Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix tests of khavee functions #854

Merged
merged 1 commit into from
Oct 23, 2023
Merged

Fix tests of khavee functions #854

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
190 changes: 139 additions & 51 deletions pythainlp/khavee/core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ def __init__(self):
KhaveeVerifier: Thai Poetry verifier
"""

def check_sara(self, word: str)-> str:
def check_sara(self, word: str) -> str:
"""
Check the vowels in the Thai word.

Expand Down Expand Up @@ -216,13 +216,17 @@ def check_marttra(self, word: str) -> str:
print(kv.check_marttra('สาว'))
# output: 'เกอว'
"""
if word[-1] == 'ร' and word[-2] in ['ต','ท'] :
if word[-1] == 'ร' and word[-2] in ['ต', 'ท']:
word = word[:-1]
word = self.handle_karun_sound_silence(word)
word = remove_tonemark(word)
if 'ำ' in word or ('ํ' in word and 'า' in word) or 'ไ' in word or 'ใ' in word:
return 'กา'
elif word[-1] in ['า','ะ','ิ','ี','ุ','ู','อ'] or ('ี' in word and 'ย' in word[-1]) or ('ื' in word and 'อ' in word[-1]):
elif (
word[-1] in ['า', 'ะ', 'ิ', 'ี', 'ุ', 'ู', 'อ'] or
('ี' in word and 'ย' in word[-1]) or
('ื' in word and 'อ' in word[-1])
):
return 'กา'
elif word[-1] in ['ง']:
return 'กง'
Expand All @@ -235,11 +239,13 @@ def check_marttra(self, word: str) -> str:
return 'เกย'
elif word[-1] in ['ว']:
return 'เกอว'
elif word[-1] in ['ก','ข','ค','ฆ']:
elif word[-1] in ['ก', 'ข', 'ค', 'ฆ']:
return 'กก'
elif word[-1] in ['จ','ช','ซ','ฎ','ฏ','ฐ','ฑ','ฒ','ด','ต','ถ','ท','ธ','ศ','ษ','ส'] :
elif word[-1] in [
'จ', 'ช', 'ซ', 'ฎ', 'ฏ', 'ฐ', 'ฑ', 'ฒ', 'ด', 'ต', 'ถ', 'ท', 'ธ', 'ศ', 'ษ', 'ส'
]:
return 'กด'
elif word[-1] in ['ญ',', ณ' ,'น' ,'ร' ,'ล' ,'ฬ']:
elif word[-1] in ['ญ', ', ณ', 'น', 'ร', 'ล', 'ฬ']:
return 'กน'
elif word[-1] in ['บ', 'ป', 'พ', 'ฟ', 'ภ']:
return 'กบ'
Expand All @@ -249,8 +255,7 @@ def check_marttra(self, word: str) -> str:
else:
return 'Cant find Marttra in this word'


def is_sumpus(self, word1: str,word2: str) -> bool:
def is_sumpus(self, word1: str, word2: str) -> bool:
"""
Check the rhyme between two words.

Expand All @@ -266,10 +271,10 @@ def is_sumpus(self, word1: str,word2: str) -> bool:

kv = KhaveeVerifier()

print(kv.is_sumpus('สรร','อัน'))
print(kv.is_sumpus('สรร', 'อัน'))
# output: True

print(kv.is_sumpus('สรร','แมว'))
print(kv.is_sumpus('สรร', 'แมว'))
# output: False
"""
marttra1 = self.check_marttra(word1)
Expand All @@ -290,13 +295,27 @@ def is_sumpus(self, word1: str,word2: str) -> bool:
marttra2 = 'กา'
return bool(marttra1 == marttra2 and sara1 == sara2)

def check_karu_lahu(self,text):
if (self.check_marttra(text) != 'กา' or (self.check_marttra(text) == 'กา' and self.check_sara(text) in ['อา','อี', 'อือ', 'อู', 'เอ', 'แอ', 'โอ', 'ออ', 'เออ', 'เอีย', 'เอือ' ,'อัว']) or self.check_sara(text) in ['อำ','ไอ','เอา']) and text not in ['บ่','ณ','ธ','ก็']:
def check_karu_lahu(self, text):
if (
(
self.check_marttra(text) != 'กา' or
(
self.check_marttra(text) == 'กา' and
self.check_sara(text) in [
'อา', 'อี', 'อือ', 'อู', 'เอ',
'แอ', 'โอ', 'ออ', 'เออ', 'เอีย',
'เอือ', 'อัว'
]
) or
self.check_sara(text) in ['อำ', 'ไอ', 'เอา']
) and
text not in ['บ่', 'ณ', 'ธ', 'ก็']
):
return 'karu'
else:
return 'lahu'

def check_klon(self, text: str,k_type: int=8) -> Union[List[str], str]:
def check_klon(self, text: str, k_type: int = 8) -> Union[List[str], str]:
"""
Check the suitability of the poem according to Thai principles.

Expand All @@ -312,11 +331,22 @@ def check_klon(self, text: str,k_type: int=8) -> Union[List[str], str]:

kv = KhaveeVerifier()

print(kv.check_klon('''ฉันชื่อหมูกรอบ ฉันชอบกินไก่ แล้วก็วิ่งไล่ หมาชื่อนํ้าทอง ลคคนเก่ง เอ๋งเอ๋งคะนอง มีคนจับจอง เขาชื่อน้องเธียร''', k_type=4))
print(kv.check_klon(
'ฉันชื่อหมูกรอบ ฉันชอบกินไก่ แล้วก็วิ่งไล่ หมาชื่อนํ้าทอง ลคคนเก่ง เอ๋งเอ๋งคะนอง \
มีคนจับจอง เขาชื่อน้องเธียร',
k_type=4
))
# output: The poem is correct according to the principle.

print(kv.check_klon('''ฉันชื่อหมูกรอบ ฉันชอบกินไก่ แล้วก็วิ่งไล่ หมาชื่อนํ้าทอง ลคคนเก่ง เอ๋งเอ๋งเสียงหมา มีคนจับจอง เขาชื่อน้องเธียร''',k_type=4))
# # -> ["Can't find rhyme between paragraphs ('หมา', 'จอง')in paragraph 2", "Can't find rhyme between paragraphs ('หมา', 'ทอง')in paragraph 2"]
print(kv.check_klon(
'ฉันชื่อหมูกรอบ ฉันชอบกินไก่ แล้วก็วิ่งไล่ หมาชื่อนํ้าทอง ลคคนเก่ง \
เอ๋งเอ๋งเสียงหมา มีคนจับจอง เขาชื่อน้องเธียร',
k_type=4
))
# output: [
"Can't find rhyme between paragraphs ('หมา', 'จอง') in paragraph 2",
"Can't find rhyme between paragraphs ('หมา', 'ทอง') in paragraph 2"
]
"""
if k_type == 8:
try:
Expand All @@ -327,34 +357,63 @@ def check_klon(self, text: str,k_type: int=8) -> Union[List[str], str]:
list_sumpus_sent3 = []
list_sumpus_sent4 = []
for i, sent in enumerate(text.split()):
sub_sent = subword_tokenize(sent,engine='dict')
sub_sent = subword_tokenize(sent, engine='dict')
if len(sub_sent) > 10:
error.append('In sentence '+str(i+2)+', there are more than 10 words. '+str(sub_sent))
if (i+1) % 4 == 1:
error.append(
'In sentence ' +
str(i + 2) +
', there are more than 10 words. ' +
str(sub_sent)
)
if (i + 1) % 4 == 1:
list_sumpus_sent1.append(sub_sent[-1])
elif (i+1) % 4 == 2:
list_sumpus_sent2h.append([sub_sent[1],sub_sent[2],sub_sent[3],sub_sent[4]])
elif (i + 1) % 4 == 2:
list_sumpus_sent2h.append(
[sub_sent[1], sub_sent[2], sub_sent[3], sub_sent[4]]
)
list_sumpus_sent2l.append(sub_sent[-1])
elif (i+1) % 4 == 3:
elif (i + 1) % 4 == 3:
list_sumpus_sent3.append(sub_sent[-1])
elif (i+1) % 4 == 0:
elif (i + 1) % 4 == 0:
list_sumpus_sent4.append(sub_sent[-1])
if len(list_sumpus_sent1) != len(list_sumpus_sent2h) or len(list_sumpus_sent2h) != len(list_sumpus_sent2l) or len(list_sumpus_sent2l) != len(list_sumpus_sent3) or len(list_sumpus_sent3) != len(list_sumpus_sent4) or len(list_sumpus_sent4) != len(list_sumpus_sent1):
if (
len(list_sumpus_sent1) != len(list_sumpus_sent2h) or
len(list_sumpus_sent2h) != len(list_sumpus_sent2l) or
len(list_sumpus_sent2l) != len(list_sumpus_sent3) or
len(list_sumpus_sent3) != len(list_sumpus_sent4) or
len(list_sumpus_sent4) != len(list_sumpus_sent1)
):
return 'The poem does not have 4 complete sentences.'
else:
for i in range(len(list_sumpus_sent1)):
countwrong = 0
for j in list_sumpus_sent2h[i]:
if self.is_sumpus(list_sumpus_sent1[i], j) is False:
countwrong +=1
if countwrong > 3:
error.append('Can\'t find rhyme between paragraphs '+str((list_sumpus_sent1[i],list_sumpus_sent2h[i]))+' in paragraph '+str(i+1))
countwrong += 1
if countwrong > 3:
error.append(
'Can\'t find rhyme between paragraphs ' +
str((list_sumpus_sent1[i], list_sumpus_sent2h[i])) +
' in paragraph ' +
str(i + 1)
)
if self.is_sumpus(list_sumpus_sent2l[i], list_sumpus_sent3[i]) is False:
# print(sumpus_sent2l,sumpus_sent3)
error.append('Can\'t find rhyme between paragraphs '+str((list_sumpus_sent2l[i],list_sumpus_sent3[i]))+' in paragraph '+str(i+1))
error.append(
'Can\'t find rhyme between paragraphs ' +
str((list_sumpus_sent2l[i], list_sumpus_sent3[i])) +
' in paragraph ' +
str(i + 1)
)
if i > 0:
if self.is_sumpus(list_sumpus_sent2l[i], list_sumpus_sent4[i-1]) is False:
error.append('Can\'t find rhyme between paragraphs '+str((list_sumpus_sent2l[i],list_sumpus_sent4[i-1]))+' in paragraph '+str(i+1))
if self.is_sumpus(
list_sumpus_sent2l[i], list_sumpus_sent4[i - 1]
) is False:
error.append(
'Can\'t find rhyme between paragraphs ' +
str((list_sumpus_sent2l[i], list_sumpus_sent4[i - 1])) +
' in paragraph ' +
str(i + 1)
)
if not error:
return 'The poem is correct according to the principle.'
else:
Expand All @@ -370,36 +429,61 @@ def check_klon(self, text: str,k_type: int=8) -> Union[List[str], str]:
list_sumpus_sent3 = []
list_sumpus_sent4 = []
for i, sent in enumerate(text.split()):
sub_sent = subword_tokenize(sent,engine='dict')
sub_sent = subword_tokenize(sent, engine='dict')
if len(sub_sent) > 5:
error.append('In sentence '+str(i+2)+', there are more than 4 words. '+str(sub_sent))
if (i+1) % 4 == 1:
error.append(
'In sentence ' +
str(i + 2) +
', there are more than 4 words. ' +
str(sub_sent)
)
if (i + 1) % 4 == 1:
list_sumpus_sent1.append(sub_sent[-1])
elif (i+1) % 4 == 2:
# print([sub_sent[1],sub_sent[2]])
list_sumpus_sent2h.append([sub_sent[1],sub_sent[2]])
elif (i + 1) % 4 == 2:
list_sumpus_sent2h.append([sub_sent[1], sub_sent[2]])
list_sumpus_sent2l.append(sub_sent[-1])
elif (i+1) % 4 == 3:
elif (i + 1) % 4 == 3:
list_sumpus_sent3.append(sub_sent[-1])
elif (i+1) % 4 == 0:
elif (i + 1) % 4 == 0:
list_sumpus_sent4.append(sub_sent[-1])
if len(list_sumpus_sent1) != len(list_sumpus_sent2h) or len(list_sumpus_sent2h) != len(list_sumpus_sent2l) or len(list_sumpus_sent2l) != len(list_sumpus_sent3) or len(list_sumpus_sent3) != len(list_sumpus_sent4) or len(list_sumpus_sent4) != len(list_sumpus_sent1):
if (
len(list_sumpus_sent1) != len(list_sumpus_sent2h) or
len(list_sumpus_sent2h) != len(list_sumpus_sent2l) or
len(list_sumpus_sent2l) != len(list_sumpus_sent3) or
len(list_sumpus_sent3) != len(list_sumpus_sent4) or
len(list_sumpus_sent4) != len(list_sumpus_sent1)
):
return 'The poem does not have 4 complete sentences.'
else:
for i in range(len(list_sumpus_sent1)):
countwrong = 0
for j in list_sumpus_sent2h[i]:
# print(list_sumpus_sent1[i],j)
if self.is_sumpus(list_sumpus_sent1[i], j) is False:
countwrong +=1
if countwrong > 1:
error.append('Can\'t find rhyme between paragraphs '+str((list_sumpus_sent1[i],list_sumpus_sent2h[i]))+'in paragraph '+str(i+1))
countwrong += 1
if countwrong > 1:
error.append(
'Can\'t find rhyme between paragraphs ' +
str((list_sumpus_sent1[i], list_sumpus_sent2h[i])) +
' in paragraph ' +
str(i + 1)
)
if self.is_sumpus(list_sumpus_sent2l[i], list_sumpus_sent3[i]) is False:
# print(sumpus_sent2l,sumpus_sent3)
error.append('Can\'t find rhyme between paragraphs '+str((list_sumpus_sent2l[i],list_sumpus_sent3[i]))+'in paragraph '+str(i+1))
error.append(
'Can\'t find rhyme between paragraphs ' +
str((list_sumpus_sent2l[i], list_sumpus_sent3[i])) +
' in paragraph ' +
str(i + 1)
)
if i > 0:
if self.is_sumpus(list_sumpus_sent2l[i], list_sumpus_sent4[i-1]) is False:
error.append('Can\'t find rhyme between paragraphs '+str((list_sumpus_sent2l[i],list_sumpus_sent4[i-1]))+' in paragraph '+str(i+1))
if self.is_sumpus(
list_sumpus_sent2l[i], list_sumpus_sent4[i - 1]
) is False:
error.append(
'Can\'t find rhyme between paragraphs ' +
str((list_sumpus_sent2l[i], list_sumpus_sent4[i - 1])) +
' in paragraph ' +
str(i + 1)
)
if not error:
return 'The poem is correct according to the principle.'
else:
Expand All @@ -410,7 +494,11 @@ def check_klon(self, text: str,k_type: int=8) -> Union[List[str], str]:
else:
return 'Something went wrong. Make sure you enter it in the correct form.'

def check_aek_too(self, text: Union[List[str], str], dead_syllable_as_aek:bool = False) -> Union[List[bool], List[str], bool, str]:
def check_aek_too(
self,
text: Union[List[str], str],
dead_syllable_as_aek: bool = False
) -> Union[List[bool], List[str], bool, str]:
"""
Checker of Thai tonal words

Expand All @@ -428,9 +516,9 @@ def check_aek_too(self, text: Union[List[str], str], dead_syllable_as_aek:bool =

# การเช็คคำเอกโท
print(kv.check_aek_too('เอง'), kv.check_aek_too('เอ่ง'), kv.check_aek_too('เอ้ง'))
## -> False, aek, too
# -> False, aek, too
print(kv.check_aek_too(['เอง', 'เอ่ง', 'เอ้ง'])) # ใช้ List ได้เหมือนกัน
## -> [False, 'aek', 'too']
# -> [False, 'aek', 'too']


"""
Expand Down
12 changes: 6 additions & 6 deletions pythainlp/khavee/example.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,27 +4,27 @@


# การเช็คสระ
print('เออ',kv.check_sara('เมอ'))
print('เออ', kv.check_sara('เมอ'))
# 'เออ'

# การเช็คมาตราตัวสะกด
print('เทอว',kv.check_marttra('เทอว'))
print('เทอว', kv.check_marttra('เทอว'))
# 'เกอว'

# การตรวจสอบคำสำผัสที่ถูกต้อง
print('สรร อัน',kv.is_sumpus('สรร','อัน'))
print('สรร อัน', kv.is_sumpus('สรร', 'อัน'))
# True

# การตรวจสอบคำสำผัสที่ผิด
print('เพื่อน ล้วน',kv.is_sumpus('เพื่อน','ล้วน'))
print('เพื่อน ล้วน', kv.is_sumpus('เพื่อน', 'ล้วน'))
# False

# การตรวจสอบคำ ครุ ลหุ
print('สรร',kv.check_karu_lahu('สรร'))
print('สรร', kv.check_karu_lahu('สรร'))
#karu

# การตรวจสอบคำ ครุ ลหุ
print('ชิชะ',kv.check_karu_lahu('ชิชะ'))
print('ชิชะ', kv.check_karu_lahu('ชิชะ'))
# lahu

# การตรวจสอบกลอน 8 ที่ถูกฉันทลักษณ์
Expand Down
Loading
Loading