def test_double_metaphone():
from whoosh.lang.dmetaphone import double_metaphone
names = {'maurice': ('MRS', None),
'aubrey': ('APR', None),
'cambrillo': ('KMPRL', 'KMPR'),
'heidi': ('HT', None),
'katherine': ('K0RN', 'KTRN'),
'Thumbail': ('0MPL', 'TMPL'),
'catherine': ('K0RN', 'KTRN'),
'richard': ('RXRT', 'RKRT'),
'bob': ('PP', None),
'eric': ('ARK', None),
'geoff': ('JF', 'KF'),
'Through': ('0R', 'TR'),
'Schwein': ('XN', 'XFN'),
'dave': ('TF', None),
'ray': ('R', None),
'steven': ('STFN', None),
'bryce': ('PRS', None),
'randy': ('RNT', None),
'bryan': ('PRN', None),
'Rapelje': ('RPL', None),
'brian': ('PRN', None),
'otto': ('AT', None),
'auto': ('AT', None),
'Dallas': ('TLS', None),
'maisey': ('MS', None),
'zhang': ('JNK', None),
'Chile': ('XL', None),
'Jose': ('HS', None),
'Arnow': ('ARN', 'ARNF'),
'solilijs': ('SLLS', None),
'Parachute': ('PRKT', None),
'Nowhere': ('NR', None),
'Tux': ('TKS', None)}
dmn = name = None
for name in names.keys():
dmn = double_metaphone(name)
assert dmn == names[name]
mf = (analysis.RegexTokenizer()
| analysis.LowercaseFilter()
| analysis.DoubleMetaphoneFilter())
results = [(t.text, t.boost) for t in mf(u("Spruce View"))]
assert results == [('SPRS', 1.0), ('F', 1.0), ('FF', 0.5)]
mf = (analysis.RegexTokenizer()
| analysis.LowercaseFilter()
| analysis.DoubleMetaphoneFilter(combine=True))
results = [(t.text, t.boost) for t in mf(u("Spruce View"))]
assert results == [('spruce', 1.0), ('SPRS', 1.0), ('view', 1.0),
('F', 1.0), ('FF', 0.5)]
namefield = fields.TEXT(analyzer=mf)
texts = list(namefield.process_text(u("Spruce View"), mode="query"))
assert texts == [u('spruce'), 'SPRS', u('view'), 'F', 'FF']
def test_double_metaphone():
from whoosh.lang.dmetaphone import double_metaphone
names = {'maurice': ('MRS', None),
'aubrey': ('APR', None),
'cambrillo': ('KMPRL', 'KMPR'),
'heidi': ('HT', None),
'katherine': ('K0RN', 'KTRN'),
'Thumbail': ('0MPL', 'TMPL'),
'catherine': ('K0RN', 'KTRN'),
'richard': ('RXRT', 'RKRT'),
'bob': ('PP', None),
'eric': ('ARK', None),
'geoff': ('JF', 'KF'),
'Through': ('0R', 'TR'),
'Schwein': ('XN', 'XFN'),
'dave': ('TF', None),
'ray': ('R', None),
'steven': ('STFN', None),
'bryce': ('PRS', None),
'randy': ('RNT', None),
'bryan': ('PRN', None),
'Rapelje': ('RPL', None),
'brian': ('PRN', None),
'otto': ('AT', None),
'auto': ('AT', None),
'Dallas': ('TLS', None),
'maisey': ('MS', None),
'zhang': ('JNK', None),
'Chile': ('XL', None),
'Jose': ('HS', None),
'Arnow': ('ARN', 'ARNF'),
'solilijs': ('SLLS', None),
'Parachute': ('PRKT', None),
'Nowhere': ('NR', None),
'Tux': ('TKS', None)}
dmn = name = None
for name in names.keys():
dmn = double_metaphone(name)
assert dmn == names[name]
mf = (analysis.RegexTokenizer()
| analysis.LowercaseFilter()
| analysis.DoubleMetaphoneFilter())
results = [(t.text, t.boost) for t in mf(u("Spruce View"))]
assert results == [('SPRS', 1.0), ('F', 1.0), ('FF', 0.5)]
mf = (analysis.RegexTokenizer()
| analysis.LowercaseFilter()
| analysis.DoubleMetaphoneFilter(combine=True))
results = [(t.text, t.boost) for t in mf(u("Spruce View"))]
assert results == [('spruce', 1.0), ('SPRS', 1.0), ('view', 1.0),
('F', 1.0), ('FF', 0.5)]
namefield = fields.TEXT(analyzer=mf)
texts = list(namefield.process_text(u("Spruce View"), mode="query"))
assert texts == [u('spruce'), 'SPRS', u('view'), 'F', 'FF']
def __call__(self, tokens):
primary_boost = self.primary_boost
secondary_boost = self.secondary_boost
combine = self.combine
for t in tokens:
if combine:
yield t
primary, secondary = double_metaphone(t.text)
b = t.boost
# Overwrite the token's text and boost and yield it
if primary:
t.text = primary
t.boost = b * primary_boost
yield t
if secondary:
t.text = secondary
t.boost = b * secondary_boost
yield t
def __call__(self, tokens):
primary_boost = self.primary_boost
secondary_boost = self.secondary_boost
combine = self.combine
for t in tokens:
if combine:
yield t
primary, secondary = double_metaphone(t.text)
b = t.boost
# Overwrite the token's text and boost and yield it
if primary:
t.text = primary
t.boost = b * primary_boost
yield t
if secondary:
t.text = secondary
t.boost = b * secondary_boost
yield t
