def test_words():
EXPECTED = 135093
w = cmudict.words()
COUNT = len(w)
if (COUNT != EXPECTED):
raise AssertionError(
'cmudict.raw(): Expected {0} bytes, got {1}.'.format(
EXPECTED, COUNT))
def getPsuedoKeyword(target_phone, already_present):
a = cmudict.dict()
b = cmudict.words()
found = False
for word in b:
for lst in a[word]:
for phone in lst:
if(target_phone == phone and already_present.get(word) == None):
already_present[word] = 1
return word
if(re.search(target_phone,phone) and len(target_phone) !=1 and already_present.get(word) == None):
already_present[word] =1
return word
foo = open(
"C:/Users/cdobb/AppData/Local/Programs/Python/Python38-32/Lib/site-packages/enchant/data/mingw32/share/enchant/hunspell/en_US.dic"
)
# First we are simply finding words of length 8 with 3 i letters
# -- some preprocessing of the dictionary entries is involved --
for bar in foo:
bar = bar.split("/")[0]
if "\n" in bar:
bar = bar[:-2]
bar = bar.lower()
if len(bar) == 8 and three_i(bar):
# now we look at pronunciation with cmudict...
temp = 0
try:
temp = cmudict.words().index(bar)
except:
temp = -1
if temp >= 0:
temp = cmudict.entries()[temp]
temp = temp[1]
#now check each syllable...
no_i_sound = True
for syl in temp:
if "IH" in syl or "AY" in syl:
no_i_sound = False
break
if no_i_sound:
# finally we want to get rid of names or obscure words by just checking
# to see if PyDictionary recognizes it...
meaning = pydict.meaning(bar, disable_errors=True)