def test03(self):
ag = Fields.AuthorGroup()
ag.append(Fields.Author("Jäger, Herbert"))
item = {'author': ag, 'title': 'Bonjour tristesse'}
r = userformat.author_title_format(item)
self.assertEqual(r, 'Jäger, H.: Bonjour tristesse')
def test02(self):
ag = Fields.AuthorGroup()
ag.append(Fields.Author("Françoise Sagan"))
item = {'author': ag, 'title': 'Bonjour tristesse'}
r = userformat.author_title_format(item)
self.assertEqual(r, 'Sagan, F.: Bonjour tristesse')
def update_content(self, entry, text):
group = Fields.AuthorGroup()
for author in string.split(text, '\n'):
author = string.strip(author)
if author == '': continue
fields = string.split(author, ',')
if len(fields) == 1:
(first, last, lineage) = (None, fields[0], None)
elif len(fields) == 2:
(first, last, lineage) = (fields[1], fields[0], None)
else:
(first, last, lineage) = (fields[2], fields[0], fields[1])
group.append(Fields.Author((None, first, last, lineage)))
entry[self.field] = group
return
def test01(self):
ag = Fields.AuthorGroup()
ag.append(Fields.Author("Bruce Rind"))
item = {'author': ag}
r = userformat.author_editor_format(item)
self.assertEqual(r, 'Rind, B.')
ag.append(Fields.Author("Otto Forster"))
r = userformat.author_editor_format(item)
self.assertEqual(r, 'Rind/Forster')
ag.append(Fields.Author("Nicolas Bourbaki"))
r = userformat.author_editor_format(item)
self.assertEqual(r, 'Rind/Forster/Bourbaki')
ag.append(Fields.Author("Léonid Kameneff"))
r = userformat.author_editor_format(item)
self.assertEqual(r, 'Rind, B. et al.')
def parse_author(self, text):
ag = Fields.AuthorGroup()
rx = re.compile('\.(?:$|\s+)')
ry = re.compile('(.)')
for name in rx.split(text):
if not name: continue
la = name.split()
if len(la) == 1:
last, first = la[0], None
else:
last = ' '.join(la[:-1])
first = la[-1]
first = ry.sub(r'\1. ', first)
auth = Fields.Author(copy=(None, first, last, None))
ag.append(auth)
return ag
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 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