().parse_file(file).entries)
if refs_string:
bibdata_entries.update(Parser().parse_string(refs_string).entries)
publications = []
publications_lists = {}
publications_untagged = []
split_by = generator.settings.get('PUBLICATIONS_SPLIT_BY', None)
untagged_title = generator.settings.get('PUBLICATIONS_UNTAGGED_TITLE', None)
# format entries
html_backend = html.Backend()
formatted_entries = style.format_entries(bibdata_entries.values())
for formatted_entry in formatted_entries:
key = formatted_entry.key
entry = bibdata_entries[key]
year = entry.fields.get('year')
# This shouldn't really stay in the field dict
# but new versions of pybtex don't support pop
pdf = entry.fields.get('pdf', None)
slides = entry.fields.get('slides', None)
poster = entry.fields.get('poster', None)
tags = []
if split_by:
tags = entry.fields.get(split_by, [])
# parse to list, and trim each string
if tags:
tags = [tag.strip() for tag in tags.split(',')]
# create keys in publications_lists if at least one
# tag is given
for tag in tags:
publications_lists[tag] = publications_lists.get(tag, [])
#render the bibtex string for the entry
bib_buf = StringIO()
bibdata_this = BibliographyData(entries={key: entry})
Writer().write_stream(bibdata_this, bib_buf)
# convert decorated html tags
# `<:bib-xyz>abc</:bib-xyz>` => `<span class="bib-xyz">abc</span>`
text = formatted_entry.text.render(html_backend)
text = replace(r'<:([^>]*)>', r'<span class="\1">', text)
text = replace(r'</:([^>]*)>', r'</span>', text)
entry_tuple = {'key': key,
'year': year,
'text': text,
'bibtex': bib_buf.getvalue(),
'pdf': pdf,
'slides': slides,
'poster': poster}
entry_tuple.update(entry.fields)
publications.append(entry_tuple)
for tag in tags:
publications_lists[tag].append(entry_tuple)
if not tags and untagged_title:
publications_untagged.append(entry_tuple)
# append untagged list if title is given
if untagged_title and publications_untagged:
publications_lists[untagged_title] = publications_untagged
# output
generator.context['publications'] = publications
generator.context['publications_lists'] = publications_lists
