def load_lexemes(self, fname):
lexDescrs = yamlReader.read_file(fname, self.errorHandler)
for dictDescr in lexDescrs:
if sys.getsizeof(self.lexemes) > self.LEX_MEMORY_LIMIT:
self.raise_error(u'Not enough memory for the lexemes.')
return
self.lexemes.append(Lexeme(dictDescr, self.errorHandler))
return len(self.lexemes)
def load_yaml_descrs(self, fnames):
"""
Load raw descriptions of lexemes, paradigms or derivations
from the file or files specified by fnames.
Return a list of descriptions.
"""
if type(fnames) == str:
fnames = [fnames]
descrs = []
for fname in fnames:
descrs += yamlReader.read_file(fname, self.errorHandler)
return descrs
def load_yaml_descrs(self, fnames):
"""
Load raw descriptions of lexemes, paradigms or derivations
from the file or files specified by fnames.
Return a list of descriptions.
"""
if type(fnames) == str:
fnames = [fnames]
descrs = []
for fname in fnames:
descrs += yamlReader.read_file(fname, self.errorHandler)
return descrs
def load_stem_conversions(self, fname):
"""Load stem conversion rules from a file.
Return the number of rules loaded."""
if len(self.lexemes) > 0:
self.raise_error(u'Loading stem conversions should occur before ' +\
u'loading stems.')
return 0
conversionDescrs = yamlReader.read_file(fname, self.errorHandler)
self.stemConversions = {} # {conversion name -> StemConversion}
for dictSC in conversionDescrs:
sc = StemConversion(dictSC, self.errorHandler)
self.stemConversions[sc.name] = sc
return len(self.stemConversions)
def load_derivations(self, fname):
"""Load derivations from a file.
Return the number of derivations loaded."""
derivDescrs = yamlReader.read_file(fname, self.errorHandler)
for dictDescr in derivDescrs:
## self.derivations[u'#deriv#' + dictDescr[u'value']] =\
## Paradigm(dictDescr, self.errorHandler)
dictDescr[u'value'] = u'#deriv#' + dictDescr[u'value']
self.derivations[dictDescr[u'value']] =\
Derivation(dictDescr, self.errorHandler)
for paradigm in self.paradigms.values():
derivations.deriv_for_paradigm(paradigm)
for derivName, deriv in self.derivations.iteritems():
if derivName.startswith(u'#deriv#paradigm#'):
deriv.build_links()
print derivName + u': build complete.'
#print unicode(self.derivations[u'#deriv#paradigm#Nctt'])
deriv.extend_leaves()
print derivName + u': leaves extended.'
#print unicode(deriv)
#print unicode(self.derivations[u'#deriv#N-fӕ#paradigm#Nct'])
print u'Leaves extended.'
#print unicode(self.derivations[u'#deriv#paradigm#Nct'])
for derivName, deriv in self.derivations.iteritems():
p = deriv.to_paradigm()
self.paradigms[derivName] = p
for derivName in self.derivations:
print u'Compiling ' + derivName + u'... ',
self.paradigms[derivName].compile_paradigm()
print u'compiled.'
gc.collect()
if derivName == u'#deriv#paradigm#Nctt':
fPara = codecs.open(u'test-ossetic/deriv-Nctt-test.txt', 'w', 'utf-8-sig')
for f in self.paradigms[derivName].flex:
fPara.write(unicode(f))
fPara.close()
print u'Derivations compiled.'
for lex in self.lexemes:
lex.add_derivations()
return len(self.derivations)
def load_paradigms(self, fname, pLst=None):
"""Load paradigms from a file.
Return the number of paradigms loaded."""
if len(self.lexemes) > 0:
self.raise_error(u'Loading paradigms should occur before ' +\
u'loading stems.')
return 0
paraDescrs = yamlReader.read_file(fname, self.errorHandler)
for dictDescr in paraDescrs:
if sys.getsizeof(self.paradigms) > self.PARADIGM_MEMORY_LIMIT:
self.raise_error(u'Not enough memory for the paradigms.')
return
self.paradigms[dictDescr[u'value']] =\
Paradigm(dictDescr, self.errorHandler)
newParadigms = {}
for pName, p in self.paradigms.iteritems():
if pLst is None or pName in pLst:
p = copy.deepcopy(p)
p.compile_paradigm()
newParadigms[pName] = p
self.paradigms = newParadigms
return len(self.paradigms)
