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 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 update (self, database, entry):
modified = False
key = string.strip (self.key.get_text ())
if key == '':
self.entry.key = None
modified = True
else:
if not key_re.match (key):
Utils.error_dialog_s(self.w.get_toplevel(),
_("Invalid key format"))
return None
key = Key.Key (database, key)
if key != self.entry.key:
if database.has_key (key):
Utils.error_dialog_s(
self.w.get_toplevel(),
_(u"Key “%s” already exists") % str (key.key))
return None
self.entry.key = key
modified = True
modified = self.type != self.entry.type or modified
for item in self.content:
try:
result = item.update(self.entry)
except UnicodeError:
f = Types.get_field(item.field)
Utils.error_dialog_s(
self.w.get_toplevel(),
_(u"The “%s” field contains a non Latin-1 symbol") %
f.name)
return None
if result == -1: return None
modified = result or modified
modified |= self.lt_update()
if not modified:
fields = self.entry.keys()
fields.sort()
if fields != self.fields: modified = 1
if modified:
return self.entry
return entry
def update(self, database, entry):
modified = False
key = string.strip(self.key.get_text())
if key == '':
self.entry.key = None
modified = True
else:
if not key_re.match(key):
Utils.error_dialog_s(self.w.get_toplevel(),
_("Invalid key format"))
return None
key = Key.Key(database, key)
if key != self.entry.key:
if database.has_key(key):
Utils.error_dialog_s(
self.w.get_toplevel(),
_(u"Key “%s” already exists") % str(key.key))
return None
self.entry.key = key
modified = True
modified = self.type != self.entry.type or modified
for item in self.content:
try:
result = item.update(self.entry)
except UnicodeError:
f = Types.get_field(item.field)
Utils.error_dialog_s(
self.w.get_toplevel(),
_(u"The “%s” field contains a non Latin-1 symbol") %
f.name)
return None
if result == -1: return None
modified = result or modified
modified |= self.lt_update()
if not modified:
fields = self.entry.keys()
fields.sort()
if fields != self.fields: modified = 1
if modified:
return self.entry
return entry
def __setitem__(self, key, value):
# First, set the cache for free
self.__text[key] = (value, 0)
if isinstance(value, Date):
return
# then, convert as bibtex.
if isinstance(value, Reference):
value = string.join(map(lambda item: item.key, value.list), ', ')
self.dict[key] = _bibtex.reverse(_fieldtype(Types.get_field(key)),
Config.get('bibtex+/braces').data,
value)
return
def lt_node_list(self, item, fields):
"""Return a list of longtexts (annotations) associated
with item """
fields = fields[:]
nodes = []
self.lt_list_add1(item, nodes, item.type.mandatory,
{'mandatory': True})
self.lt_list_add1(item, nodes, item.type.optional,
{'mandatory': False})
remaining = list_remove(fields, [x['key'] for x in nodes])
self.lt_list_add1(item, nodes, [Types.get_field(x) for x in remaining],
{'mandatory': False})
return nodes
def __setitem__(self, key, value):
# First, set the cache for free
self.__text[key] = (value, 0)
if isinstance(value, Date):
return
# then, convert as bibtex.
if isinstance(value, Reference):
value = string.join(map(lambda item: item.key, value.list), ', ')
self.dict[key] = _bibtex.reverse(_fieldtype(Types.get_field(key)),
Config.get('bibtex+/braces').data,
value)
return
def lt_node_list (self, item, fields):
"""Return a list of longtexts (annotations) associated
with item """
fields = fields[:]
nodes = []
self.lt_list_add1 (item, nodes,
item.type.mandatory, {'mandatory': True})
self.lt_list_add1 (item, nodes,
item.type.optional, {'mandatory': False})
remaining = list_remove (fields, [ x['key'] for x in nodes])
self.lt_list_add1 (
item, nodes, [Types.get_field(x) for x in remaining],
{'mandatory': False})
return nodes
def update_notebook (self):
if self.notebook_init:
self.notebook.foreach(
lambda x: self.notebook.remove_page(0))
self.notebook_init = True
self.current_page = None
names = (_("Mandatory"), _("Optional"), _("Notes"), _("Extra"))
self.fields = map (string.lower, self.entry.keys ())
self.content = []
for i in range (len(names)):
label = gtk.Label (names [i])
if i == 0:
table = [x.name.lower() for x
in self.entry.type.mandatory
if x.type != Fields.LongText]
self.add_type1_widget (label, table, i)
list_remove (self.fields, table)
elif i == 1:
table = [x.name.lower() for x
in self.entry.type.optional
if x.type != Fields.LongText]
self.add_type1_widget (label, table, i)
list_remove (self.fields, table)
elif i == 2:
self.lt_nodes = self.lt_node_list (
self.entry, self.fields)
self.lt_listvw = LT_Widget_1 (
self, i, self.dialogue)
self.lt_detail = LT_Widget_2 (
self.notebook, i, self.dialogue)
list_remove (self.fields,
[x['key'] for x in self.lt_nodes])
else:
table = [x for x in self.fields if
Types.get_field(x).type != Fields.LongText]
self.add_type1_widget (label, table, i)
self.notebook.set_border_width(6)
self.notebook.show ()
return
def create_field(self, *arg):
text = string.strip(string.lower(self.newfield.get_text()))
if not re.match(r"[a-z][\w_-]*$", text):
if not Utils.Callback(
"The fieldname '%s' looks invalid.\nReally proceed?" % text,
parent=self.w.get_toplevel()).answer():
return
# update the current entry
current = self.update(self.database, copy.deepcopy(self.entry))
if current is None:
return
self.entry = current
newtype = Types.get_field(text).type
self.entry[text] = newtype(_newcontent[newtype])
self.update_notebook()
def update_notebook(self):
if self.notebook_init:
self.notebook.foreach(lambda x: self.notebook.remove_page(0))
self.notebook_init = True
self.current_page = None
names = (_("Mandatory"), _("Optional"), _("Notes"), _("Extra"))
self.fields = map(string.lower, self.entry.keys())
self.content = []
for i in range(len(names)):
label = gtk.Label(names[i])
if i == 0:
table = [
x.name.lower() for x in self.entry.type.mandatory
if x.type != Fields.LongText
]
self.add_type1_widget(label, table, i)
list_remove(self.fields, table)
elif i == 1:
table = [
x.name.lower() for x in self.entry.type.optional
if x.type != Fields.LongText
]
self.add_type1_widget(label, table, i)
list_remove(self.fields, table)
elif i == 2:
self.lt_nodes = self.lt_node_list(self.entry, self.fields)
self.lt_listvw = LT_Widget_1(self, i, self.dialogue)
self.lt_detail = LT_Widget_2(self.notebook, i, self.dialogue)
list_remove(self.fields, [x['key'] for x in self.lt_nodes])
else:
table = [
x for x in self.fields
if Types.get_field(x).type != Fields.LongText
]
self.add_type1_widget(label, table, i)
self.notebook.set_border_width(6)
self.notebook.show()
return
def create_field(self, *arg):
text = string.strip(string.lower(self.newfield.get_text()))
if not re.match(r"[a-z][\w_-]*$", text):
if not Utils.Callback(
"The fieldname '%s' looks invalid.\nReally proceed?" %
text,
parent=self.w.get_toplevel()).answer():
return
# update the current entry
current = self.update(self.database, copy.deepcopy(self.entry))
if current is None:
return
self.entry = current
newtype = Types.get_field(text).type
self.entry[text] = newtype(_newcontent[newtype])
self.update_notebook()
def writer (iter, output, **argh):
entry = iter.first ()
mapping = Config.get ("refer/mapping").data
while entry:
for key in mapping.keys ():
# some fields are not to be used in output, as we
# lost their content
if not mapping [key] [1]: continue
field = mapping [key] [0]
if field == "label" and entry.key:
output.write ('%' + key + ' ')
output.write (Utils.format (str (entry.key.key), 75, 0, 0))
output.write ('\n')
continue
elif entry.has_key (field):
type = Types.get_field (field).type
if type == Fields.AuthorGroup:
# one field per author
for auth in entry [field]:
output.write ('%' + key + ' ')
output.write (Utils.format (str (auth), 75, 0, 0))
output.write ('\n')
continue
# general case
output.write ('%' + key + ' ')
output.write (Utils.format (str (entry [field]), 75, 0, 0))
output.write ('\n')
entry = iter.next ()
if entry: output.write ('\n')
return
def writer(iter, output, **argh):
entry = iter.first()
mapping = Config.get("refer/mapping").data
while entry:
for key in mapping.keys():
# some fields are not to be used in output, as we
# lost their content
if not mapping[key][1]: continue
field = mapping[key][0]
if field == "label" and entry.key:
output.write('%' + key + ' ')
output.write(Utils.format(str(entry.key.key), 75, 0, 0))
output.write('\n')
continue
elif entry.has_key(field):
type = Types.get_field(field).type
if type == Fields.AuthorGroup:
# one field per author
for auth in entry[field]:
output.write('%' + key + ' ')
output.write(Utils.format(str(auth), 75, 0, 0))
output.write('\n')
continue
# general case
output.write('%' + key + ' ')
output.write(Utils.format(str(entry[field]), 75, 0, 0))
output.write('\n')
entry = iter.next()
if entry: output.write('\n')
return
def __init__(self, parent=None):
self.dialog = gtk.Dialog(_('New Annotation Name'), parent, 0,
(gtk.STOCK_OK, gtk.RESPONSE_ACCEPT,
gtk.STOCK_CANCEL, gtk.RESPONSE_REJECT))
self.dialog.vbox.set_border_width(24)
self.dialog.vbox.pack_start(
gtk.Label(_('Name of the new annotation:')), True, True, 6)
self.options = gtk.combo_box_new_text()
self.dialog.vbox.pack_start(self.options, True, True, 6)
self.fields = [
x for x in Config.get('base/fields').data
if Types.get_field(x).type == Fields.LongText
]
self.fields.sort()
for i in self.fields:
self.options.append_text(i)
self.options.set_active(0)
self.options.connect('changed', self.changed)
self.dialog.set_default_response(gtk.RESPONSE_ACCEPT)
self.value = self.fields[0]
def __init__ (self, parent=None):
self.dialog = gtk.Dialog(
_('New Annotation Name'), parent, 0,
(gtk.STOCK_OK, gtk.RESPONSE_ACCEPT,
gtk.STOCK_CANCEL, gtk.RESPONSE_REJECT))
self.dialog.vbox.set_border_width (24)
self.dialog.vbox. pack_start (
gtk.Label (
_('Name of the new annotation:')),
True, True, 6)
self.options = gtk.combo_box_new_text ()
self.dialog.vbox.pack_start (self.options, True, True, 6)
self.fields = [ x for x in Config.get ('base/fields').data
if Types.get_field(x).type == Fields.LongText]
self.fields.sort()
for i in self.fields:
self.options.append_text (i)
self.options.set_active (0)
self.options.connect ('changed', self.changed)
self.dialog.set_default_response(gtk.RESPONSE_ACCEPT)
self.value = self.fields[0]
def entry_write(entry, output):
'''Print a single entry as BiBTeX code.'''
native = isinstance(entry, Entry)
tp = entry.type
# write the type and key
output.write('@%s{%s,\n' % (tp.name, entry.key.key))
# create a hash containing all the keys, to keep track
# of those who have been already written
dico = {}
datefields = Config.get('bibtex/datefield').data
convert = Config.get('bibtex/months').data
# we have to handle the special case of the dates
# create the list of months
monthlist = range(0, 12)
for key in convert.keys():
monthlist[convert[key] - 1] = key
dateformat = Config.get('bibtex+/dateformat').data
if native:
# loop over all the fields
for field in entry.keys():
if datefields.has_key(field):
# we are processing a date...
date = entry[field]
dico[datefields[field][0]] = str(date.year)
if date.month:
month = monthlist[date.month - 1]
if date.day:
month = dateformat % {'day': date.day, 'month': month}
dico[datefields[field][1]] = month
else:
# we are processing a normal entry
dico[field] = _bibtex.get_native(entry.dict[field])
else:
for field in entry.keys():
# convert the field in a bibtex form
if datefields.has_key(field):
# we are processing a date...
date = entry[field]
dico[datefields[field][0]] = str(date.year)
if date.month:
month = monthlist[date.month - 1]
if date.day:
month = dateformat % {'day': date.day, 'month': month}
dico[datefields[field][1]] = month
else:
# we are processing a normal entry
value = entry[field]
# eventually convert the crossref
if isinstance(value, Reference):
value = string.join(map(lambda item: item.key, value.list),
', ')
fieldtype = _fieldtype(Types.get_field(field))
dico[field] = _nativify(value, fieldtype)
first = True
# write according to the type order
for f in tp.mandatory + tp.optional:
# dico contains all the available fields
field = string.lower(f.name)
if not dico.has_key(field):
continue
if not first: output.write(',\n')
else: first = False
output.write(' %-14s = ' % f.name)
output.write(Utils.format(dico[field], 75, 19, 19)[19:])
del dico[field]
keys = dico.keys()
keys.sort()
for f in keys:
if not first: output.write(',\n')
else: first = False
output.write(' %-14s = ' % f)
output.write(Utils.format(dico[f], 75, 19, 19)[19:])
output.write('\n}\n\n')
return
def display(self, iterator):
# clear the access table
self.access = []
Utils.set_cursor(self.w, 'clock')
self.model.clear()
for entry in iterator:
row = []
i = 0
for f in self.fields:
row.append(i)
i = i + 1
if f == '-key-':
row.append((str(entry.key.key)).decode('latin-1'))
elif f == '-type-':
row.append(str(entry.type.name)) ## ascii
elif f == '-author/editor-':
row.append(
userformat.author_editor_format(entry).decode(
'latin-1'))
elif f == '-author/title-':
row.append(
userformat.author_title_format(entry).decode(
'latin-1'))
elif entry.has_key(f):
if Types.get_field(f).type == Fields.AuthorGroup:
text = join(map(lambda a: str(a.last), entry[f]), ', ')
elif Types.get_field(f).type == Fields.Date:
text = str(entry[f].year)
else:
text = str(entry[f])
row.append(text.decode('latin-1'))
else:
row.append('')
if True:
row.append(i)
if Resource.is_viewable(entry):
row.append(self.gvpixbuf)
else:
row.append(None)
iter = self.model.append()
apply(self.model.set, [iter] + row)
self.access.append(entry)
entry = iterator.next()
Utils.set_cursor(self.w, 'normal')
return
def get_latex(self, key):
''' Returns latex part '''
return _bibtex.get_latex(self.parser, self.dict[key],
_fieldtype(Types.get_field(key)))
def get_field (self, entry):
try:
return entry [self.field]
except KeyError:
return Types.get_field (self.field).type ('')
def next(self):
dict = {}
# read entry till next blank line
text = []
field = ''
while 1:
line = self.file.readline()
if line == '': break
line = string.rstrip(line)
# starting with a blank ?
if line == '' or line[0] == ' ':
# ...then we continue the current text
text.append(string.lstrip(line))
continue
# new entry ?
if separator_re.match(line): break
# else, this is a new field
if field:
# save the previous one if needed
dict[field] = '\n'.join(text)
text = []
# store the name of this new field
field = string.lower(line)
# don't waste the last field content
if field:
dict[field] = '\n'.join(text)
# did we parse a field ?
if len(dict) == 0: return None
# create the entry content
entry = Base.Entry(type=self.deftype)
for key in dict.keys():
if not self.mapping.has_key(key):
#print "warning: unused key `%s'" % key
continue
(name, type) = self.mapping[key]
text_type = Types.get_field(name).type
# parse a simple text field
if type == SimpleField:
entry[name] = text_type(string.strip(dict[key]))
elif type == KeywordField:
text = string.strip(dict[key])
if entry.has_key(name):
text = str(entry[name]) + '\n ' + text
entry[name] = text_type(text)
# parse an author field
elif type == AuthorField:
entry[name] = self.parse_author(dict[key])
continue
# parse a source field
elif type == SourceField:
dict_key = ' '.join(dict[key].split('\n'))
m = self.source_re.match(dict_key.strip())
if m:
year, month, day = None, None, None
j, v, s, n, p, o, y, d = m.group('journal', 'volume',
'inseries', 'number',
'pages', 'other', 'year',
'month')
if s: ### article in a monograph series
entry['booktitle'] = Fields.Text(j)
if d:
entry['beigabevermerk'] = Fields.LongText(d)
entry.type = Types.get_entry('incollection')
elif j:
entry['journal'] = Fields.Text(j)
if d and not d.isspace():
dates = d.split()
try:
month = long_month[dates[0]]
except KeyError:
pass
## import warnings
## warnings.filterwarnings ('once',
## message='date',
## module='OvidLike')
## warnings.warn (
## 'OVID: %s is not representable in date '
## 'field %s' %(dates[0], d), stacklevel=2)
if len(dates) > 1:
day = int(dates[1])
if v:
entry['volume'] = Fields.Text(v)
if n:
entry['number'] = Fields.Text(n)
if p:
entry['pages'] = Fields.Text(p)
if o:
entry['other-note'] = Fields.Text(o)
if y:
year = int(y)
entry['date'] = Fields.Date((year, month, day))
else:
print '>>> Error: Source field does not parse correctly:'
print dict_key
print entry
continue
return entry
def display (self, iterator):
# clear the access table
self.access = []
Utils.set_cursor (self.w, 'clock')
self.model.clear ()
for entry in iterator:
row = []
i = 0
for f in self.fields:
row.append (i)
i = i + 1
if f == '-key-':
row.append ((str (entry.key.key)).decode ('latin-1'))
elif f == '-type-':
row.append (str (entry.type.name)) ## ascii
elif f == '-author/editor-':
row.append (userformat.author_editor_format
(entry).decode ('latin-1'))
elif f == '-author/title-':
row.append (userformat.author_title_format
(entry).decode ('latin-1'))
elif entry.has_key (f):
if Types.get_field (f).type == Fields.AuthorGroup:
text = join (map (lambda a: str (a.last), entry [f]), ', ')
elif Types.get_field (f).type == Fields.Date:
text = str (entry [f].year)
else:
text = str (entry [f])
row.append (text.decode ('latin-1'))
else:
row.append ('')
if True:
row.append (i)
if Resource.is_viewable (entry):
row.append (self.gvpixbuf)
else:
row.append (None)
iter = self.model.append ()
apply (self.model.set, [iter] + row)
self.access.append (entry)
entry = iterator.next ()
Utils.set_cursor (self.w, 'normal')
return
def next (self):
dict = {}
# read entry till next blank line
text = []
field = ''
while 1:
line = self.file.readline ()
if line == '': break
line = string.rstrip (line)
# starting with a blank ?
if line == '' or line [0] == ' ':
# ...then we continue the current text
text.append (string.lstrip (line))
continue
# new entry ?
if separator_re.match (line): break
# else, this is a new field
if field:
# save the previous one if needed
dict [field] = '\n'.join(text)
text = []
# store the name of this new field
field = string.lower (line)
# don't waste the last field content
if field:
dict [field] = '\n'.join(text)
# did we parse a field ?
if len (dict) == 0: return None
# create the entry content
entry = Base.Entry (type = self.deftype)
for key in dict.keys ():
if not self.mapping.has_key (key):
#print "warning: unused key `%s'" % key
continue
(name, type) = self.mapping [key]
text_type = Types.get_field (name).type
# parse a simple text field
if type == SimpleField:
entry [name] = text_type (string.strip (dict [key]))
elif type == KeywordField:
text = string.strip (dict [key])
if entry.has_key (name):
text = str (entry [name]) + '\n ' + text
entry [name] = text_type (text)
# parse an author field
elif type == AuthorField:
entry [name] = self.parse_author (dict[key])
continue
# parse a source field
elif type == SourceField:
dict_key = ' '.join(dict[key].split('\n'))
m = self.source_re.match (dict_key.strip())
if m:
year, month, day = None, None, None
j, v, s, n, p, o, y, d = m.group(
'journal', 'volume', 'inseries', 'number',
'pages', 'other', 'year', 'month')
if s: ### article in a monograph series
entry['booktitle'] = Fields.Text (j)
if d:
entry['beigabevermerk'] = Fields.LongText (d)
entry.type = Types.get_entry('incollection')
elif j:
entry ['journal'] = Fields.Text (j)
if d and not d.isspace():
dates = d.split ()
try:
month = long_month [dates[0]]
except KeyError:
pass
## import warnings
## warnings.filterwarnings ('once',
## message='date',
## module='OvidLike')
## warnings.warn (
## 'OVID: %s is not representable in date '
## 'field %s' %(dates[0], d), stacklevel=2)
if len(dates) > 1:
day = int (dates[1])
if v:
entry ['volume'] = Fields.Text (v)
if n:
entry ['number'] = Fields.Text (n)
if p:
entry ['pages'] = Fields.Text (p)
if o:
entry ['other-note'] = Fields.Text(o)
if y:
year = int(y)
entry ['date'] = Fields.Date((year, month, day))
else:
print '>>> Error: Source field does not parse correctly:'
print dict_key
print entry
continue
return entry
def next(self):
current = None
data = ''
table = {}
# Skip whitespace
while 1:
line = self.file.readline()
if line == '': return table
line = string.rstrip(line)
if line != '': break
while 1:
head = header.match(line)
if head:
if current:
if table.has_key(current):
table[current].append(data)
else:
table[current] = [data]
current = string.strip(head.group(1))
data = head.group(2)
else:
cont = contin.match(line)
if cont:
data = data + ' ' + cont.group(1)
line = self.file.readline()
if line == '': break
line = string.rstrip(line)
if line == '': break
# don't forget the last item
if current:
if table.has_key(current):
table[current].append(data)
else:
table[current] = [data]
# create the entry with the actual fields
norm = {}
type = Types.get_entry('article')
if table.has_key('PMID'):
norm['url'] = Fields.URL(medurl + table['PMID'][0])
norm['medline-pmid'] = Fields.Text(table['PMID'][0])
del table['PMID']
if table.has_key('UI'):
norm[one_to_one['UI']] = Fields.Text(table['UI'][0])
del table['UI']
if table.has_key('AU'):
group = Fields.AuthorGroup()
for au in table['AU']:
# analyze the author by ourself.
first, last, lineage = [], [], []
for part in string.split(au, ' '):
if part.isupper():
# in upper-case, this is a first name
if len(last) > 0:
first.append(part)
else:
# if there is no last name, there can't be a first name
last.append(part)
else:
if len(first) > 0:
# there was a first name, this must be a lineage
lineage.append(part)
else:
last.append(part)
if len(first) > 1:
print "medline: long first name found. skipping."
first = first[0:1]
if len(first) > 0:
first = string.join(first[0], '. ') + '.'
else:
first = None
if len(last) > 0:
last = string.join(last, ' ')
else:
last = None
if len(lineage) > 0:
lineage = string.join(lineage, ' ')
else:
lineage = None
group.append(Fields.Author((None, first, last, lineage)))
norm[one_to_one['AU']] = group
del table['AU']
if table.has_key('DP'):
fields = string.split(table['DP'][0], ' ')
norm[one_to_one['DP']] = Fields.Date(fields[0])
del table['DP']
# The simple fields...
for f in table.keys():
f_mapped = one_to_one.get(f, 'medline-%s' % (f.lower()))
text_type = Types.get_field(f_mapped).type
norm[f_mapped] = text_type(string.join(table[f], " ; "))
return Base.Entry(None, type, norm)
def next (self):
lines = {}
in_table = {}
file_notes, file_time, file_version, file_format = ('','','','')
while 1:
line = self.file.readline()
if line == '': return lines # what does that mean ??
head = xheader.match(line)
if not head :
pass
elif head.group(1) == 'Date:':
file_time = time.strftime(
"%Y-%m-%d %H:%M", rfc822.parsedate(head.group(2)))
elif head.group(1) == 'Notes:':
file_notes = string.strip(head.group(2))
elif head.group(1) == 'FN':
file_format = head.group(2)
elif head.group(1) == 'VR':
file_version = head.group(2)
elif len(head.group(1)) == 2 :
break
else :
pass
self.extraneous.append(line)
self.isifileformat = self.isifileformat or "Isifile format %s(%s)" % (
file_format, file_version)
self.isifileinfo = self.isifileinfo or "ISI %s (%s) %s" %(
file_time, file_notes, login_name)
while 1:
if line == 'ER':break
if head :
key = head.group(1)
if key == 'ER': break
val = head.group(3)
if lines.has_key(key):
lines[key].append(val)
else:
lines[key] = [val]
else:
cont = contin.match(line)
if cont :
val = cont.group(1)
lines[key].append(val)
else: break
line = self.file.readline()
if line == '': break # error situation
head = header.match (line)
key = 'PT'
if lines.has_key(key):
if string.strip(lines[key][0])[0] == 'J':
del lines [key]
else:
print 'Warning: Unknown type of entry (%s) -- may need editing.' %(
lines[key])
type = Types.get_entry ('article')
for key in ( 'AU', 'ED'):
if lines.has_key(key):
group = Fields.AuthorGroup()
for item in lines[key]:
if string.strip(item) =='[Anon]' :
auth = [item]
else:
name, firstn = string.split (item, ',')
auth = ["%s, " % name]
for i in string.strip (firstn):
auth.append ("%s. " % i)
group.append (Fields.Author("".join(auth)))
if key == 'AU':
in_table['author'] = group
elif key == 'ED':
in_table['editor'] = group
del lines[key]
key, key1, key2 = 'PG', 'BP', 'EP'
if lines.has_key(key1) and lines.has_key(key2):
if len(lines[key1]) == len(lines[key2]):
pages = []
for i in range(len(lines[key1])):
firstpg = lines[key1] [i]
lastpg = lines[key2] [i]
pages.append(('%s -- %s' % (firstpg, lastpg)))
in_table['pages'] = Fields.Text (string.join(pages, '; '))
del lines[key1]; del lines[key2]
else: print 'inconsistent BP, EP fields found'
if lines.has_key(key):
in_table['size'] = Fields.Text ('%s p.' %(lines[key][0]))
del lines[key]
key = 'PY'
if lines.has_key(key):
val = lines[key][0]
in_table['date'] = Fields.Date(val)
del lines[key]
key = 'ID'
if lines. has_key(key):
val = "[ISI:] %s ;;" %(string.lower(string.join(lines[key], ' ')))
if lines.has_key('DE'):
lines['DE'].append ('; ' + val)
else :
lines['DE'] = [val]
del lines[key]
# journal titles come in various forms
if lines.has_key ('SO'):
uc_title = ' '.join(lines['SO'])
in_table ['journal'] = Fields.Text (uc_title)
if lines.has_key('JI'):
uc_title = re.split(r"([- .,/]+)", uc_title)
ca_title = re.split(r"[- .,/]+", ' '.join(lines['JI']))
i , Title = 0, []
for word in uc_title:
Word = string.capitalize(word)
if word == ca_title [i]:
Title.append (word)
i += 1
elif Word.startswith(ca_title[i]):
Title.append(Word)
i += 1
else:
Title.append(string.lower(word))
del lines['JI']
in_table['journal'] = Fields.Text ("".join(Title))
del lines['SO']
for key in lines.keys():
mapped_key, joiner = key_map.get(
key, ('isifile-%s' %(key.lower()), ' ; '))
text_type = Types.get_field (mapped_key).type
in_table [mapped_key] = text_type (joiner.join(lines[key]))
return Base.Entry ( None, type, in_table)
def entry_write(entry, output):
'''Print a single entry as BiBTeX code.'''
native = isinstance(entry, Entry)
tp = entry.type
# write the type and key
output.write('@%s{%s,\n' % (tp.name, entry.key.key))
# create a hash containing all the keys, to keep track
# of those who have been already written
dico = {}
datefields = Config.get('bibtex/datefield').data
convert = Config.get('bibtex/months').data
# we have to handle the special case of the dates
# create the list of months
monthlist = range(0, 12)
for key in convert.keys():
monthlist[convert[key]-1] = key
dateformat = Config.get('bibtex+/dateformat').data
if native:
# loop over all the fields
for field in entry.keys():
if datefields.has_key(field):
# we are processing a date...
date = entry[field]
dico[datefields[field][0]] = str(date.year)
if date.month:
month = monthlist[date.month - 1]
if date.day:
month = dateformat % {'day': date.day,
'month' : month}
dico[datefields[field][1]] = month
else:
# we are processing a normal entry
dico[field] = _bibtex.get_native(entry.dict[field])
else:
for field in entry.keys():
# convert the field in a bibtex form
if datefields.has_key(field):
# we are processing a date...
date = entry[field]
dico[datefields[field][0]] = str(date.year)
if date.month:
month = monthlist[date.month - 1]
if date.day:
month = dateformat % {'day': date.day,
'month' : month}
dico[datefields[field][1]] = month
else:
# we are processing a normal entry
value = entry[field]
# eventually convert the crossref
if isinstance(value, Reference):
value = string.join(map(lambda item: item.key, value.list), ', ')
fieldtype = _fieldtype(Types.get_field(field))
dico[field] = _nativify(value, fieldtype)
first = True
# write according to the type order
for f in tp.mandatory + tp.optional:
# dico contains all the available fields
field = string.lower(f.name)
if not dico.has_key(field):
continue
if not first: output.write (',\n')
else: first = False
output.write(' %-14s = ' % f.name)
output.write(Utils.format (dico[field],
75, 19, 19)[19:])
del dico[field]
keys = dico.keys()
keys.sort()
for f in keys:
if not first: output.write (',\n')
else: first = False
output.write(' %-14s = ' % f)
output.write(Utils.format(dico[f],
75, 19, 19)[19:])
output.write ('\n}\n\n')
return
def next(self):
data = ''
fields = {}
type = None
label = None
while 1:
line = self.file.readline()
if line == '' and not fields:
return None
line = string.strip(line)
if line == '' and fields:
# store the current field
if type:
if type == "label":
label = string.join(string.split(data), ' ')
elif fields.has_key(type):
fields[type].append(
string.join(string.split(data), ' '))
else:
fields[type] = [string.join(string.split(data), ' ')]
# determine the real type
while 1:
if fields.has_key('journal'):
type = 'article'
break
if fields.has_key('booktitle'):
type = 'inbook'
break
if fields.has_key('volume') or fields.has_key('number'):
type = 'inproceedings'
break
if fields.has_key('publisher'):
type = 'book'
break
if fields.has_key('author') and fields.has_key('title'):
type = 'unpublished'
break
type = 'misc'
break
entry = Types.get_entry(type)
for f in fields.keys():
type = Types.get_field(f).type
if type == Fields.AuthorGroup:
group = Fields.AuthorGroup()
for auth in fields[f]:
group.append(Fields.Author(auth))
fields[f] = group
else:
if len(fields[f]) > 1:
sys.stderr.write(
"warning: field `%s' is defined more than once"
% f)
continue
fields[f] = type(fields[f][0])
if label:
key = Key.Key(None, label)
return Base.Entry(key, entry, fields)
else:
return Base.Entry(None, entry, fields)
t = tag_re.match(line)
# we matched a new field start
if t:
if type:
if type == "label":
label = string.join(string.split(data), ' ')
elif fields.has_key(type):
fields[type].append(
string.join(string.split(data), ' '))
else:
fields[type] = [string.join(string.split(data), ' ')]
type = t.group(1)
if not self.mapping.has_key(type):
print "warning: key `%s' has been skipped" % (type)
type = None
data = ''
else:
# store the current field
type = self.mapping[type][0]
data = t.group(2)
continue
# in the general case, append the new text
data = data + ' ' + line
def get_latex(self, key):
''' Returns latex part '''
return _bibtex.get_latex(self.parser,
self.dict[key],
_fieldtype(Types.get_field(key)))
def next (self):
data = ''
fields = {}
type = None
label = None
while 1:
line = self.file.readline ()
if line == '' and not fields:
return None
line = string.strip (line)
if line == '' and fields:
# store the current field
if type:
if type == "label":
label = string.join (string.split (data), ' ')
elif fields.has_key (type):
fields [type].append (string.join (string.split (data), ' '))
else:
fields [type] = [string.join (string.split (data), ' ')]
# determine the real type
while 1:
if fields.has_key ('journal'):
type = 'article'
break
if fields.has_key ('booktitle'):
type = 'inbook'
break
if fields.has_key ('volume') or fields.has_key ('number'):
type = 'inproceedings'
break
if fields.has_key ('publisher'):
type = 'book'
break
if fields.has_key ('author') and fields.has_key ('title'):
type = 'unpublished'
break
type = 'misc'
break
entry = Types.get_entry (type)
for f in fields.keys ():
type = Types.get_field (f).type
if type == Fields.AuthorGroup:
group = Fields.AuthorGroup ()
for auth in fields [f]:
group.append (Fields.Author (auth))
fields [f] = group
else:
if len (fields [f]) > 1:
sys.stderr.write ("warning: field `%s' is defined more than once" % f)
continue
fields [f] = type (fields [f] [0])
if label:
key = Key.Key (None, label)
return Base.Entry (key, entry, fields)
else:
return Base.Entry (None, entry, fields)
t = tag_re.match (line)
# we matched a new field start
if t:
if type:
if type == "label":
label = string.join (string.split (data), ' ')
elif fields.has_key (type):
fields [type].append (string.join (string.split (data), ' '))
else:
fields [type] = [string.join (string.split (data), ' ')]
type = t.group (1)
if not self.mapping.has_key (type):
print "warning: key `%s' has been skipped" % (type)
type = None
data = ''
else:
# store the current field
type = self.mapping [type] [0]
data = t.group (2)
continue
# in the general case, append the new text
data = data + ' ' + line
def next (self):
current = None
data = ''
table = {}
# Skip whitespace
while 1:
line = self.file.readline ()
if line == '': return table
line = string.rstrip (line)
if line != '': break
while 1:
head = header.match (line)
if head:
if current:
if table.has_key (current):
table [current].append (data)
else:
table [current] = [data]
current = string.strip (head.group (1))
data = head.group (2)
else:
cont = contin.match (line)
if cont:
data = data + ' ' + cont.group (1)
line = self.file.readline ()
if line == '': break
line = string.rstrip (line)
if line == '': break
# don't forget the last item
if current:
if table.has_key (current):
table [current].append (data)
else:
table [current] = [data]
# create the entry with the actual fields
norm = {}
type = Types.get_entry ('article')
if table.has_key ('PMID'):
norm ['url'] = Fields.URL (medurl + table ['PMID'] [0])
norm ['medline-pmid'] = Fields.Text (table ['PMID'] [0])
del table ['PMID']
if table.has_key ('UI'):
norm [one_to_one ['UI']] = Fields.Text (table ['UI'] [0])
del table ['UI']
if table.has_key ('AU'):
group = Fields.AuthorGroup ()
for au in table ['AU']:
# analyze the author by ourself.
first, last, lineage = [], [], []
for part in string.split (au, ' '):
if part.isupper ():
# in upper-case, this is a first name
if len (last) > 0:
first.append (part)
else:
# if there is no last name, there can't be a first name
last.append (part)
else:
if len (first) > 0:
# there was a first name, this must be a lineage
lineage.append (part)
else:
last.append (part)
if len (first) > 1:
print "medline: long first name found. skipping."
first = first [0:1]
if len (first) > 0:
first = string.join (first [0], '. ') + '.'
else:
first = None
if len (last) > 0:
last = string.join (last, ' ')
else:
last = None
if len (lineage) > 0:
lineage = string.join (lineage, ' ')
else:
lineage = None
group.append (Fields.Author ((None, first, last, lineage)))
norm [one_to_one ['AU']] = group
del table ['AU']
if table.has_key ('DP'):
fields = string.split (table ['DP'][0], ' ')
norm [one_to_one ['DP']] = Fields.Date (fields [0])
del table ['DP']
# The simple fields...
for f in table.keys ():
f_mapped = one_to_one.get(f, 'medline-%s' %(f.lower()))
text_type = Types.get_field(f_mapped).type
norm [f_mapped] = text_type (string.join (table [f], " ; "))
return Base.Entry (None, type, norm)
def next (self):
current = None
data = ''
table = {}
# Skip whitespace
while 1:
line = self.file.readline ()
if line == '': return table
line = string.rstrip (line)
if line != '': break
while 1:
head = header.match (line)
if head:
if current:
if table.has_key (current):
table [current].append (data)
else:
table [current] = [data]
current = string.strip (head.group (1))
data = head.group (2)
else:
cont = contin.match (line)
if cont:
data = data + ' ' + cont.group (1)
line = self.file.readline ()
if line == '': break
line = string.rstrip (line)
if line == '': break
# don't forget the last item
if current:
if table.has_key (current):
table [current].append (data)
else:
table [current] = [data]
# create the entry with the actual fields
norm = {}
type = Types.get_entry ('article')
if table.has_key ('PMID'):
norm ['url'] = Fields.URL (medurl + table ['PMID'] [0])
norm ['medline-pmid'] = Fields.Text (table ['PMID'] [0])
del table ['PMID']
if table.has_key ('UI'):
norm [one_to_one ['UI']] = Fields.Text (table ['UI'] [0])
del table ['UI']
if table.has_key ('AU'):
group = Fields.AuthorGroup ()
for au in table ['AU']:
# analyze the author names
first, last, lineage = [], [], []
parts = string.split (au, ' ')
# if the last part is not uppercase, it is a lineage ('Jr', '3rd', etc.)
if not parts [-1] . isupper():
lineage.append (parts.pop ())
else:
lineage = None
# after removing lineage from list, last part should be the first initial(s)
if len (parts) > 1:
first = parts.pop ()
first = string.join (first, '. ') + '.'
else:
first = None
# join remaining parts to form the last name. if it's one initial, give it a '.'
for part in parts:
if len (part) == 1:
last.append (part + '.')
else:
last.append (part)
if len (parts) > 0:
last = string.join(last, ' ')
else:
last = None
group.append (Fields.Author ((None, first, last, lineage)))
norm [one_to_one ['AU']] = group
del table ['AU']
if table.has_key ('DP'):
fields = string.split (table ['DP'][0], ' ')
norm [one_to_one ['DP']] = Fields.Date (fields [0])
del table ['DP']
# The simple fields...
for f in table.keys ():
f_mapped = one_to_one.get(f, 'medline-%s' %(f.lower()))
text_type = Types.get_field(f_mapped).type
norm [f_mapped] = text_type (string.join (table [f], " ; "))
return Base.Entry (None, type, norm)