def convertLaTeX2Unicode(self, source):
for latex_entity in _latex2utf8enc_mapping_simple.keys():
source = _encode(_decode(source).replace(latex_entity, _latex2utf8enc_mapping_simple[latex_entity]))
for latex_entity in _latex2utf8enc_mapping.keys():
source = _encode(_decode(source).replace(latex_entity, _latex2utf8enc_mapping[latex_entity]))
return source
def convertLaTeX2Unicode(self, source):
for latex_entity in _latex2utf8enc_mapping_simple.keys():
source = _encode(
_decode(source).replace(
latex_entity, _latex2utf8enc_mapping_simple[latex_entity]))
for latex_entity in _latex2utf8enc_mapping.keys():
source = _encode(
_decode(source).replace(latex_entity,
_latex2utf8enc_mapping[latex_entity]))
return source
def _bibitem_links(self, hit):
key = hit.group(1).strip()
catalog = getToolByName(self, 'portal_catalog')
brains = catalog(getId=key, portal_type=self.getReferenceTypes())
if brains:
brain = brains[0]
encoding = self.getProperty('default_charset') or 'utf-8'
authors = _encode(brain.Authors, encoding)
year = _encode(brain.publication_year, encoding)
title = _encode(brain.Title, encoding)
source = _encode(brain.Source, encoding)
url = brain.getURL()
link = '%s (%s) <a href="%s">%s</a>, %s' % \
(authors, year, url, title, source)
return link
else:
return key
def _bibitem_links(self, hit):
key = hit.group(1).strip()
catalog = getToolByName(self, 'portal_catalog')
brains = catalog(getId = key,
portal_type = self.getReferenceTypes()
)
if brains:
brain = brains[0]
encoding = self.getProperty('default_charset') or 'utf-8'
authors = _encode(brain.Authors, encoding)
year = _encode(brain.publication_year, encoding)
title = _encode(brain.Title, encoding)
source = _encode(brain.Source, encoding)
url = brain.getURL()
link = '%s (%s) <a href="%s">%s</a>, %s' % \
(authors, year, url, title, source)
return link
else:
return key
def pre_validate(self, REQUEST, errors):
reference_catalog = getToolByName(self, 'reference_catalog')
authors = REQUEST.get('authors', [])
result = []
references = []
# deduce author names from member reference
for author in authors:
reference = author.get('reference', None)
if reference == 'None':
author.reference = ''
elif reference:
reference_object = reference_catalog.lookupObject(reference)
if reference_object.isTranslatable():
references.append(reference_object.getCanonical().UID())
reference_object = reference_object.getCanonical()
else:
references.append(reference)
# obtain author data from privileged fields in the reference object
data = self.getAuthorDataFromMember(reference_object)
only_requested_data = not not author.get('lastname', None)
if data:
for key in [
key for key in data.keys()
if key not in ('middlename', )
]:
if not only_requested_data or (author.get(key, None)
== '?'):
if key == 'firstname':
author.firstnames = _decode(
data['firstname']) + ' ' + _decode(
data['middlename'])
else:
exec('author.%s = _decode(data[key])' % key)
# if this doesn't help, we try to derive the author name from the Title of reference... (YUK)
if not author.get('lastname', None):
firstnames, lastname = self._name_from_reference(reference)
author.firstnames = firstnames
author.lastname = lastname
if ''.join([
_decode(_encode(val)) for val in dict(author).values()
]).strip():
result.append(author)
REQUEST.form['authors'] = result[:]
REQUEST.form['member_publication_authors'] = references[:]
def getCookedBibRefId(self, ref, use_pid_on_import=True, **kwargs):
"""
cook id for ref dict or object, ref dict / object may be a single reference only
"""
isReferenceObject = (type(ref) != type({}))
if isReferenceObject:
ref = self._object2ref(ref)
new_id = 'nobody1000'
if use_pid_on_import and ref.get('pid'):
new_id = ref['pid']
else:
new_id = self._cookIdCore(ref, new_id=new_id)
return _encode(_decode(self._cleanId(new_id)))
def getCookedBibRefId(self, ref, use_pid_on_import=True, **kwargs):
"""
cook id for ref dict or object, ref dict / object may be a single reference only
"""
isReferenceObject = (type(ref) != type({}))
if isReferenceObject:
ref = self._object2ref(ref)
new_id = 'nobody1000'
if use_pid_on_import and ref.get('pid'):
new_id = ref['pid']
else:
new_id = self._cookIdCore(ref, new_id=new_id)
return _encode(_decode(self._cleanId(new_id)))
def _inline_links(self, hit):
keys = [k.strip() for k in hit.group(1).split(',')]
results = []
catalog = getToolByName(self, 'portal_catalog')
encoding = self.getProperty('default_charset') or 'utf-8'
for key in keys:
brains = catalog(getId=key, portal_type=self.getReferenceTypes())
if brains:
url = brains[0].getURL()
label = _encode(brains[0].citationLabel, encoding) \
or 'no label'
link = '<a href="%s">%s</a>' % (url, label)
results.append(link)
else:
results.append(key)
return '; '.join(results)
def _inline_links(self, hit):
keys = [k.strip() for k in hit.group(1).split(',')]
results = []
catalog = getToolByName(self, 'portal_catalog')
encoding = self.getProperty('default_charset') or 'utf-8'
for key in keys:
brains = catalog(getId = key,
portal_type = self.getReferenceTypes()
)
if brains:
url = brains[0].getURL()
label = _encode(brains[0].citationLabel, encoding) \
or 'no label'
link = '<a href="%s">%s</a>' % (url, label)
results.append(link)
else:
results.append(key)
return '; '.join(results)
def pre_validate(self, REQUEST, errors):
reference_catalog = getToolByName(self, 'reference_catalog')
authors = REQUEST.get('authors',[])
result = []
references=[]
# deduce author names from member reference
for author in authors:
reference = author.get('reference', None)
if reference == 'None':
author.reference = ''
elif reference:
reference_object = reference_catalog.lookupObject(reference)
if reference_object.isTranslatable():
references.append(reference_object.getCanonical().UID())
reference_object = reference_object.getCanonical()
else:
references.append(reference)
# obtain author data from privileged fields in the reference object
data = self.getAuthorDataFromMember(reference_object)
only_requested_data = not not author.get('lastname', None)
if data:
for key in [ key for key in data.keys() if key not in ('middlename',) ]:
if not only_requested_data or (author.get(key, None) == '?'):
if key == 'firstname':
author.firstnames = _decode(data['firstname']) + ' ' + _decode(data['middlename'])
else:
exec('author.%s = _decode(data[key])' % key)
# if this doesn't help, we try to derive the author name from the Title of reference... (YUK)
if not author.get('lastname', None):
firstnames, lastname = self._name_from_reference(reference)
author.firstnames = firstnames
author.lastname = lastname
if ''.join([_decode(_encode(val)) for val in dict(author).values()]).strip():
result.append(author)
REQUEST.form['authors'] = result[:]
REQUEST.form['member_publication_authors'] = references[:]
def convertLaTeX2Unicode(self, source):
# This function converts LaTeX entities to Unicode characters.
# This function was improved by JB, 2011-06-05.
#
# Changes:
# 1. Rather than:
# for k in d.keys():
# s = s.replace(k, d[k])
# use:
# for k, v in d.items():
# s = s.replace(k, v)
#
# 2. Rather than (decode -> replace -> encode) *every* *iteration*,
# decode once at the beginning, and encode again once at the end.
#
# This yielded an 8x speedup for the function (profiled by cProfile):
# 9.615 CPU secs per call on the ~1M benchmark BibTeX file (19.230 secs total),
# down to 1.198 CPU secs/call (2.397 secs total).
#
# Next, the functions '_build_mapping_tree' and '_replace_using_mapping_tree'
# were used to optimise-dowm the number of 'string.replace' calls.
# (For more info, read the comment immediately above '_build_mapping_tree'.)
#
# This optimisation yielded a 4.5x speedup for this function:
# 1.198 CPU secs per call (on the ~1M benchmark BibTeX file) down to 0.259.
# The total speedup of this function was now >35x: 9.615 down to 0.259.
source = _decode(source)
#for latex_entity, unicode_code_point in _latex2utf8enc_mapping_simple.items():
#source = source.replace(latex_entity, unicode_code_point)
source = _replace_using_mapping_tree(
source, LATEX2UTF8ENC_MAPPING_SIMPLE_TREE)
#for latex_entity, unicode_code_point in _latex2utf8enc_mapping.items():
#source = source.replace(latex_entity, unicode_code_point)
source = _replace_using_mapping_tree(source,
LATEX2UTF8ENC_MAPPING_TREE)
source = _encode(source)
return source
def convertLaTeX2Unicode(self, source):
# This function converts LaTeX entities to Unicode characters.
# This function was improved by JB, 2011-06-05.
#
# Changes:
# 1. Rather than:
# for k in d.keys():
# s = s.replace(k, d[k])
# use:
# for k, v in d.items():
# s = s.replace(k, v)
#
# 2. Rather than (decode -> replace -> encode) *every* *iteration*,
# decode once at the beginning, and encode again once at the end.
#
# This yielded an 8x speedup for the function (profiled by cProfile):
# 9.615 CPU secs per call on the ~1M benchmark BibTeX file (19.230 secs total),
# down to 1.198 CPU secs/call (2.397 secs total).
#
# Next, the functions '_build_mapping_tree' and '_replace_using_mapping_tree'
# were used to optimise-dowm the number of 'string.replace' calls.
# (For more info, read the comment immediately above '_build_mapping_tree'.)
#
# This optimisation yielded a 4.5x speedup for this function:
# 1.198 CPU secs per call (on the ~1M benchmark BibTeX file) down to 0.259.
# The total speedup of this function was now >35x: 9.615 down to 0.259.
source = _decode(source)
#for latex_entity, unicode_code_point in _latex2utf8enc_mapping_simple.items():
#source = source.replace(latex_entity, unicode_code_point)
source = _replace_using_mapping_tree(source, LATEX2UTF8ENC_MAPPING_SIMPLE_TREE)
#for latex_entity, unicode_code_point in _latex2utf8enc_mapping.items():
#source = source.replace(latex_entity, unicode_code_point)
source = _replace_using_mapping_tree(source, LATEX2UTF8ENC_MAPPING_TREE)
source = _encode(source)
return source
def log(msg, severity=INFO, detail='', error=None):
# XXX is this used somewhere?
LOG('CMFBibliographyAT', severity, _encode(msg), _encode(detail), error)
def render(self, data, source_format, target_format, output_encoding=None):
""" Transform data from 'source_format'
to 'target_format'
We have nothing, so we do nothing :)
>>> if _getCommand('bib', 'end', None) is not None:
... result = ExternalTransformUtility().render('', 'bib', 'end')
... assert result == ''
>>> data = '''
... @Book{bookreference.2008-02-04.7570607450,
... author = {Werner, kla{\"u}s},
... title = {H{\"a}rry Motter},
... year = {1980},
... publisher = {Diogenes}
... }'''
This should work. (If external bibutils are installed!)
We transform the `bib`-format into the `end`-format
>>> if _hasCommands(commands.get('bib2end')):
... result = ExternalTransformUtility().render(data, 'bib', 'end')
... # We need to take care of any stray Windows carriage returns.
... result = result.replace('\r', '')
... assert '''
... %0 Book
... %A Werner, kla"us title =. H"arry Motter
... %D 1980
... %I Diogenes
... %F bookreference.2008-02-04.7570607450 '''.strip() in result
This one is not allowed. No valid transformer exists for
`foo` and `bar` (foo2bar)
>>> ExternalTransformUtility().render(data, 'foo', 'bar')
Traceback (most recent call last):
...
ValueError: No transformation from 'foo' to 'bar' found.
"""
command = _getCommand(source_format, target_format)
if not command:
return ''
orig_path = os.environ['PATH']
if os.environ.has_key('BIBUTILS_PATH'):
os.environ['PATH'] = os.pathsep.join([orig_path,
os.environ['BIBUTILS_PATH']])
ts = time.time()
# This is a stinking workaround with hanging subprocesses on Linux.
# We had the case where "end2xml | xml2bib " was just hanging
# while reading the results from the output pipeline. So we fall
# back in a safe way to os.system() on Linux
if sys.platform == 'linux2':
input_filename = tempfile.mktemp()
error_filename = tempfile.mktemp()
output_filename = tempfile.mktemp()
file(input_filename, 'wb').write(_encode(data))
command = 'cat "%s" | %s 2>"%s" 1>"%s"' % (input_filename, command, error_filename, output_filename)
st = os.system(command)
error = file(output_filename, 'rb').read()
result = file(output_filename, 'rb').read()
os.unlink(input_filename)
os.unlink(output_filename)
os.unlink(error_filename)
else:
ts = time.time()
log.info(command)
p = Popen(command, shell=True, stdin=PIPE, stdout=PIPE, stderr=PIPE,
close_fds=False)
(fi, fo, fe) = (p.stdin, p.stdout, p.stderr)
fi.write(_encode(data))
fi.close()
result = fo.read()
fo.close()
error = fe.read()
fe.close()
log.info('Execution time: %2.2f seconds' % (time.time() - ts))
if error:
# command could be like 'ris2xml', or 'ris2xml | xml2bib'. It
# seems unlikely, but we'll code for an arbitrary number of
# pipes...
command_list = command.split(' | ')
for each in command_list:
if each in error and not result:
log.error("'%s' not found. Make sure 'bibutils' is installed.",
command)
if output_encoding is None:
return result
else:
return _convertToOutputEncoding(result,
output_encoding=output_encoding)
os.environ['PATH'] = orig_path
def log(msg, severity=INFO, detail='', error=None):
# XXX is this used somewhere?
LOG('CMFBibliographyAT', severity, _encode(msg), _encode(detail), error)
class ImportView(BrowserView):
template = ViewPageTemplateFile('import.pt')
def __call__(self):
self.processed = False
self.errors = {}
start_time = self.context.ZopeTime().timeTime()
if not self.request.form.get('form.submitted'):
return self.template()
# fetch value from request
input_encoding = self.request.form.get('input_encoding', 'utf-8')
span_of_search = self.request.form.get('span_of_search', None)
format = self.request.form.get('format', 'bib')
# process source
filename = None
source = self.request.form.get('up_text')
if not source:
upfile = self.request.form.get('file')
filename = upfile and getattr(upfile, 'filename', None)
if not filename:
self.errors['file'] = _(u'You must import a file or enter a'
' text.')
addStatusMessage(self.request,
_(u"Please correct the indicated errors."))
return self.template()
source = upfile.read()
if not source or not isinstance(source, basestring):
msg = "Could not read the file '%s'." % filename
self.errors['file'] = msg
addStatusMessage(self.request, _(unicode(msg)))
return self.template()
# skip DOS line breaks
source = source.replace('\r', '')
# get parsed entries from the Bibliography Tool
bibtool = getToolByName(self.context, 'portal_bibliography')
try:
entries = bibtool.getEntries(source,
format,
filename,
input_encoding=input_encoding)
except ImportParseError:
msg = """%s Parser's 'checkFormat' and guessing the format""" \
""" from the file name '%s' failed.""" % (format,
filename)
self.errors['format'] = msg
addStatusMessage(self.request, _(unicode(msg)))
return self.template()
except UnicodeError:
msg = """The choosen input encoding does not match the real """ \
"""encoding of your input data in order to convert it to """\
"""unicode internally."""
self.errors['input_encoding'] = msg
addStatusMessage(self.request, _(unicode(msg)))
return self.template()
except RuntimeError, e:
addStatusMessage(self.request, _(unicode(e)))
return self.template()
# debug message if entries is not a python list
if not entries or not isinstance(entries, (list, tuple)):
msg = "There must be something wrong with the parser"
addStatusMessage(self.request, _(unicode(msg)))
return self.template()
# start building the report
mtool = getToolByName(self.context, 'portal_membership')
member = mtool.getAuthenticatedMember()
fullname = member.getProperty('fullname', None)
if fullname:
username = '******' % (_encode(fullname), _encode(member.getId()))
else:
username = _encode(member.getId())
tmp_report = '[%s] Imported by %s' % (self.context.ZopeTime(),
username)
if filename is not None:
tmp_report += ' from file %s' % _encode(filename)
tmp_report += ':\n\n'
# process import for each entry
processedEntries = 0
importErrors = 0
logger.info('Start import of %s raw entries.' % len(entries))
counter = 0
for entry in entries:
counter += 1
count = '#%05i: ' % counter
logger.info(count + 'processing entry')
# Workaround for #36 where an entry represents
# an error from parser instead of a dict containing
# importable data
if isinstance(entry, basestring):
msg = 'Entry could not be parsed! %s' % _encode(entry)
upload = (msg, 'error')
logger.error(count + msg)
elif entry.get('title'):
logger.info(count + 'Normal processing')
upload = self.context.processSingleImport(
entry, span_of_search=span_of_search)
else:
formated = '; '.join([
'%s=%s' % (key, entry[key]) for key in sorted(entry.keys())
if key == key.lower()
])
upload = ('Found entry without title: %s\n' % formated,
'error')
logger.error(count + upload[0])
if upload[1] == 'ok':
processedEntries += 1
else:
importErrors += 1
state, msg = _encode(upload[1].upper()), _encode(upload[0])
tmp_report += '%s: %s\n' % (state, msg)
self.context.logImportReport(tmp_report)
self.processed = True
# set the portal status message up
msg = "Processed %i entries. There were %i errors. "\
"Import processed in %f seconds. See import report below." \
% (processedEntries, importErrors,
self.context.ZopeTime().timeTime() - start_time)
logger.info(msg)
addStatusMessage(self.request, _(unicode(msg)))
return self.template()
