class CSLRenderer(object):
"""
Interface with citeproc-py CSL library
"""
def __init__(self, doc, style):
self.citations={}
self.doc=doc
style=style.lower()
bib_source = self.convertReferencesToJSON()
bib_source = CiteProcJSON(bib_source)
bib_style = CitationStylesStyle(os.path.join(CSL_PATH,style), validate=False)
self.bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.html)
self.prepareCitations()
def convertReferencesToJSON(self):
"""
Converts a SciDoc's references into the JSON format expected
"""
def copyFields(dict1, dict2, field_list):
for field in field_list:
if field[0] in dict1:
dict2[field[1]]=dict1[field[0]]
res=[]
for ref in self.doc.references:
newref={}
copyFields(ref, newref,
# new = old
[("id","id"),
("title","title"),
("authors","author"),
("publisher","publisher-name"),
("title","title"),
("type","type"),
("URL","url"),
])
newref["issued"]={"date-parts":[(ref.get("year","0"),)]}
res.append(newref)
return res
def prepareCitations(self):
"""
"""
for cit in self.doc.citations:
self.citations[cit["id"]]=Citation([CitationItem(cit["ref_id"])])
self.bibliography.register(self.citations[cit["id"]])
def getCitationText(self, cit):
"""
"""
warn=lambda x:None
return self.bibliography.cite(self.citations[cit["id"]], warn)
def getBibliography(self):
"""
Returns the formatted bibliography
"""
return [str(item) for item in self.bibliography.bibliography()]
def display_citation(bibtex_metadata, bib_stylename, formatter=formatter.html):
# valid style names: plos, apa, pnas, nature, bmj, harvard1
# full list is here: https://github.com/citation-style-language/styles
bib_style = EnhancedCitationStyle(bib_stylename)
bibliography = CitationStylesBibliography(bib_style, bibtex_metadata, formatter) #could be formatter.html
id = "ITEM-1"
citation = Citation([CitationItem(id)])
bibliography.register(citation)
citation_parts = "".join(bibliography.bibliography()[0])
citation_text = "".join(citation_parts)
if bib_stylename == 'apa':
# strip extra periods and spaces that can occur in APA format
citation_text = citation_text.replace('..', '.')
citation_text = citation_text.replace(' ', ' ')
citation_text = citation_text.strip()
# strip leading comma
if citation_text.startswith(','):
citation_text = citation_text.lstrip(',').strip()
citation_text = strip_duplicate_apa_title(bibtex_metadata, citation_text)
citation_text = html.unescape(citation_text)
return citation_text
def getReferencesFromJupyterNotebook(notebook):
metadata = notebook.get('metadata')
references = []
bibliography = []
filterReferences = []
inline_references_table = dict()
try:
references = metadata.get('cite2c').get('citations')
# logger.info("Loging references ---> {0}".format(references))
bib_source = CiteProcJSON(references.values())
bib_style = CitationStylesStyle(
'jdhseo/styles/modern-language-association.csl', validate=False)
bib = CitationStylesBibliography(bib_style, bib_source, formatter.html)
# register citation
for key, entry in bib_source.items():
# exclude "undefined" due to bug cite2c
if key != "undefined":
bib.register(Citation([CitationItem(key)]))
for item in bib.bibliography():
bibliography.append(str(item))
for k, entry in references.items():
inline_references_table[k] = getAuthorDateFromReference(entry)
except Exception as e:
logger.exception(e)
pass
# caseless matching
#return references, sorted(bibliography, key=str.casefold), inline_references_table
return references, sorted(bibliography,
key=lambda x: re.sub('[^A-Za-z]+', '', x).lower(
)), inline_references_table
class RFIC2009Paper(Document):
rngschema = 'rfic.rng'
namespace = 'http://www.mos6581.org/ns/rficpaper'
def __init__(self, filename, bibliography_source):
super().__init__(backend=pdf)
self.styles = styles
parser = xml_frontend.Parser(CustomElement,
self.namespace,
schema=self.rngschema)
xml_tree = parser.parse(filename)
self.root = xml_tree.getroot()
bibliography_style = CitationStylesStyle('ieee.csl')
self.bibliography = CitationStylesBibliography(bibliography_style,
bibliography_source,
csl_formatter)
authors = [author.text for author in self.root.head.authors.author]
if len(authors) > 1:
self.author = ', '.join(authors[:-1]) + ', and ' + authors[-1]
else:
self.author = authors[0]
self.title = self.root.head.title.text
self.keywords = [term.text for term in self.root.head.indexterms.term]
self.parse_input()
def parse_input(self):
self.title_par = Paragraph(self.title, style='title')
self.author_par = Paragraph(self.author, style='author')
self.affiliation_par = Paragraph(self.root.head.affiliation.text,
style='affiliation')
self.content = Chain(self)
self.content << Abstract(self.root.head.abstract.text)
self.content << IndexTerms(self.keywords)
toc_heading = Heading('Table of Contents', style='unnumbered')
toc = TableOfContents()
self.content << rt.Section([toc_heading, toc])
for element in self.root.body.getchildren():
self.content << element.process()
bib_heading = Heading('References', style='unnumbered')
self.bibliography.sort()
bib = Bibliography(self.bibliography)
self.content << rt.Section([bib_heading, bib])
def setup(self):
self.page_count = 1
page = RFICPage(self, first=True)
self.add_page(page, self.page_count)
page.title_box << self.title_par
page.title_box << self.author_par
page.title_box << self.affiliation_par
def new_page(self, chains):
page = RFICPage(self)
self.page_count += 1
self.add_page(page, self.page_count)
def format_bibliography(json_data):
""" Format CSL-JSON to HTML APA format """
json_data = _uniqueify(_flatten(json_data))
bib_source = CiteProcJSON(json_data)
style_path = get_style_filepath('apa')
bib_style = CitationStylesStyle(style_path, validate=False)
# Create the citeproc-py bibliography, passing it the:
# * CitationStylesStyle,
# * BibliographySource (CiteProcJSON in this case), and
# * a formatter (plain, html, or you can write a custom formatter)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.html)
# Processing citations in a document needs to be done in two passes as for
# some CSL styles, a citation can depend on the order of citations in the
# bibliography and thus on citations following the current one.
# For this reason, we first need to register all citations with the
# CitationStylesBibliography.
for c in json_data:
bibliography.register(Citation([CitationItem(c['id'])]))
items = []
for item in bibliography.bibliography():
items.append(str(item))
return items
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
csl = None
if isinstance(node, PreprintService):
csl = preprint_csl(node, node.node)
data = [csl, ]
else:
data = [node.csl, ]
bib_source = CiteProcJSON(data)
custom = CUSTOM_CITATIONS.get(style, False)
path = os.path.join(BASE_PATH, 'static', custom) if custom else os.path.join(CITATION_STYLES_PATH, style)
bib_style = CitationStylesStyle(path, validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
bib = bibliography.bibliography()
cit = unicode(bib[0] if len(bib) else '')
title = csl['title'] if csl else node.csl['title']
if cit.count(title) == 1:
i = cit.index(title)
prefix = clean_up_common_errors(cit[0:i])
suffix = clean_up_common_errors(cit[i + len(title):])
cit = prefix + title + suffix
elif cit.count(title) == 0:
cit = clean_up_common_errors(cit)
return cit
class RFIC2009Paper(Document):
rngschema = 'rfic.rng'
namespace = 'http://www.mos6581.org/ns/rficpaper'
def __init__(self, filename, bibliography_source):
super().__init__(backend=pdf)
self.styles = styles
parser = xml_frontend.Parser(CustomElement, self.namespace,
schema=self.rngschema)
xml_tree = parser.parse(filename)
self.root = xml_tree.getroot()
bibliography_style = CitationStylesStyle('ieee.csl')
self.bibliography = CitationStylesBibliography(bibliography_style,
bibliography_source,
csl_formatter)
authors = [author.text for author in self.root.head.authors.author]
if len(authors) > 1:
self.author = ', '.join(authors[:-1]) + ', and ' + authors[-1]
else:
self.author = authors[0]
self.title = self.root.head.title.text
self.keywords = [term.text for term in self.root.head.indexterms.term]
self.parse_input()
def parse_input(self):
self.title_par = Paragraph(self.title, style='title')
self.author_par = Paragraph(self.author, style='author')
self.affiliation_par = Paragraph(self.root.head.affiliation.text,
style='affiliation')
self.content = Chain(self)
self.content << Abstract(self.root.head.abstract.text)
self.content << IndexTerms(self.keywords)
toc_heading = Heading('Table of Contents', style='unnumbered')
toc = TableOfContents()
self.content << rt.Section([toc_heading, toc])
for element in self.root.body.getchildren():
self.content << element.process()
bib_heading = Heading('References', style='unnumbered')
self.bibliography.sort()
bib = Bibliography(self.bibliography)
self.content << rt.Section([bib_heading, bib])
def setup(self):
self.page_count = 1
page = RFICPage(self, first=True)
self.add_page(page, self.page_count)
page.title_box << self.title_par
page.title_box << self.author_par
page.title_box << self.affiliation_par
def new_page(self, chains):
page = RFICPage(self)
self.page_count += 1
self.add_page(page, self.page_count)
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
reformat_styles = ['apa', 'chicago-author-date', 'modern-language-association']
csl = node.csl
data = [csl, ]
bib_source = CiteProcJSON(data)
custom = CUSTOM_CITATIONS.get(style, False)
path = os.path.join(BASE_PATH, 'static', custom) if custom else os.path.join(CITATION_STYLES_PATH, style)
try:
bib_style = CitationStylesStyle(path, validate=False)
except ValueError:
citation_style = CitationStyle.load(style)
if citation_style is not None and citation_style.has_parent_style:
parent_style = citation_style.parent_style
parent_path = os.path.join(CITATION_STYLES_PATH, parent_style)
bib_style = CitationStylesStyle(parent_path, validate=False)
else:
raise ValueError('Unable to find a dependent or independent parent style related to {}.csl'.format(style))
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
bib = bibliography.bibliography()
cit = unicode(bib[0] if len(bib) else '')
title = csl['title'] if csl else node.csl['title']
title = title.rstrip('.')
if cit.count(title) == 1:
i = cit.index(title)
prefix = clean_up_common_errors(cit[0:i])
suffix = clean_up_common_errors(cit[i + len(title):])
if (style in reformat_styles):
if suffix[0:1] == '.':
suffix = suffix[1:]
if title[-1] != '.':
title += '.'
cit = prefix + title + suffix
elif cit.count(title) == 0:
cit = clean_up_common_errors(cit)
if (style in reformat_styles):
cit = add_period_to_title(cit)
if style == 'apa':
cit = apa_reformat(node, cit)
if style == 'chicago-author-date':
cit = chicago_reformat(node, cit)
if style == 'modern-language-association':
cit = mla_reformat(node, cit)
return cit
def cite(self):
style_path = get_style_filepath(CSL_STYLE)
print(style_path)
bib_style = CitationStylesStyle(style_path, validate=False)
bibliography = CitationStylesBibliography(bib_style,
CiteProcJSON([self.fields]),
formatter.html)
citation = Citation([CitationItem(self.instance.slug)])
bibliography.register(citation)
return str(bibliography.bibliography()[0])
def apa_html_utf8(json_data, alocale='en-US'):
# CiteProcJSON receives a [{}] not a {}
bib_source = CiteProcJSON([json_data])
apa = CitationStylesStyle('apa', locale=alocale, validate=False)
bibliography = CitationStylesBibliography(apa, bib_source, formatter.html)
citation = Citation([CitationItem(json_data['id'])])
bibliography.register(citation)
# handle python weird string type
return str(''.join(bibliography.bibliography()[0]).encode('utf-8'))
def _get_citation(library, document_id, style):
bib_source = CiteProcJSON(library)
bib_style = CitationStylesStyle(style)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
for id in range(0, len(document_id)):
citation = Citation([CitationItem(library[id]['id'])])
bibliography.register(citation)
return bibliography.bibliography()
def make_citation_by_string(publications):
bib_str = render_to_string('semsite/export/publications.bib',
{'publications': publications})
bib = bibtex.BibTeX(io.StringIO(bib_str), encoding='utf-8')
bib_style = CitationStylesStyle('gost-r-7-0-5-2008',
'ru-ru',
validate=False)
bibliography = CitationStylesBibliography(bib_style, bib)
citation_items = [CitationItem(key) for key in bib]
citation = Citation(citation_items)
bibliography.register(citation)
return bibliography.cite(citation, warn)
def __init__(self, filename):
with open(filename, 'rt', encoding='UTF-8') as f:
self.json_data = json.load(f)
csl_io = io.BytesIO(utf_8_encode(self.json_data['csl'])[0])
self.style = CitationStylesStyle(csl_io)
self._fix_input(self.json_data['input'])
self.references = [item['id'] for item in self.json_data['input']]
self.references_dict = CiteProcJSON(self.json_data['input'])
self.bibliography = CitationStylesBibliography(self.style,
self.references_dict)
self.expected = self.json_data['result'].splitlines()
def __init__(self, filename):
self.csl_test = CslTest({}, None, (filename, ),
os.path.basename(filename))
self.csl_test.parse()
csl_io = io.BytesIO(utf_8_encode(self.data['csl'])[0])
self.style = CitationStylesStyle(csl_io)
self._fix_input(self.data['input'])
self.references = [item['id'] for item in self.data['input']]
self.references_dict = CiteProcJSON(self.data['input'])
self.bibliography = CitationStylesBibliography(self.style,
self.references_dict)
self.expected = self.data['result'].splitlines()
def __init__(self,
for_cls,
csl_style,
export_format=adsFormatter.unicode,
journal_format=adsJournalFormat.default):
"""
:param for_cls: input data for this class
:param csl_style: export journal style
:param export_format: export format
"""
self.for_cls = for_cls
self.csl_style = csl_style
self.export_format = export_format
self.journal_format = journal_format
self.citation_item = []
self.bibcode_list = []
self.__update_title()
# Process the JSON data to generate a citaproc-py BibliographySource.
bib_source = CiteProcJSON(self.for_cls)
csl_style_fullpath = os.path.realpath(__file__ + "/../../cslstyles")
# load a CSL style (from the current directory)
bib_style = CitationStylesStyle(os.path.join(csl_style_fullpath + '/' +
csl_style),
validate=False)
# Create the citaproc-py bibliography, passing it the:
# * CitationStylesStyle,
# * BibliographySource (CiteProcJSON in this case), and
# * a formatter (plain, html, or you can write a custom formatter)
# we are going to have CSL format everything using html and then format it as we need to match the
# classic output
self.bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.html)
# Processing citations in a document needs to be done in two passes as for some
# CSL styles, a citation can depend on the order of citations in the
# bibliography and thus on citations following the current one.
# For this reason, we first need to register all citations with the
# CitationStylesBibliography.
for item in self.for_cls:
citation = Citation([CitationItem(item['id'])])
self.citation_item.append(citation)
self.bibliography.register(citation)
# this is actually a bibcode that was passed in, but we have to use
# one of CSLs predefined ids
self.bibcode_list.append(''.join(item.get('locator', '')))
def __init__(self, doc, style):
self.citations = {}
self.doc = doc
style = style.lower()
bib_source = self.convertReferencesToJSON()
bib_source = CiteProcJSON(bib_source)
bib_style = CitationStylesStyle(os.path.join(CSL_PATH, style),
validate=False)
self.bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.html)
self.prepareCitations()
def generate_html(bib_csl_data, style_name='harvard1'):
assert bib_csl_data
bib_source = bib_csl_data
bib_style = CitationStylesStyle(style_name, validate=False)
bibliography = CitationStylesBibliography(
bib_style, bib_source, formatter.html)
bib_cites = [Citation([CitationItem(item)]) for item in bib_source]
for item in bib_cites:
bibliography.register(item)
# FIXME: Can this be avoided?
for item in bib_cites:
bibliography.cite(item, _cite_warn)
out = ''
bibliography.sort()
for item in bibliography.bibliography():
line = '<p>' + ''.join(item) + '</p>\n'
out += line
return out
def generate_md(db, min_year=float('-inf'), max_year=float('inf'), group_by_year=True):
directory = os.path.dirname(os.path.abspath(__file__))
file_location = directory + os.path.sep + "static" + os.path.sep + "SasView_linktitle"
bib_style = CitationStylesStyle(file_location, validate=False)
# Create the citeproc-py bibliography, passing it the:
# * CitationStylesStyle,
# * BibliographySource (CiteProcJSON in this case), and
# * a formatter (plain, html, or you can write a custom formatter)
# add id to each item:
items = db.items()
for k, v in items:
v['id'] = k
bib_source = CiteProcJSON([v for k,v in items])
year_lookup = sort_year(items)
years = list(year_lookup.keys())
years.sort()
output_years = [year for year in years if year <= max_year and year >=min_year]
year_cite_items = []
year_link_items = []
patent_items = []
for year in output_years:
year_link_items.append('[{0}](#{0})'.format(year))
keys = year_lookup[year]
keys.sort(key=lambda l: get_date_string(db[l]), reverse=True)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
citations = [Citation([CitationItem(k)]) for k in keys]
for c in citations:
bibliography.register(c)
bib = bibliography.bibliography()
cite_bib = []
for k, b in zip(keys, bib):
if db[k].get('type', None) == 'patent':
patent_items.append(unescape("".join(b)))
else:
cite_bib.append(b)
bib_output = [unescape("".join(b)) for b in cite_bib]
year_output = ['## {0}\n'.format(year)]
year_output.append('<small>[top](#acknowledgements-and-contacts)</small>\n')
year_output.append('---\n')
for i, bib_i in enumerate(bib_output):
bib_final = ("{0}. " + bib_i).format(i)
year_output.append(bib_final)
year_cite_items.append("\n".join(year_output))
return {"citations": year_cite_items, "links": year_link_items, "patents": patent_items}
def serialize(self, pid, record, links_factory=None, **kwargs):
"""Serialize a single record.
:param pid: Persistent identifier instance.
:param record: Record instance.
:param links_factory: Factory function for record links.
"""
data = self.serializer.serialize(pid, record, links_factory)
source = self._get_source(data)
style = CitationStylesStyle(validate=False, **self._get_args(**kwargs))
bib = CitationStylesBibliography(style, source, formatter.plain)
citation = Citation([CitationItem(pid.pid_value)])
bib.register(citation)
return self._clean_result(''.join(bib.bibliography()[0]))
def serialize(self, pid, record, links_factory=None, **kwargs):
"""Serialize a single record.
:param pid: Persistent identifier instance.
:param record: Record instance.
:param links_factory: Factory function for record links.
"""
data = self.serializer.serialize(pid, record, links_factory)
source = self._get_source(data)
style = CitationStylesStyle(validate=False, **self._get_args(**kwargs))
bib = CitationStylesBibliography(style, source, formatter.plain)
citation = Citation([CitationItem(pid.pid_value)])
bib.register(citation)
return self._clean_result(''.join(bib.bibliography()[0]))
def bibToCite(bibfile, ID):
bib_source = BibTeX(bibfile)
style_path = get_style_filepath('modern-language-association')
bib_style = CitationStylesStyle(style_path, validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.plain)
citation1 = Citation([CitationItem(ID)])
bibliography.register(citation1)
with open('out.txt', 'w') as f:
for item in bibliography.bibliography():
print(str(item), file=f)
class Bibliography:
def __init__(self, bib_file, csl_file):
self.bib_file = bib_file
self.csl_file = csl_file
self.bibtex = BibTeX(bib_file)
self.style = CitationStylesStyle(csl_file, validate=False)
self.bibliography = CitationStylesBibliography(bib_style, bib_source,formatter.plain)
def citation(self, pandocname):
c = Citation([CitationItem('whole-collection')])
self.bibliography.register(c)
return self.bibliography.cite(c)
def bibliography(self):
for item in self.bibliography.bibliography():
str(item)
def to_CSL(project_data, citation_style, formatter_Style):
"""Format an OSF project as a citation.
:param project_data: JSON-style dictionary of project information
:param citation_style: File name of citation xml document
:param formatter: Citeproc formatter (e.g. formatter.plain, formatter.html)
"""
bib_source = CiteProcJSON([project_data])
bib_style = CitationStylesStyle(citation_style)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter_Style)
citation1 = Citation([CitationItem(project_data['id'])])
bibliography.register(citation1)
return bibliography.bibliography()
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
csl = None
if isinstance(node, PreprintService):
csl = preprint_csl(node, node.node)
data = [csl, ]
else:
data = [node.csl, ]
bib_source = CiteProcJSON(data)
custom = CUSTOM_CITATIONS.get(style, False)
path = os.path.join(BASE_PATH, 'static', custom) if custom else os.path.join(CITATION_STYLES_PATH, style)
bib_style = CitationStylesStyle(path, validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
bib = bibliography.bibliography()
cit = unicode(bib[0] if len(bib) else '')
title = csl['title'] if csl else node.csl['title']
if cit.count(title) == 1:
i = cit.index(title)
prefix = clean_up_common_errors(cit[0:i])
suffix = clean_up_common_errors(cit[i + len(title):])
cit = prefix + title + suffix
elif cit.count(title) == 0:
cit = clean_up_common_errors(cit)
if isinstance(node, PreprintService):
cit_node = node.node
else:
cit_node = node
if style == 'apa':
cit = apa_reformat(cit_node, cit)
if style == 'chicago-author-date':
cit = chicago_reformat(cit_node, cit)
if style == 'modern-language-association':
cit = mla_reformat(cit_node, cit)
return cit
def get_citation_string(json, id, style, locale):
"""Get the citation string from CiteProc library."""
def _clean_result(text):
"""Remove double spaces, punctuation."""
text = re.sub(r"\s\s+", " ", text)
text = re.sub(r"\.\.+", ".", text)
return text
source = CiteProcJSON([json])
citation_style = CitationStylesStyle(validate=False,
style=style,
locale=locale)
bib = CitationStylesBibliography(citation_style, source, formatter.plain)
citation = Citation([CitationItem(id)])
bib.register(citation)
return _clean_result(str(bib.bibliography()[0]))
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
if isinstance(node, Node):
data = [
node.csl,
]
elif isinstance(node, PreprintService):
csl = preprint_csl(node, node.node)
data = [
csl,
]
else:
raise ValueError
bib_source = CiteProcJSON(data)
bib_style = CitationStylesStyle(os.path.join(CITATION_STYLES_PATH, style),
validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
def warn(citation_item):
pass
bibliography.cite(citation, warn)
bib = bibliography.bibliography()
return unicode(bib[0] if len(bib) else '')
def display_citation(bibtex_metadata, bib_stylename, formatter=formatter.html):
# valid style names: plos, apa, pnas, nature, bmj, harvard1
# full list is here: https://github.com/citation-style-language/styles
bib_style = EnhancedCitationStyle(bib_stylename)
bibliography = CitationStylesBibliography(
bib_style, bibtex_metadata, formatter) #could be formatter.html
id = "ITEM-1"
citation = Citation([CitationItem(id)])
bibliography.register(citation)
citation_parts = u"".join(bibliography.bibliography()[0])
citation_text = u"".join(citation_parts)
html_parser = HTMLParser()
citation_text = html_parser.unescape(citation_text)
return citation_text
def ama_html(self):
if not self.citeproc_json:
return ""
bibliography = CitationStylesBibliography(
CitationStylesStyle(
str(
Path(__file__).parent / "styles" /
"american-medical-association-no-url.csl")),
self.bib_source,
formatter.html,
)
bibliography.register(Citation([CitationItem(self.doi)]))
# The bibliography only contains 1 element
citation = str(bibliography.bibliography()[0])
citation = re.sub(r"^1\. ", "", citation)
return clean(citation)
def __init__(self, filename, bibliography_source):
super().__init__(backend=pdf)
parser = xml_frontend.Parser(CustomElement,
self.namespace,
schema=self.rngschema)
xml_tree = parser.parse(filename)
self.root = xml_tree.getroot()
bibliography_style = CitationStylesStyle('ieee.csl')
self.bibliography = CitationStylesBibliography(bibliography_style,
bibliography_source,
csl_formatter)
authors = [author.text for author in self.root.head.authors.author]
if len(authors) > 1:
self.author = ', '.join(authors[:-1]) + ', and ' + authors[-1]
else:
self.author = authors[0]
self.title = self.root.head.title.text
self.keywords = [term.text for term in self.root.head.indexterms.term]
self.parse_input()
def __init__(self, doc, style):
self.citations={}
self.doc=doc
style=style.lower()
bib_source = self.convertReferencesToJSON()
bib_source = CiteProcJSON(bib_source)
bib_style = CitationStylesStyle(os.path.join(CSL_PATH,style), validate=False)
self.bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.html)
self.prepareCitations()
def __init__(self, filename):
self.csl_test = CslTest({}, None, (filename, ),
os.path.basename(filename))
self.csl_test.parse()
csl_io = io.BytesIO(utf_8_encode(self.data['csl'])[0])
self.style = CitationStylesStyle(csl_io)
self._fix_input(self.data['input'])
self.references = [item['id'] for item in self.data['input']]
self.references_dict = CiteProcJSON(self.data['input'])
self.bibliography = CitationStylesBibliography(self.style,
self.references_dict)
self.expected = self.data['result'].splitlines()
def __init__(self, filename):
with open(filename, encoding='UTF-8') as f:
self.json_data = json.load(f)
csl_io = io.BytesIO(utf_8_encode(self.json_data['csl'])[0])
self.style = CitationStylesStyle(csl_io)
self._fix_input(self.json_data['input'])
self.references = [item['id'] for item in self.json_data['input']]
self.references_dict = CiteProcJSON(self.json_data['input'])
self.bibliography = CitationStylesBibliography(self.style,
self.references_dict)
self.expected = self.json_data['result'].splitlines()
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
if isinstance(node, Node):
data = [node.csl, ]
elif isinstance(node, PreprintService):
csl = preprint_csl(node, node.node)
data = [csl, ]
else:
raise ValueError
bib_source = CiteProcJSON(data)
bib_style = CitationStylesStyle(os.path.join(CITATION_STYLES_PATH, style), validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
def warn(citation_item):
pass
bibliography.cite(citation, warn)
bib = bibliography.bibliography()
return unicode(bib[0] if len(bib) else '')
def print_using_citeproc(csl_json, keys, style):
from citeproc import CitationStylesStyle, CitationStylesBibliography
from citeproc import Citation, CitationItem
from citeproc import formatter
from citeproc.source.json import CiteProcJSON
def warn(citation_item):
raise RuntimeError(
"Reference with key '{}' not found".format(citation_item.key)
)
bib_source = CiteProcJSON([csl_json])
bib_style = CitationStylesStyle(style, validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.html)
citations = []
# the following lines just do whatever example in citeproc repo does
for key in keys:
citation = Citation([CitationItem(key)])
bibliography.register(citation)
citations.append(citation)
for citation in citations:
# unused = bibliography.cite(citation, warn_missing_key)
unused = bibliography.cite(citation, warn)
for item in bibliography.bibliography():
print(str(item))
def citeproc_csl(csl_json: Dict[str, Any],
style: str,
html: bool = False) -> str:
"""
Renders a release entity to a styled citation.
Notable styles include:
- 'csl-json': special case to JSON encode the structured CSL object (via
release_to_csl())
- bibtext: multi-line bibtext format (used with LaTeX)
Returns a string; if the html flag is set, and the style isn't 'csl-json'
or 'bibtex', it will be HTML. Otherwise plain text.
"""
if not csl_json.get("id"):
csl_json["id"] = "unknown"
if style == "csl-json":
return json.dumps(csl_json)
bib_src = CiteProcJSON([csl_json])
form = formatter.plain
if html:
form = formatter.html
style_path = get_style_filepath(style)
bib_style = CitationStylesStyle(style_path, validate=False)
bib = CitationStylesBibliography(bib_style, bib_src, form)
bib.register(Citation([CitationItem(csl_json["id"])]))
lines = bib.bibliography()[0]
if style == "bibtex":
out = ""
for line in lines:
if line.startswith(" @"):
out += "@"
elif line.startswith(" "):
out += "\n " + line
else:
out += line
return "".join(out)
else:
return "".join(lines)
def apply(self):
"""Process Bib and Csl nodes, set up bibliography opbects, register and then cite the
in-text citations, and then construct the bibliography."""
# Find Bib and Csl nodes and extract the relevant information from them.
bibnodes = [bib for bib in self.document.traverse() if isinstance(bib, nodes.Bib)]
if bibnodes:
self.bib_source = bibnodes[0]["bibfile"]
cslnodes = [csl for csl in self.document.traverse() if isinstance(csl, nodes.Csl)]
if cslnodes:
self.bib_style = cslnodes[0]["cslfile"]
# Remove the extraneous nodes now we're done with them.
for node in bibnodes + cslnodes:
node.parent.remove(node)
citations = [cite for cite in self.document.traverse() \
if isinstance(cite, nodes.citation_reference)]
if citations:
# Set up the bibliography objects.
self.bib_source = os.path.join(self.cwd, self.bib_source + \
(".bib" if not self.bib_source.endswith('.bib') else ''))
self.bib_source = BibTeX(self.bib_source)
self.bib_style = os.path.join(self.cwd, self.bib_style + \
(".csl" if not self.bib_style.endswith('.csl') else ''))
self.bib_style = CitationStylesStyle(self.bib_style, validate=False)
self.bibliography = CitationStylesBibliography(self.bib_style,
self.bib_source, formatter)
# Register each in-text citation, then cite each one.
for citation in citations:
self.register_item(citation)
for citation in citations:
self.cite_item(citation)
# Find the places where a Bibliography has been requested.
bnodes = [bib for bib in self.document.traverse() \
if isinstance(bib, nodes.Bibliography)]
self.document.b = self.bibliography
biblist = nodes.enumerated_list()
bibtexts = self.bibliography.bibliography()
for btext in bibtexts:
list_item = nodes.list_item()
item_node = publish_doctree(escape(str(btext)))[0]
list_item.append(item_node)
biblist.append(list_item)
for bnode in bnodes:
bnode.parent.replace(bnode, biblist)
def parse_zotero_bibtex(items, citation_style_path):
# separate the bibtex items and extract the key
p = re.compile(ur'{([^,]+)')
bibtexItems = []
for item in items.split('@'):
if len(item) > 1:
# make a tuple containing bibtex key and bibtex string
# splitting removed the @ of each entry so we need to add it back to get valid bibtex
bibtexItems.append((re.search(p, item).group(1), "@" + item))
# citeproc_py expects a bibtex file for input so we cannot just pass the bibtex string,
# so we need to write it into a virtual file using StringIO
output = StringIO.StringIO()
output.write(items)
output.seek(0)
itemDicts = []
try:
# initialize citeproc-py
bib_source = BibTeX(output)
bib_style = CitationStylesStyle(citation_style_path, validate=False)
# parse each bibtex entry into a html formatted bibliography entry
# instead of generating a bibliography of all entries at once, create a
# bibliography for each of the single items to retaing mapping to the
# corresponding bibtex string
for (key, bibtex) in bibtexItems:
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.html)
citationItem = CitationItem(key)
citation = Citation([citationItem])
bibliography.register(citation)
bibliography.cite(citation, warning)
renderedItem = ""
# the rendered bibliography item returned by citeproc-py is split
# inton an array of separate tokens, so we need to join them together
for token in bibliography.bibliography()[0]:
renderedItem += token
itemDicts.append({
'key': key,
'bibtex': bibtex,
'html': renderedItem
})
except:
warning("ERROR PARSING")
traceback.print_exc(file=sys.stderr)
finally:
output.close()
return itemDicts
def run_citeproc_releases(args):
for line in args.json_input:
line = line.strip()
if not line:
continue
entity = entity_from_json(line, ReleaseEntity, api_client=args.api.api_client)
csl_json = release_to_csl(entity)
# XXX:
csl_json['id'] = "release:" + (entity.ident or "unknown")
if args.style == "csl-json":
args.json_output.write(json.dumps(csl_json) + "\n")
continue
bib_src = CiteProcJSON([csl_json])
form = formatter.plain
if args.html:
form = formatter.html
style_path = get_style_filepath(args.style)
bib_style = CitationStylesStyle(style_path, validate=False)
bib = CitationStylesBibliography(bib_style, bib_src, form)
bib.register(Citation([CitationItem(csl_json['id'])]))
# XXX:
#args.json_output.write(
# json.dumps(release_to_csl(entity)) + '\n')
lines = bib.bibliography()[0]
if args.style == "bibtex":
for l in lines:
if l.startswith(" @"):
args.json_output.write("\n@")
elif l.startswith(" "):
#print("line: START|{}|END".format(l))
args.json_output.write("\n " + l)
else:
args.json_output.write(l)
else:
args.json_output.write(''.join(lines) + "\n")
print()
def __init__(self, filename, bibliography_source):
super().__init__(backend=pdf)
parser = xml_frontend.Parser(CustomElement, self.namespace,
schema=self.rngschema)
xml_tree = parser.parse(filename)
self.root = xml_tree.getroot()
bibliography_style = CitationStylesStyle('ieee.csl')
self.bibliography = CitationStylesBibliography(bibliography_style,
bibliography_source,
csl_formatter)
authors = [author.text for author in self.root.head.authors.author]
if len(authors) > 1:
self.author = ', '.join(authors[:-1]) + ', and ' + authors[-1]
else:
self.author = authors[0]
self.title = self.root.head.title.text
self.keywords = [term.text for term in self.root.head.indexterms.term]
self.parse_input()
def get_bib():
bib_source = BibTeX('notebooks/bibliography.bib', encoding='utf8')
bib_style = CitationStylesStyle('notebooks/springer.csl', validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.html)
bib_cites = [Citation([CitationItem(item)]) for item in bib_source]
for item in bib_cites:
bibliography.register(item)
for item in bib_cites:
bibliography.cite(item, _cite_warn)
num = len(bibliography.keys)
bib_entries = dict()
for i in range(num):
bib = ''.join(bibliography.bibliography()[i])
# remove beginning digits and \. from bib entries
bib = '{}. ' + re.sub("^\d+\.", "", bib)
bib_entries[bibliography.keys[i]] = bib
return bib_entries
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
if isinstance(node, Node):
data = [node.csl, ]
elif isinstance(node, PreprintService):
csl = preprint_csl(node, node.node)
data = [csl, ]
else:
raise ValueError
bib_source = CiteProcJSON(data)
bib_style = CitationStylesStyle(os.path.join(CITATION_STYLES_PATH, style), validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
def warn(citation_item):
pass
bibliography.cite(citation, warn)
bib = bibliography.bibliography()
if len(bib):
doi = data[0].get('DOI')
if style == 'apa':
first_segment = [list(bib[0])[0][:-2]]
return ''.join(first_segment + list(bib[0])[1:13]) if doi else ''.join(first_segment + list(bib[0])[1:12] + list(bib[0])[13:])
elif style == 'modern-language-association':
return ''.join(list(bib[0])[:4] + ['.'] + list(bib[0])[4:5] + list(bib[0])[6:-2])
elif style == 'chicago-author-date':
return ''.join(list(bib[0])[0:3] + ['.'] + list(bib[0])[3:4] + [' '] + list(bib[0])[5:])
else:
return unicode(bib[0])
else:
return ''
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
data = [node.csl, ]
bib_source = CiteProcJSON(data)
bib_style = CitationStylesStyle(os.path.join(CITATION_STYLES_PATH, style), validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
def warn(citation_item):
pass
bibliography.cite(citation, warn)
bib = bibliography.bibliography()
return unicode(bib[0] if len(bib) else '')
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
data = [
node.csl,
]
bib_source = CiteProcJSON(data)
bib_style = CitationStylesStyle(os.path.join(CITATION_STYLES_PATH, style),
validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
def warn(citation_item):
pass
bibliography.cite(citation, warn)
bib = bibliography.bibliography()
return unicode(bib[0] if len(bib) else '')
class Transform(docutils.transforms.Transform):
"""The PyCite Transform handles the recognition and processing of bib-backed citations"""
default_priority = 208
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.cwd, _ = os.path.split(self.document.transformer.components['input'].source_path)
self.bib_source = "bibliography"
self.bib_style = "apa"
self.bibliography = None
self.citations = {}
def register_item(self, item):
"""Registers the citation with the Bibliography"""
citation = Citation([CitationItem(item["refname"])])
self.citations[item["refname"]] = citation
self.bibliography.register(citation)
def cite_item(self, item):
"""Inserts the in-text citation"""
try:
i = publish_doctree(escape(self.bibliography.cite(self.citations[item["refname"]],
fail)))[0][0]
except ValueError:
return
item.parent.replace(item, i)
def apply(self):
"""Process Bib and Csl nodes, set up bibliography opbects, register and then cite the
in-text citations, and then construct the bibliography."""
# Find Bib and Csl nodes and extract the relevant information from them.
bibnodes = [bib for bib in self.document.traverse() if isinstance(bib, nodes.Bib)]
if bibnodes:
self.bib_source = bibnodes[0]["bibfile"]
cslnodes = [csl for csl in self.document.traverse() if isinstance(csl, nodes.Csl)]
if cslnodes:
self.bib_style = cslnodes[0]["cslfile"]
# Remove the extraneous nodes now we're done with them.
for node in bibnodes + cslnodes:
node.parent.remove(node)
citations = [cite for cite in self.document.traverse() \
if isinstance(cite, nodes.citation_reference)]
if citations:
# Set up the bibliography objects.
self.bib_source = os.path.join(self.cwd, self.bib_source + \
(".bib" if not self.bib_source.endswith('.bib') else ''))
self.bib_source = BibTeX(self.bib_source)
self.bib_style = os.path.join(self.cwd, self.bib_style + \
(".csl" if not self.bib_style.endswith('.csl') else ''))
self.bib_style = CitationStylesStyle(self.bib_style, validate=False)
self.bibliography = CitationStylesBibliography(self.bib_style,
self.bib_source, formatter)
# Register each in-text citation, then cite each one.
for citation in citations:
self.register_item(citation)
for citation in citations:
self.cite_item(citation)
# Find the places where a Bibliography has been requested.
bnodes = [bib for bib in self.document.traverse() \
if isinstance(bib, nodes.Bibliography)]
self.document.b = self.bibliography
biblist = nodes.enumerated_list()
bibtexts = self.bibliography.bibliography()
for btext in bibtexts:
list_item = nodes.list_item()
item_node = publish_doctree(escape(str(btext)))[0]
list_item.append(item_node)
biblist.append(list_item)
for bnode in bnodes:
bnode.parent.replace(bnode, biblist)
class ProcessorTest(object):
def __init__(self, filename):
with open(filename, encoding='UTF-8') as f:
self.json_data = json.load(f)
csl_io = io.BytesIO(utf_8_encode(self.json_data['csl'])[0])
self.style = CitationStylesStyle(csl_io)
self._fix_input(self.json_data['input'])
self.references = [item['id'] for item in self.json_data['input']]
self.references_dict = CiteProcJSON(self.json_data['input'])
self.bibliography = CitationStylesBibliography(self.style,
self.references_dict)
self.expected = self.json_data['result'].splitlines()
@staticmethod
def _fix_input(input_data):
for i, ref in enumerate(input_data):
if 'id' not in ref:
ref['id'] = i
if 'type' not in ref:
ref['type'] = 'book'
def execute(self):
if self.json_data['citation_items']:
citations = [self.parse_citation(item)
for item in self.json_data['citation_items']]
elif self.json_data['citations']:
citations = []
for cit in self.json_data['citations']:
cit = cit[0]
citation_items = [self.parse_citation_item(cititem)
for cititem in cit['citationItems']]
citation = Citation(citation_items)
citation.key = cit['citationID']
citation.note_index = cit['properties']['noteIndex']
citations.append(citation)
elif self.json_data['bibentries']:
citation_items = [self.parse_citation_item({'id': entry})
for entry in self.json_data['bibentries'][-1]]
citations = [Citation(citation_items)]
else:
citation_items = [self.parse_citation_item({'id': ref})
for ref in self.references]
citations = [Citation(citation_items)]
for citation in citations:
self.bibliography.register(citation)
if self.style.has_bibliography():
self.bibliography.sort()
results = []
do_nothing = lambda x: None # callback passed to cite()
if self.json_data['mode'] == 'citation':
if self.json_data['citations']:
for i, citation in enumerate(citations):
if i == len(citations) - 1:
dots_or_other = '>>'
else:
dots_or_other = '..'
results.append('{}[{}] '.format(dots_or_other, i) +
self.bibliography.cite(citation, do_nothing))
else:
for citation in citations:
results.append(self.bibliography.cite(citation, do_nothing))
elif self.json_data['mode'] in ('bibliography', 'bibliography-nosort'):
results.append(self.bibliography.bibliography())
return results
def parse_citation(self, citation_data):
citation_items = []
for item in citation_data:
citation_item = self.parse_citation_item(item)
citation_items.append(citation_item)
return Citation(citation_items)
def parse_citation_item(self, citation_item_data):
options = {}
for key, value in citation_item_data.items():
python_key = key.replace('-', '_')
if python_key == 'id':
reference_key = str(value)
continue
elif python_key == 'locator':
try:
options['locator'] = Locator(citation_item_data['label'],
value)
except KeyError:
# some tests don't specify the label
options['locator'] = Locator('page', value)
elif python_key == 'label':
pass
else:
options[python_key] = value
return CitationItem(reference_key, **options)
# Parse the BibTeX database.
bib_source = BibTeX('xampl.bib')
# load a CSL style (from the current directory)
bib_style = CitationStylesStyle('harvard1', validate=False)
# Create the citeproc-py bibliography, passing it the:
# * CitationStylesStyle,
# * BibliographySource (BibTeX in this case), and
# * a formatter (plain, html, or you can write a custom formatter)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.html)
# Processing citations in a document needs to be done in two passes as for some
# CSL styles, a citation can depend on the order of citations in the
# bibliography and thus on citations following the current one.
# For this reason, we first need to register all citations with the
# CitationStylesBibliography.
citation1 = Citation([CitationItem('whole-collection')])
citation2 = Citation([CitationItem('whole-set'), CitationItem('misc-full')])
citation3 = Citation([CitationItem('techreport-full')])
citation4 = Citation([CitationItem('mastersthesis-minimal')])
citation5 = Citation([CitationItem('inproceedings-full'),
CitationItem('unpublished-full')])
# Parse the BibTeX database.
bib_source = BibTeX('xampl.bib')
# load a CSL style (from the current directory)
bib_style = CitationStylesStyle('harvard1.csl')
# Create the citeproc-py bibliography, passing it the:
# * CitationStylesStyle,
# * BibliographySource (BibTeX in this case), and
# * a formatter (plain, html, or you can write a custom formatter)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.plain)
# Processing citations in a document need to be done in two passes as for some
# CSL styles, a citation can depend on the order of citations in the
# bibliography and thus on citations following the current one.
# For this reason, we first need to register all citations with the
# CitationStylesBibliography.
citation1 = Citation([CitationItem('whole-collection')])
citation2 = Citation([CitationItem('whole-set'), CitationItem('misc-full')])
citation3 = Citation([CitationItem('techreport-minimal')])
bibliography.register(citation1)
bibliography.register(citation2)
bibliography.register(citation3)
# Parse the BibTeX database.
bib_source = BibTeX('xampl.bib')
# load a CSL style (from the current directory)
bib_style = CitationStylesStyle('harvard1', validate=False)
# Create the citeproc-py bibliography, passing it the:
# * CitationStylesStyle,
# * BibliographySource (BibTeX in this case), and
# * a formatter (plain, html, or you can write a custom formatter)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.plain)
# Processing citations in a document needs to be done in two passes as for some
# CSL styles, a citation can depend on the order of citations in the
# bibliography and thus on citations following the current one.
# For this reason, we first need to register all citations with the
# CitationStylesBibliography.
citation1 = Citation([CitationItem('whole-collection')])
citation2 = Citation([CitationItem('whole-set'), CitationItem('misc-full')])
citation3 = Citation([CitationItem('techreport-full')])
citation4 = Citation([CitationItem('mastersthesis-minimal')])
citation5 = Citation([CitationItem('inproceedings-full'),
CitationItem('unpublished-full')])
parse_file.closed
def write_to_txt(filename, bibliography):
i=1
f = open(filename,'w', encoding='utf8')
for item in bibliography.bibliography():
if(numbering_txt):
f.write(unicode(str(i))+'. ')
i=i+1
f.write(unicode(str(item))+'\n')
f.close()
def write_to_bbl(filename, bibliography):
f = open(filename,'w', encoding='utf8')
f.write('\\begin{thebibliography}{}\n')
i=0
for item in bibliography.bibliography():
f.write('\\bibitem{'+my_cities[i]+'}')
f.write(unicode(str(item))+'\n')
i=i+1
f.write('\\end{thebibliography}')
f.close()
bib_source = BibTeX('literature.bib', encoding='utf8')
bib_style = CitationStylesStyle('gost-r-7-0-5-2008', locale='ru-RU', validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source, formatter.plain)
get_cities('literature.bib')
for i in range(len(my_cities)):
bibliography.register(Citation([CitationItem(my_cities[i])]))
write_to_txt('my_bibliography.txt',bibliography)
write_to_bbl('my_bibliography.bbl',bibliography)
def bib(style_path,
ds,
return_cites_and_keys = False,
formatter = "chocolate",
dumb_quotes = True,
# Turning this off won't educate any straight quotes in
# the data, but leaving it on will stupefy all the
# smart quotes in the output.
apa_tweaks = True,
# The below options are ignored unless apa_tweaks is on.
always_include_issue = False,
include_isbn = False,
url_after_doi = False,
publisher_website = True,
abbreviate_given_names = True):
if isinstance(formatter, str):
try: formatter = formatter_from_name[formatter]
except KeyError: raise ValueError('Unknown formatter "{}"'.format(formatter))
style = get_style(style_path, apa_tweaks,
include_isbn, url_after_doi, abbreviate_given_names)
ds = deepcopy(ds)
if apa_tweaks:
# Distinguish entries that would have identical authors and years
# by adding suffixes to the years.
# Group works by author and year.
ay = defaultdict(list)
for d in ds:
k = repr(d.get('author') or d.get('editor')) + '/' + str(d['issued']['date-parts'][0][0])
if not any(d is v for v in ay[k]):
ay[k].append(d)
# If any group has more than one element, add suffixes.
for v in ay.values():
if len(v) > 1:
for i, d in enumerate(sorted(v, key = title_sort_key)):
d['year_suffix'] = ascii_lowercase[i]
for d in ds:
if 'id' not in d:
d['id'] = str(random())
for k in list(d.keys()):
if d[k] is None: del d[k]
if apa_tweaks:
# By default, don't include the issue number for
# journal articles.
if not always_include_issue and d['type'] == 'article-journal':
delf(d, 'issue')
# Use the weird "Retrieved from Dewey, Cheatem, &
# Howe website: http://example.com" format prescribed
# for reports.
if publisher_website and d['type'] == 'report' and 'publisher' in d and 'URL' in d:
d['URL'] = '{} website: {}'.format(
d.pop('publisher'), d['URL'])
# Add structure words for presentations and include
# the event place.
if d['type'] == 'speech' and d['genre'] == 'paper':
d['event'] = 'meeting of the {}, {}'.format(
d.pop('publisher'), d['event-place'])
if d['type'] == 'speech' and d['genre'] == 'video':
d['medium'] = 'Video file'
del d['genre']
# When abbreviating given names, remove hyphens
# preceding lowercase letters. Otherwise, weird
# stuff happens.
if abbreviate_given_names and 'author' in d:
for a in d['author']:
if 'given' in a:
a['given'] = sub(
'-(.)',
lambda mo:
("" if mo.group(1).islower() else "-") +
mo.group(1),
a['given'])
# Abbreviate a long list of authors with an ellipsis
# and the final author.
if 'author' in d and len(d['author']) > 7:
d['author'] = (
d['author'][0:6] +
[{'given': '', 'family': '⣥<ellipsis>⣥'}] +
d['author'][-1:])
bibliography = CitationStylesBibliography(
style,
{ref.key: ref for ref in parse_references(ds)},
formatter)
cites = [ Citation([CitationItem(d['id'])]) for d in ds ]
for c in cites: bibliography.register(c)
def sort_key_f(item):
ref = item.reference
names = [(name['family'].lower(), name['given'][0].lower() if 'given' in name else '')
for name in ref.get('author') or ref.get('editor')
if name.family != '⣥<ellipsis>⣥']
return (names, ref['issued']['year'],
title_sort_key(ref),
ref['page']['first'] if 'page' in ref else '')
if len(ds) > 1:
# Sort the bibliography
# bibliography.sort() # Doesn't appear to handle leading "the"s correctly.
bibliography.items = sorted(bibliography.items, key = sort_key_f)
bibliography.keys = [item.key for item in bibliography.items]
bibl = bibliography.bibliography()
for i, s in enumerate(bibl):
s = ''.join(s)
# Fix spacing and punctuation issues.
s = s.replace(' ', ' ')
s = sub(r'([.!?…])\.', r'\1', s)
if dumb_quotes:
s = s.replace('‘', "'").replace('’', "'").replace('“', '"').replace('”', '"')
if apa_tweaks:
if formatter is citeproc.formatter.html or formatter is chocolate:
# Italicize the stuff between a journal name and a volume
# number.
s = sub(r'</i>, <i>(\d)', r', \1', s)
# Make "p." into "pp." when more than one page is cited.
s = sub(r'(\W)p\. (\S+[,–])', r'\1pp. \2', s)
# Replace the ellipsis placeholder.
s = s.replace('⣥<ellipsis>⣥, ., &', '…')
bibl[i] = s
if return_cites_and_keys:
fcites = [bibliography.cite(c, lambda x: None) for c in cites]
return (fcites, bibliography.keys, bibl)
else:
return bibl
meta = doc[0]['unMeta']
result = meta.get('bibliography', {})
if result:
bibliography_path = value_of_metadata(result)
result = meta.get('csl', {})
if result:
csl_path = value_of_metadata(result)
if bibliography_path == None or csl_path == None:
raise Exception('Metadata variables must be set for both bibliography and csl.')
# Parse the BibTeX database.
bib_source = BibTeX(bibliography_path, encoding='utf-8')
# load a CSL style
bib_style = CitationStylesStyle(csl_path, validate=False)
bibliography = CitationStylesBibliography(bib_style, bib_source, f)
altered = walk(doc, citation_register, format, doc[0]['unMeta'])
second = walk(altered, citation_replace, format, doc[0]['unMeta'])
references = []
for item, key in zip(bibliography.bibliography(), bibliography.keys):
attrs = {'id': key, 'class':'h-cite'}
references.append(Div(attributes(attrs),[Para([render(str(item))])]))
second[1].extend(references) # add more paragraphs to the end of the main document list of blocks
json.dump(second, sys.stdout)
class RFIC2009Paper(Document):
rngschema = 'rfic.rng'
namespace = 'http://www.mos6581.org/ns/rficpaper'
def __init__(self, filename, bibliography_source):
super().__init__(backend=pdf)
parser = xml_frontend.Parser(CustomElement, self.namespace,
schema=self.rngschema)
xml_tree = parser.parse(filename)
self.root = xml_tree.getroot()
bibliography_style = CitationStylesStyle('ieee.csl')
self.bibliography = CitationStylesBibliography(bibliography_style,
bibliography_source,
csl_formatter)
authors = [author.text for author in self.root.head.authors.author]
if len(authors) > 1:
self.author = ', '.join(authors[:-1]) + ', and ' + authors[-1]
else:
self.author = authors[0]
self.title = self.root.head.title.text
self.keywords = [term.text for term in self.root.head.indexterms.term]
self.parse_input()
def parse_input(self):
self.title_par = Paragraph(self.title, style=styles['title'])
self.author_par = Paragraph(self.author, style=styles['author'])
self.affiliation_par = Paragraph(self.root.head.affiliation.text,
style=styles['affiliation'])
self.content = Chain(self)
self.content << Abstract(self.root.head.abstract.text)
self.content << IndexTerms(self.keywords)
self.content << Heading(self, 'Table of Contents',
style=styles['unnumbered'], level=1)
toc = TableOfContents(style=styles['toc'],
styles=[styles['toc1'], styles['toc2'],
styles['toc3']])
self.content << toc
for section in self.root.body.section:
for flowable in section.process(self):
toc.register(flowable)
self.content << flowable
try:
for flowable in self.root.body.acknowledgement.process(self):
toc.register(flowable)
self.content << flowable
except AttributeError:
pass
self.content << Heading(self, 'References', style=styles['unnumbered'],
level=1)
self.bibliography.sort()
self.content << self.bibliography.bibliography()
def setup(self):
self.page_count = 1
page = RFICPage(self, first=True)
self.add_page(page, self.page_count)
page.title_box << self.title_par
page.title_box << self.author_par
page.title_box << self.affiliation_par
def new_page(self, chains):
page = RFICPage(self)
self.page_count += 1
self.add_page(page, self.page_count)
def render_citation(node, style='apa'):
"""Given a node, return a citation"""
reformat_styles = [
'apa', 'chicago-author-date', 'modern-language-association'
]
csl = None
if isinstance(node, PreprintService):
csl = preprint_csl(node, node.node)
data = [
csl,
]
else:
data = [
node.csl,
]
bib_source = CiteProcJSON(data)
custom = CUSTOM_CITATIONS.get(style, False)
path = os.path.join(BASE_PATH, 'static',
custom) if custom else os.path.join(
CITATION_STYLES_PATH, style)
try:
bib_style = CitationStylesStyle(path, validate=False)
except ValueError:
citation_style = CitationStyle.load(style)
if citation_style is not None and citation_style.has_parent_style:
parent_style = citation_style.parent_style
parent_path = os.path.join(CITATION_STYLES_PATH, parent_style)
bib_style = CitationStylesStyle(parent_path, validate=False)
else:
raise ValueError(
'Unable to find a dependent or independent parent style related to {}.csl'
.format(style))
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.plain)
citation = Citation([CitationItem(node._id)])
bibliography.register(citation)
bib = bibliography.bibliography()
cit = unicode(bib[0] if len(bib) else '')
title = csl['title'] if csl else node.csl['title']
title = title.rstrip('.')
if cit.count(title) == 1:
i = cit.index(title)
prefix = clean_up_common_errors(cit[0:i])
suffix = clean_up_common_errors(cit[i + len(title):])
if (style in reformat_styles):
if suffix[0:1] == '.':
suffix = suffix[1:]
if title[-1] != '.':
title += '.'
cit = prefix + title + suffix
elif cit.count(title) == 0:
cit = clean_up_common_errors(cit)
if (style in reformat_styles):
cit = add_period_to_title(cit)
if isinstance(node, PreprintService):
cit_node = node.node
else:
cit_node = node
if style == 'apa':
cit = apa_reformat(cit_node, cit)
if style == 'chicago-author-date':
cit = chicago_reformat(cit_node, cit)
if style == 'modern-language-association':
cit = mla_reformat(cit_node, cit)
return cit
class ProcessorTest(object):
"""Parses atest fixture and provides a method for processing the tests
defined in it."""
bib_prefix = '<div class="csl-bib-body">'
bib_suffix = '</div>'
item_prefix = ' <div class="csl-entry">'
item_suffix = '</div>'
def __init__(self, filename):
self.csl_test = CslTest({}, None, (filename, ),
os.path.basename(filename))
self.csl_test.parse()
csl_io = io.BytesIO(utf_8_encode(self.data['csl'])[0])
self.style = CitationStylesStyle(csl_io)
self._fix_input(self.data['input'])
self.references = [item['id'] for item in self.data['input']]
self.references_dict = CiteProcJSON(self.data['input'])
self.bibliography = CitationStylesBibliography(self.style,
self.references_dict)
self.expected = self.data['result'].splitlines()
@property
def data(self):
return self.csl_test.data
@staticmethod
def _fix_input(input_data):
for i, ref in enumerate(input_data):
if 'id' not in ref:
ref['id'] = i
if 'type' not in ref:
ref['type'] = 'book'
def execute(self):
if self.data['citation_items']:
citations = [
self.parse_citation(item)
for item in self.data['citation_items']
]
elif self.data['citations']:
citations = []
for cit in self.data['citations']:
cit = cit[0]
citation_items = [
self.parse_citation_item(cititem)
for cititem in cit['citationItems']
]
citation = Citation(citation_items)
if 'citationID' in cit:
citation.key = cit['citationID']
citation.note_index = cit['properties']['noteIndex']
citations.append(citation)
elif self.data['bibentries']:
citation_items = [
self.parse_citation_item({'id': entry})
for entry in self.data['bibentries'][-1]
]
citations = [Citation(citation_items)]
else:
citation_items = [
self.parse_citation_item({'id': ref})
for ref in self.references
]
citations = [Citation(citation_items)]
for citation in citations:
self.bibliography.register(citation)
if self.style.has_bibliography():
self.bibliography.sort()
results = []
do_nothing = lambda x: None # callback passed to cite()
if self.data['mode'] == 'citation':
if self.data['citations']:
for i, citation in enumerate(citations):
if i == len(citations) - 1:
dots_or_other = '>>'
else:
dots_or_other = '..'
results.append(
'{}[{}] '.format(dots_or_other, i) +
self.bibliography.cite(citation, do_nothing))
else:
for citation in citations:
results.append(self.bibliography.cite(
citation, do_nothing))
elif self.data['mode'] in ('bibliography', 'bibliography-nosort'):
results.append(self.bib_prefix)
for entry in self.bibliography.bibliography():
text = self.item_prefix + str(entry) + self.item_suffix
results.append(text)
results.append(self.bib_suffix)
return results
def parse_citation(self, citation_data):
citation_items = []
for item in citation_data:
citation_item = self.parse_citation_item(item)
citation_items.append(citation_item)
return Citation(citation_items)
def parse_citation_item(self, citation_item_data):
options = {}
for key, value in citation_item_data.items():
python_key = key.replace('-', '_')
if python_key == 'id':
reference_key = str(value)
continue
elif python_key == 'locator':
try:
options['locator'] = Locator(citation_item_data['label'],
value)
except KeyError:
# some tests don't specify the label
options['locator'] = Locator('page', value)
elif python_key == 'label':
pass
else:
options[python_key] = value
return CitationItem(reference_key, **options)
def __init__(self, bib_file, csl_file):
self.bib_file = bib_file
self.csl_file = csl_file
self.bibtex = BibTeX(bib_file)
self.style = CitationStylesStyle(csl_file, validate=False)
self.bibliography = CitationStylesBibliography(bib_style, bib_source,formatter.plain)
# Save it and reopen it, because the documentation doesn't describe a way to load a string
with open('bibtex.bib', 'w') as bibtex_file:
bibtexparser.dump(bibTeX, bibtex_file)
bib_source = BibTeX('bibtex.bib', encoding='utf-8')
# load a CSL style
fCSL = open('static/csl/mla-isawbib-full.csl', "rb")
fCSL = open('static/csl/mla-isawbib-authorless.csl', "rb")
bib_style = CitationStylesStyle(fCSL, validate=False)
# Create the citeproc-py bibliography, passing it the:
# * CitationStylesStyle,
# * BibliographySource (CiteProcJSON in this case), and
# * a formatter (plain, html, or you can write a custom formatter)
bibliography = CitationStylesBibliography(bib_style, bib_source,
formatter.html)
# Processing citations in a document needs to be done in two passes as for some
# CSL styles, a citation can depend on the order of citations in the
# bibliography and thus on citations following the current one.
# For this reason, we first need to register all citations with the
# CitationStylesBibliography.
for item in bibTeX.entries:
print(item['ID'])
citation = Citation([CitationItem(item['ID'])])
bibliography.register(citation)
print('')
print('Bibliography')
print('------------')
def bib(style_path,
ds,
return_cites_and_keys = False,
formatter = "chocolate",
apa_tweaks = True,
# The below options are ignored unless apa_tweaks is on.
always_include_issue = False,
include_isbn = False,
url_after_doi = False,
publisher_website = True,
abbreviate_given_names = True):
if isinstance(formatter, str):
try: formatter = formatter_from_name[formatter]
except KeyError: raise ValueError('Unknown formatter "{}"'.format(formatter))
style = get_style(style_path, apa_tweaks,
include_isbn, url_after_doi, abbreviate_given_names)
ds = deepcopy(ds)
if apa_tweaks:
# Distinguish entries that would have identical authors and years
# by adding suffixes to the years.
# Group works by author and year.
#
# (Actually, we use only an initial subset of authors,
# the same number that would be included in an inline citation
# after the first inline citation. This is 2 for 2 authors
# and 1 otherwise.)
ay = defaultdict(list)
for d in ds:
names = d.get('author') or d.get('editor')
if len(names) != 2:
names = [names[0]]
k = repr(names) + '/' + str(d['issued']['date-parts'][0][0])
if not any(d is v for v in ay[k]):
ay[k].append(d)
# If any group has more than one element, add suffixes.
for v in ay.values():
if len(v) > 1:
for i, d in enumerate(sorted(v, key = title_sort_key)):
d['year_suffix'] = ascii_lowercase[i]
for d in ds:
if 'id' not in d:
d['id'] = str(random())
for k in list(d.keys()):
if d[k] is None: del d[k]
if apa_tweaks:
# By default, don't include the issue number for
# journal articles.
if not always_include_issue and d['type'] == 'article-journal':
delf(d, 'issue')
# Use the weird "Retrieved from Dewey, Cheatem, &
# Howe website: http://example.com" format prescribed
# for reports.
if publisher_website and d['type'] == 'report' and 'publisher' in d and 'URL' in d:
d['URL'] = '{} website: {}'.format(
d.pop('publisher'), d['URL'])
# Add structure words for presentations and include
# the event place.
if d['type'] == 'speech' and d['genre'] == 'paper':
d['event'] = 'meeting of the {}, {}'.format(
d.pop('publisher'), d['event-place'])
if d['type'] == 'speech' and d['genre'] == 'video':
d['medium'] = 'Video file'
del d['genre']
# Format encyclopedia entries like book chapters.
if d['type'] == 'entry-encyclopedia':
d['type'] = 'chapter'
# When abbreviating given names, remove hyphens
# preceding lowercase letters. Otherwise, weird
# stuff happens.
if abbreviate_given_names and 'author' in d:
for a in d['author']:
if 'given' in a:
a['given'] = sub(
'-(.)',
lambda mo:
("" if mo.group(1).islower() else "-") +
mo.group(1),
a['given'])
bibliography = CitationStylesBibliography(
style,
CiteProcJSON(ds),
formatter)
cites = [ Citation([CitationItem(d['id'])]) for d in ds ]
for c in cites: bibliography.register(c)
def sort_key_f(item):
ref = item.reference
names = [(name['family'].lower(), name['given'][0].lower() if 'given' in name else '')
for name in ref.get('author') or ref.get('editor')]
return (names, ref['issued']['year'],
title_sort_key(ref),
ref['page']['first'] if 'page' in ref else '')
if len(ds) > 1:
# Sort the bibliography
# bibliography.sort() # Doesn't appear to handle leading "the"s correctly.
bibliography.items = sorted(bibliography.items, key = sort_key_f)
bibliography.keys = [item.key for item in bibliography.items]
bibl = bibliography.bibliography()
for i, s in enumerate(bibl):
s = ''.join(s)
# Fix spacing and punctuation issues.
s = s.replace(' ', ' ')
s = sub(r'([.!?…])\.', r'\1', s)
if apa_tweaks:
if formatter is citeproc.formatter.html or formatter is chocolate:
# Italicize the stuff between a journal name and a volume
# number.
s = sub(r'</i>, <i>(\d)', r', \1', s)
# Remove redundant periods that are separated
# from the first end-of-sentence mark by an </i>
# tag.
s = sub(r'([.!?…]</i>)\.', r'\1', s)
# If there are two authors and the first is a mononym,
# remove the comma after it.
s = sub('^([^.,]+), &', r'\1 &', s)
bibl[i] = s
if return_cites_and_keys:
fcites = [bibliography.cite(c, lambda x: None) for c in cites]
return (fcites, bibliography.keys, bibl)
else:
return bibl
