def register(self, id, instance=None):
""" map an id to an instance """
if self.data.has_key(id):
pass
#raise RegistryException, '"%s" already registered' % id
for iface in self.ifaces:
try:
if isinstance(instance, InstanceType):
verifyClass(iface, instance.__class__)
else:
verifyClass(iface, instance)
except:
LOG('TextIndexNG', WARNING,
'interface broken for %s' % str(instance),
error=sys.exc_info())
raise
self.data[id] = instance
def __init__(self, id, extra=None, caller=None):
def _get(o, k, default):
""" return a value for a given key of a dict/record 'o' """
if isinstance(o, dict):
return o.get(k, default)
else:
return getattr(o, k, default)
self.id = id
# check parameters
if extra:
for k in extra.keys():
if not k in self._all_options:
raise TXNGError,'unknown parameter "%s"' % k
if caller is not None:
self.catalog_path = '/'.join(caller.getPhysicalPath())
else:
self.catalog_path = None
# indexed attributes
self._indexed_fields = _get(extra, 'indexed_fields', '').split(',')
self._indexed_fields = [ attr.strip() for attr in self._indexed_fields if attr ]
if not self._indexed_fields:
self._indexed_fields = [ self.id ]
# splitter to be used
self.use_splitter = _get(extra, 'use_splitter', 'TXNGSplitter')
# max len of splitted words
self.splitter_max_len= _get(extra, 'splitter_max_len', 64)
# allow single characters
self.splitter_single_chars = _get(extra,'splitter_single_chars',0)
# valid word separators
self.splitter_separators = _get(extra, 'splitter_separators','.+-_@')
# allow single characters
self.splitter_casefolding = _get(extra,'splitter_casefolding',1)
# left truncation
self.truncate_left = _get(extra, 'truncate_left', 0)
# Term autoexpansion
self.autoexpand = _get(extra, 'autoexpand', 0)
self.autoexpand_limit = _get(extra, 'autoexpand_limit', 4)
# maximum number of hits
self.numhits = _get(extra, 'numhits', 999999999)
# default maximum distance for words with near search
self.near_distance = _get(extra,'near_distance', 5)
# Stopwords: either filename or StopWord object
self.use_stopwords = _get(extra, 'use_stopwords', None) or None
if self.use_stopwords:
verifyClass(StopwordsInterface, self.use_stopwords.__class__)
# Normalizer
self.use_normalizer = _get(extra,'use_normalizer', None) or None
# use converters from the ConvertersRegistry
self.use_converters = _get(extra,'use_converters',0)
# Storage to be used
self.use_storage = _get(extra,'use_storage', 'StandardStorage')
# encoding
self.default_encoding = _get(extra,'default_encoding', 'iso-8859-15')
# check Parser
self.use_parser = _get(extra, 'use_parser','PyQueryParser')
# Thesaurus
self.use_thesaurus = _get(extra, 'use_thesaurus', None)
self.thesaurus_mode = _get(extra, 'thesaurus_mode', None)
self.use_lexicon = 'StandardLexicon'
self.clear()
