Commit c845e3aa authored by Nong Hoang Tu's avatar Nong Hoang Tu
Browse files

Update upstream source from tag 'upstream/1.6.4'

Update to upstream version '1.6.4'
with Debian dir 21300c2c32c01fb0384c234535a11c4222970615
parents 8e3025a4 45160b25
This diff is collapsed.
......@@ -25,13 +25,6 @@
# 02110-1301 USA
######################### END LICENSE BLOCK #########################
from .compat import wrap_ord
NUM_OF_CATEGORY = 6
DONT_KNOW = -1
ENOUGH_REL_THRESHOLD = 100
MAX_REL_THRESHOLD = 1000
MINIMUM_DATA_THRESHOLD = 4
# This is hiragana 2-char sequence table, the number in each cell represents its frequency category
jp2CharContext = (
......@@ -120,24 +113,35 @@ jp2CharContext = (
(0,4,0,3,0,3,0,3,0,3,5,5,3,3,3,3,4,3,4,3,3,3,4,4,4,3,3,3,3,4,3,5,3,3,1,3,2,4,5,5,5,5,4,3,4,5,5,3,2,2,3,3,3,3,2,3,3,1,2,3,2,4,3,3,3,4,0,4,0,2,0,4,3,2,2,1,2,0,3,0,0,4,1),
)
class JapaneseContextAnalysis:
class JapaneseContextAnalysis(object):
NUM_OF_CATEGORY = 6
DONT_KNOW = -1
ENOUGH_REL_THRESHOLD = 100
MAX_REL_THRESHOLD = 1000
MINIMUM_DATA_THRESHOLD = 4
def __init__(self):
self._total_rel = None
self._rel_sample = None
self._need_to_skip_char_num = None
self._last_char_order = None
self._done = None
self.reset()
def reset(self):
self._mTotalRel = 0 # total sequence received
# category counters, each interger counts sequence in its category
self._mRelSample = [0] * NUM_OF_CATEGORY
self._total_rel = 0 # total sequence received
# category counters, each integer counts sequence in its category
self._rel_sample = [0] * self.NUM_OF_CATEGORY
# if last byte in current buffer is not the last byte of a character,
# we need to know how many bytes to skip in next buffer
self._mNeedToSkipCharNum = 0
self._mLastCharOrder = -1 # The order of previous char
self._need_to_skip_char_num = 0
self._last_char_order = -1 # The order of previous char
# If this flag is set to True, detection is done and conclusion has
# been made
self._mDone = False
self._done = False
def feed(self, aBuf, aLen):
if self._mDone:
def feed(self, byte_str, num_bytes):
if self._done:
return
# The buffer we got is byte oriented, and a character may span in more than one
......@@ -147,81 +151,83 @@ class JapaneseContextAnalysis:
# well and analyse the character once it is complete, but since a
# character will not make much difference, by simply skipping
# this character will simply our logic and improve performance.
i = self._mNeedToSkipCharNum
while i < aLen:
order, charLen = self.get_order(aBuf[i:i + 2])
i += charLen
if i > aLen:
self._mNeedToSkipCharNum = i - aLen
self._mLastCharOrder = -1
i = self._need_to_skip_char_num
while i < num_bytes:
order, char_len = self.get_order(byte_str[i:i + 2])
i += char_len
if i > num_bytes:
self._need_to_skip_char_num = i - num_bytes
self._last_char_order = -1
else:
if (order != -1) and (self._mLastCharOrder != -1):
self._mTotalRel += 1
if self._mTotalRel > MAX_REL_THRESHOLD:
self._mDone = True
if (order != -1) and (self._last_char_order != -1):
self._total_rel += 1
if self._total_rel > self.MAX_REL_THRESHOLD:
self._done = True
break
self._mRelSample[jp2CharContext[self._mLastCharOrder][order]] += 1
self._mLastCharOrder = order
self._rel_sample[jp2CharContext[self._last_char_order][order]] += 1
self._last_char_order = order
def got_enough_data(self):
return self._mTotalRel > ENOUGH_REL_THRESHOLD
return self._total_rel > self.ENOUGH_REL_THRESHOLD
def get_confidence(self):
# This is just one way to calculate confidence. It works well for me.
if self._mTotalRel > MINIMUM_DATA_THRESHOLD:
return (self._mTotalRel - self._mRelSample[0]) / self._mTotalRel
if self._total_rel > self.MINIMUM_DATA_THRESHOLD:
return (self._total_rel - self._rel_sample[0]) / self._total_rel
else:
return DONT_KNOW
return self.DONT_KNOW
def get_order(self, aBuf):
def get_order(self, byte_str):
return -1, 1
class SJISContextAnalysis(JapaneseContextAnalysis):
def __init__(self):
self.charset_name = "SHIFT_JIS"
super(SJISContextAnalysis, self).__init__()
self._charset_name = "SHIFT_JIS"
def get_charset_name(self):
return self.charset_name
@property
def charset_name(self):
return self._charset_name
def get_order(self, aBuf):
if not aBuf:
def get_order(self, byte_str):
if not byte_str:
return -1, 1
# find out current char's byte length
first_char = wrap_ord(aBuf[0])
if ((0x81 <= first_char <= 0x9F) or (0xE0 <= first_char <= 0xFC)):
charLen = 2
first_char = byte_str[0]
if (0x81 <= first_char <= 0x9F) or (0xE0 <= first_char <= 0xFC):
char_len = 2
if (first_char == 0x87) or (0xFA <= first_char <= 0xFC):
self.charset_name = "CP932"
self._charset_name = "CP932"
else:
charLen = 1
char_len = 1
# return its order if it is hiragana
if len(aBuf) > 1:
second_char = wrap_ord(aBuf[1])
if len(byte_str) > 1:
second_char = byte_str[1]
if (first_char == 202) and (0x9F <= second_char <= 0xF1):
return second_char - 0x9F, charLen
return second_char - 0x9F, char_len
return -1, charLen
return -1, char_len
class EUCJPContextAnalysis(JapaneseContextAnalysis):
def get_order(self, aBuf):
if not aBuf:
def get_order(self, byte_str):
if not byte_str:
return -1, 1
# find out current char's byte length
first_char = wrap_ord(aBuf[0])
first_char = byte_str[0]
if (first_char == 0x8E) or (0xA1 <= first_char <= 0xFE):
charLen = 2
char_len = 2
elif first_char == 0x8F:
charLen = 3
char_len = 3
else:
charLen = 1
char_len = 1
# return its order if it is hiragana
if len(aBuf) > 1:
second_char = wrap_ord(aBuf[1])
if len(byte_str) > 1:
second_char = byte_str[1]
if (first_char == 0xA4) and (0xA1 <= second_char <= 0xF3):
return second_char - 0xA1, charLen
return second_char - 0xA1, char_len
return -1, char_len
return -1, charLen
# flake8: noqa
......@@ -210,20 +210,19 @@ BulgarianLangModel = (
)
Latin5BulgarianModel = {
'charToOrderMap': Latin5_BulgarianCharToOrderMap,
'precedenceMatrix': BulgarianLangModel,
'mTypicalPositiveRatio': 0.969392,
'keepEnglishLetter': False,
'charsetName': "ISO-8859-5"
'char_to_order_map': Latin5_BulgarianCharToOrderMap,
'precedence_matrix': BulgarianLangModel,
'typical_positive_ratio': 0.969392,
'keep_english_letter': False,
'charset_name': "ISO-8859-5",
'language': 'Bulgairan',
}
Win1251BulgarianModel = {
'charToOrderMap': win1251BulgarianCharToOrderMap,
'precedenceMatrix': BulgarianLangModel,
'mTypicalPositiveRatio': 0.969392,
'keepEnglishLetter': False,
'charsetName': "windows-1251"
'char_to_order_map': win1251BulgarianCharToOrderMap,
'precedence_matrix': BulgarianLangModel,
'typical_positive_ratio': 0.969392,
'keep_english_letter': False,
'charset_name': "windows-1251",
'language': 'Bulgarian',
}
# flake8: noqa
......@@ -27,7 +27,7 @@
# KOI8-R language model
# Character Mapping Table:
KOI8R_CharToOrderMap = (
KOI8R_char_to_order_map = (
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, # 00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, # 10
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, # 20
......@@ -46,7 +46,7 @@ KOI8R_CharToOrderMap = (
35, 43, 45, 32, 40, 52, 56, 33, 61, 62, 51, 57, 47, 63, 50, 70, # f0
)
win1251_CharToOrderMap = (
win1251_char_to_order_map = (
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, # 00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, # 10
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, # 20
......@@ -65,7 +65,7 @@ win1251_CharToOrderMap = (
9, 7, 6, 14, 39, 26, 28, 22, 25, 29, 54, 18, 17, 30, 27, 16,
)
latin5_CharToOrderMap = (
latin5_char_to_order_map = (
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, # 00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, # 10
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, # 20
......@@ -84,7 +84,7 @@ latin5_CharToOrderMap = (
239, 68,240,241,242,243,244,245,246,247,248,249,250,251,252,255,
)
macCyrillic_CharToOrderMap = (
macCyrillic_char_to_order_map = (
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, # 00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, # 10
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, # 20
......@@ -103,7 +103,7 @@ macCyrillic_CharToOrderMap = (
9, 7, 6, 14, 39, 26, 28, 22, 25, 29, 54, 18, 17, 30, 27,255,
)
IBM855_CharToOrderMap = (
IBM855_char_to_order_map = (
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, # 00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, # 10
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, # 20
......@@ -122,7 +122,7 @@ IBM855_CharToOrderMap = (
250, 18, 62, 20, 51, 25, 57, 30, 47, 29, 63, 22, 50,251,252,255,
)
IBM866_CharToOrderMap = (
IBM866_char_to_order_map = (
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, # 00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, # 10
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, # 20
......@@ -279,51 +279,55 @@ RussianLangModel = (
)
Koi8rModel = {
'charToOrderMap': KOI8R_CharToOrderMap,
'precedenceMatrix': RussianLangModel,
'mTypicalPositiveRatio': 0.976601,
'keepEnglishLetter': False,
'charsetName': "KOI8-R"
'char_to_order_map': KOI8R_char_to_order_map,
'precedence_matrix': RussianLangModel,
'typical_positive_ratio': 0.976601,
'keep_english_letter': False,
'charset_name': "KOI8-R",
'language': 'Russian',
}
Win1251CyrillicModel = {
'charToOrderMap': win1251_CharToOrderMap,
'precedenceMatrix': RussianLangModel,
'mTypicalPositiveRatio': 0.976601,
'keepEnglishLetter': False,
'charsetName': "windows-1251"
'char_to_order_map': win1251_char_to_order_map,
'precedence_matrix': RussianLangModel,
'typical_positive_ratio': 0.976601,
'keep_english_letter': False,
'charset_name': "windows-1251",
'language': 'Russian',
}
Latin5CyrillicModel = {
'charToOrderMap': latin5_CharToOrderMap,
'precedenceMatrix': RussianLangModel,
'mTypicalPositiveRatio': 0.976601,
'keepEnglishLetter': False,
'charsetName': "ISO-8859-5"
'char_to_order_map': latin5_char_to_order_map,
'precedence_matrix': RussianLangModel,
'typical_positive_ratio': 0.976601,
'keep_english_letter': False,
'charset_name': "ISO-8859-5",
'language': 'Russian',
}
MacCyrillicModel = {
'charToOrderMap': macCyrillic_CharToOrderMap,
'precedenceMatrix': RussianLangModel,
'mTypicalPositiveRatio': 0.976601,
'keepEnglishLetter': False,
'charsetName': "MacCyrillic"
};
'char_to_order_map': macCyrillic_char_to_order_map,
'precedence_matrix': RussianLangModel,
'typical_positive_ratio': 0.976601,
'keep_english_letter': False,
'charset_name': "MacCyrillic",
'language': 'Russian',
}
Ibm866Model = {
'charToOrderMap': IBM866_CharToOrderMap,
'precedenceMatrix': RussianLangModel,
'mTypicalPositiveRatio': 0.976601,
'keepEnglishLetter': False,
'charsetName': "IBM866"
'char_to_order_map': IBM866_char_to_order_map,
'precedence_matrix': RussianLangModel,
'typical_positive_ratio': 0.976601,
'keep_english_letter': False,
'charset_name': "IBM866",
'language': 'Russian',
}
Ibm855Model = {
'charToOrderMap': IBM855_CharToOrderMap,
'precedenceMatrix': RussianLangModel,
'mTypicalPositiveRatio': 0.976601,
'keepEnglishLetter': False,
'charsetName': "IBM855"
'char_to_order_map': IBM855_char_to_order_map,
'precedence_matrix': RussianLangModel,
'typical_positive_ratio': 0.976601,
'keep_english_letter': False,
'charset_name': "IBM855",
'language': 'Russian',
}
# flake8: noqa
......@@ -31,7 +31,7 @@
# 252: 0 - 9
# Character Mapping Table:
Latin7_CharToOrderMap = (
Latin7_char_to_order_map = (
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, # 00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, # 10
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, # 20
......@@ -50,7 +50,7 @@ Latin7_CharToOrderMap = (
9, 8, 14, 7, 2, 12, 28, 23, 42, 24, 64, 75, 19, 26, 27,253, # f0
)
win1253_CharToOrderMap = (
win1253_char_to_order_map = (
255,255,255,255,255,255,255,255,255,255,254,255,255,254,255,255, # 00
255,255,255,255,255,255,255,255,255,255,255,255,255,255,255,255, # 10
253,253,253,253,253,253,253,253,253,253,253,253,253,253,253,253, # 20
......@@ -207,19 +207,19 @@ GreekLangModel = (
)
Latin7GreekModel = {
'charToOrderMap': Latin7_CharToOrderMap,
'precedenceMatrix': GreekLangModel,
'mTypicalPositiveRatio': 0.982851,
'keepEnglishLetter': False,
'charsetName': "ISO-8859-7"
'char_to_order_map': Latin7_char_to_order_map,
'precedence_matrix': GreekLangModel,
'typical_positive_ratio': 0.982851,
'keep_english_letter': False,
'charset_name': "ISO-8859-7",
'language': 'Greek',
}
Win1253GreekModel = {
'charToOrderMap': win1253_CharToOrderMap,
'precedenceMatrix': GreekLangModel,
'mTypicalPositiveRatio': 0.982851,
'keepEnglishLetter': False,
'charsetName': "windows-1253"
'char_to_order_map': win1253_char_to_order_map,
'precedence_matrix': GreekLangModel,
'typical_positive_ratio': 0.982851,
'keep_english_letter': False,
'charset_name': "windows-1253",
'language': 'Greek',
}
# flake8: noqa
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment