def expand(file, entry):
items = entry[4]
for k in items.keys():
items[k] = _bibtex.expand(file, items[k], typemap.get(k, -1))
items['key'] = entry[0]
items['type'] = entry[1]
return items
def open(self, filename):
"""Read and parse bibtex file using python-bibtex."""
# figure out what the second argument means
file = _bibtex.open_file(filename, 1)
while 1:
entry = _bibtex.next(file)
if entry == None: break
eprops = {}
for k, v in entry[4].iteritems():
# figure out what the last argument really does
# leaving in -1 seems to be save
value = _bibtex.expand(file, v, 0)[2]
try:
value = unicode(value, 'utf-8')
except UnicodeDecodeError, e:
print "ERROR: Failed to decode string '%s'" % value
raise
if k.lower() == 'author':
value = value.split(' and ')
if k.lower() == 'pages':
value = tuple(value.replace('-', ' ').split())
eprops[k] = value
# bibtex key is dict key
self[entry[0]] = (entry[1], eprops)
def expand(file, entry, type=-1):
"""Inline the expanded respresentation of each field."""
bibkey, bibtype, a, b, items = entry
results = []
for k in sorted(items):
results.append((k, _bibtex.expand(file, items[k], typemap.get(k, -1))))
return (bibkey, bibtype, a, b, results)
def __init__(self, context, t):
"""Constructs a BibtexRecord using the _bibtex mystery tuple."""
## A hack to deal with _bibtex' incomplete handling of the "month" field
self.date = [0,0] # Month, Year
##
self.handle = t[0]
self.entry_type = t[1]
t[2] #Mystery!
t[3] #Mystery!
self.typemap = {} #If we know that some fields should be a particular type.
self.data = {}
items = t[4]
for k in items.keys():
ty = self.typemap.get(k, -1)
x = _bibtex.expand(context, items[k], ty)
if k == "month":
month = BIBTEX_MONTHS[_bibtex.get_native(items[k])]
self.date[0] = month
self.data["month"] = month
continue
if ty == -1: ty = x[0]
self.data[k] = self.build(ty, x)
if 'year' in self.data.keys():
self.date[1] = self.data['year']
else:
self.date[1] = 1900
def field_and_loss(self, key):
# look in the cache first
if self.__text.has_key(key):
return self.__text[key]
obj = self.dict[key]
# search its declared type
fieldtype = Types.get_field(key)
ret = _bibtex.expand(self.parser, obj, _fieldtype(fieldtype))
fieldtype = fieldtype.type
if fieldtype == AuthorGroup:
# Author
val = AuthorGroup()
for aut in ret[3]:
val.append(Author(aut))
elif fieldtype == Date:
# Date
val = Date((ret[3], None, None))
elif fieldtype == LongText:
# Annotation text
val = BibLongTextField(ret[2], self.get_latex(key))
elif fieldtype == Text:
# Any other text
val = BibTextField(ret[2], self.get_latex(key))
elif fieldtype == Reference:
# a reference on the same database
val = Reference(ret[2], self.key.base)
else:
# specific fields, like URL
val = fieldtype(ret[2])
self.__text[key] = (val, ret[1])
return (val, ret[1])
def open(self, filename):
"""Read and parse bibtex file using python-bibtex."""
# figure out what the second argument means
file = _bibtex.open_file(filename, 1)
while 1:
entry = _bibtex.next(file)
if entry == None: break
eprops = {}
for k,v in entry[4].iteritems():
# figure out what the last argument really does
# leaving in -1 seems to be save
value = _bibtex.expand(file, v, 0)[2]
try:
value = unicode(value, 'utf-8')
except UnicodeDecodeError, e:
print "ERROR: Failed to decode string '%s'" % value
raise
if k.lower() == 'author':
value = value.split(' and ')
if k.lower() == 'pages':
value = tuple(value.replace('-', ' ').split())
eprops[k] = value
# bibtex key is dict key
self[entry[0]] = (entry[1],eprops)
def __init__(self, context, t):
"""Constructs a BibtexRecord using the _bibtex mystery tuple."""
## A hack to deal with _bibtex' incomplete handling of the "month" field
self.date = [0, 0] # Month, Year
##
self.handle = t[0]
self.entry_type = t[1]
t[2] #Mystery!
t[3] #Mystery!
self.typemap = {
} #If we know that some fields should be a particular type.
self.data = {}
items = t[4]
for k in items.keys():
ty = self.typemap.get(k, -1)
x = _bibtex.expand(context, items[k], ty)
if k == "month":
month = BIBTEX_MONTHS[self._strip_braces(
_bibtex.get_native(items[k]))]
self.date[0] = month
self.data["month"] = month
continue
if ty == -1: ty = x[0]
self.data[k] = self.build(ty, x)
self.date[1] = self.data["year"]
def field_and_loss(self, key):
# look in the cache first
if self.__text.has_key(key):
return self.__text[key]
obj = self.dict[key]
# search its declared type
fieldtype = Types.get_field(key)
ret = _bibtex.expand(self.parser, obj,
_fieldtype(fieldtype))
fieldtype = fieldtype.type
if fieldtype == AuthorGroup:
# Author
val = AuthorGroup ()
for aut in ret[3]:
val.append (Author(aut))
elif fieldtype == Date:
# Date
val = Date((ret[3], None, None))
elif fieldtype == LongText:
# Annotation text
val = BibLongTextField(ret[2], self.get_latex(key))
elif fieldtype == Text:
# Any other text
val = BibTextField(ret[2], self.get_latex(key))
elif fieldtype == Reference:
# a reference on the same database
val = Reference(ret[2], self.key.base)
else:
# specific fields, like URL
val = fieldtype(ret[2])
self.__text[key] = (val, ret[1])
return (val, ret[1])
def expand(file, entry):
if entry[0] in ("preamble", "string"):
return entry
bibkind, (bibkey, bibtype, a, b, items) = entry
results = [(k, _bibtex.expand(file, items[k], -1)) for k in sorted(items)]
return (bibkind, (bibkey, bibtype, a, b, results))
def expand(file, entry, type=-1):
"""Inline the expanded respresentation of each field."""
bibkey, bibtype, a, b, items = entry
results = []
for k in sorted(items):
results.append(
(k, _bibtex.expand(file, items[k], typemap.get(k, -1))))
return (bibkey, bibtype, a, b, results)
def expand (f, entry, typemap = -1):
items = entry [4]
for k in items.keys ():
items [k] = _bibtex.expand (f, items [k], typemap)
if k!='year':
items[k]=items[k][-1]
items['type']=entry[1]
items['bibTex']=entry[0]
return items
def expand(file, entry):
if entry[0] in ('preamble', 'string'):
return entry
bibkind, (bibkey, bibtype, a, b, items) = entry
results = [(k, _bibtex.expand(file, items[k], -1))
for k in sorted(items)]
return (bibkind, (bibkey, bibtype, a, b, results))
def readBib(filename):
## _bibtex has horrible interface
bibfile=bib.open_file(filename,1) # 2nd argument: strict mode
db={}
#rq=bibRecode.request('latex..latin1')
while True:
entry=bib.next(bibfile)
if entry is None: break
key,type,dta=entry[0],entry[1],entry[4]
item={'type':type}
for field in dta.keys():
expanded=bib.expand(bibfile,dta[field],-1)
#conv=bibRecode.recode(rq,expanded[2])
item[field]=expanded[2].strip()
db[key]=item
## now we don't need _bibtex anymore, everything is in our dicts
return db
def readBib(filename):
## _bibtex has horrible interface
bibfile = bib.open_file(filename, 1) # 2nd argument: strict mode
db = {}
#rq=bibRecode.request('latex..latin1')
while True:
entry = bib.next(bibfile)
if entry is None: break
key, type, dta = entry[0], entry[1], entry[4]
item = {'type': type}
for field in dta.keys():
expanded = bib.expand(bibfile, dta[field], -1)
#conv=bibRecode.recode(rq,expanded[2])
item[field] = expanded[2].strip()
db[key] = item
## now we don't need _bibtex anymore, everything is in our dicts
return db
def parse(bibfilename, strict=0):
print "opening bibtex file ", bibfilename
file = _bibtex.open_file(bibfilename, strict)
entry = _bibtex.next(file)
counter = 0
answer = {}
while (entry <> None):
counter += 1
key = entry[0] # such as lemire-kaser2003
type = entry[1] #article, proc...
items = entry[4]
myentry = {}
myentry['type'] = type
for itemkey in items.keys():
myentry[itemkey.lower()] = cgi.escape(
_bibtex.expand(file, items[itemkey], -1)[2])
answer[key] = myentry
entry = _bibtex.next(file)
print "parsed ", counter, " entries"
return answer
if False:
print importer("http://portal.acm.org/citation.cfm?id=277650.277719").load_bibtex()
print importer("http://portal.acm.org/citation.cfm?id=324550").load_bibtex()
print importer("http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.23.414").load_bibtex()
test = importer("http://eprint.iacr.org/2009/137").load_bibtex()
print test
else:
test = """@misc{cryptoeprint:2009:137,
author = {Nicolas T. Courtois},
title = {The Dark Side of Security by Obscurity and Cloning MiFare Classic Rail and Building Passes Anywhere, Anytime},
howpublished = {Cryptology ePrint Archive, Report 2009/137},
year = {2009},
note = {\url{http://eprint.iacr.org/}},
}"""
if "_bibtex" in sys.modules:
b = _bibtex.open_string("foo", test, True)
print b
_bibtex.first(b)
i = _bibtex.next(b)
while i:
print i
items = i[4]
for k in items.keys():
items[k] = _bibtex.expand(b, items[k], -1)
print items
i = _bibtex.next(b)
for m in matches:
for k in m.split(','):
start_keys.add(k)
keys_to_papers = {}
while True:
be = _bibtex.next(b)
if not be:
break
key = be[0]
print >> sys.stderr, "===" + str(key)
bh = be[4]
h = {}
for k in be[4].keys ():
h[k] = _bibtex.expand(b,bh[k],-1)
if ("title" in h) and ("year" in h):
t = h["title"]
y = h["year"]
p = Paper(key,t[2],y[2])
bibtex_papers.add(p)
keys_to_papers[key] = p
paper_pdf_file = os.path.join(pdf_directory,key+".pdf")
if os.path.exists(paper_pdf_file):
papers_with_pdf_versions.add(p)
else:
print >> sys.stderr, "Warning: no PDF file "+paper_pdf_file
if not tex_file:
start_keys.add(key)
earliest_year = 10000
for m in matches:
for k in m.split(','):
start_keys.add(k)
keys_to_papers = {}
while True:
be = _bibtex.next(b)
if not be:
break
key = be[0]
print >> sys.stderr, "===" + str(key)
bh = be[4]
h = {}
for k in be[4].keys():
h[k] = _bibtex.expand(b, bh[k], -1)
if ("title" in h) and ("year" in h):
t = h["title"]
y = h["year"]
p = Paper(key, t[2], y[2])
bibtex_papers.add(p)
keys_to_papers[key] = p
paper_pdf_file = os.path.join(pdf_directory, key + ".pdf")
if os.path.exists(paper_pdf_file):
papers_with_pdf_versions.add(p)
else:
print >> sys.stderr, "Warning: no PDF file " + paper_pdf_file
if not tex_file:
start_keys.add(key)
earliest_year = 10000
