1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
|
import sys, re, traceback
from string import split
repl = {}
tests = []
comp = []
modes = ["[Word]", "[word]", "[Char]", "[char]", "[code]"]
mode = "[Word]"
code = u""
language = ""
oldlinenums = {}
def prepare_for_eval(s):
s = re.sub("(affix|spell|morph|stem|option|suggest|generate)\(", r'\1(LOCALE,', s)
s = re.sub(r"word\(\s*(\d)", r'word(s[m.end():],\1', s) # word(n)
s = re.sub(r"word\(\s*-(\d)", r'wordmin(s[:m.start()],\1', s) # word(-n)
s = re.sub(r"[\\](\d)", r'm.group(\1)', s)
s = re.sub("[{]([^}]+)}", r'm.group("\1_1")', s)
return s
def mysplit(s, line, oldline, debug):
global repl
global tests
global comp
global modes
global mode
orig = s
if s[0:1] == '[':
if s.strip() in modes:
mode = s.strip()
return None
elif re.match(r"\[\w+\]$", s.strip()):
#sys.stderr.write("Unknown mode: " + s + "\n")
return oldline
dec = 0
exprep = 0 # replacement is a Python expression (beginning with sign =)
condition = False
ngroup = 0 # back reference group number that will be used for error positioning
# description
c = re.search("\s#\s", s)
com = u""
if c:
try:
c = c.start(c.lastindex - 1)
except:
c = c.start()
com = s[c+2:].strip()
if com[0:1] == "=":
com = prepare_for_eval(com)
s = s[:c]
m1 = re.search("<-", s)
m2 = re.search("-\d*>", s)
if m1 and m2:
condition = prepare_for_eval(s[m1.end(0): m2.start(0)].strip())
s = s[0:m1.start(0)] + s[m2.start(0):]
if s[0:1] == '"':
# quoted
pos = s[1:].find('"')
while pos > 0 and s[pos] == '\\':
pos = s[pos:].find('"')
s1 = s[1:pos+1]
s2 = s[pos+2:].strip()
else:
m = re.compile("-\d*>").search(s)
if not m:
m = re.compile("[_a-zA-Z][_a-zA-Z0-9]*").match(s)
if not m:
# syntax error
return oldline
s1 = m.group(0)
s2 = s[m.end(0):].strip()
# replace previous definitions
for i in repl:
ire = re.compile("[{]" + i + "}")
if re.search(ire, s2):
s2 = ire.sub(repl[i], s2)
# make named group
s2 = "(?P<" + m.group(0) + ">" + s2 + ")"
dec = 1
else:
s1 = s[0:m.start(0)].strip()
if re.match("TEST: ", s1): # deprecated
tests += [[s1[5:].strip(), s[m.end(0):].strip(), oldline]]
return None
s2 = s[m.start(0):].strip()
m = re.compile("-(\d*)>").match(s2)
if dec!= 1 and m:
s2 = s2[m.end(0):].strip()
if m.group(1):
ngroup = int(m.group(1))
elif dec!=1:
# syntax error
return oldline
if s2[0:1] == '=':
exprep = 1
if s2[0:1] == '"' and s2[-1:]=='"':
s2 = s2[1:-1]
if dec==1:
repl[s1] = s2
return None
else:
for i in repl:
s1 = re.sub("[{]" + i + "}", repl[i], s1)
# modes
if mode == "[Word]" or mode == "[word]":
if s1[0] == '^':
s1 = ur"((?<=[!?.] )|^)" + s1[1:] + ur"(?![-\w\u2013\u00AD])"
else:
s1 = ur"(?<![-\w\u2013.,\u00AD])" + s1 + ur"(?![-\w\u2013\u00AD])"
# modes with casing ([Word] or [Char])
if mode == "[Word]" or mode == "[Char]":
s1 = "(?i)" + s1
# add Unicode flag to the regex by (?u)
s1 = re.sub("[(][?][iI][)]", "(?iu)", s1)
if not re.match("[(][?][iI][uU][)]", s1):
s1 = "(?u)" + s1
else:
# casing
lu = re.compile("(?u)\w")
s3 = u""
state = 0
for i in range(0, len(s1)):
c = s1[i]
if c == "[":
state = 1
if state == 1 and c == "]":
state = 0
if c == "<" and i > 3 and s1[i-3:i]=="(?P":
state = 2
if state == 2 and c == ">":
state = 0
if c == "?" and i > 0 and s1[i-1:i]=="(":
state = 5
if state == 5 and c == ")":
state = 0
if lu.match(c) and c.islower() and state == 0:
if c=="i" and (language == "tr" or language == "az"):
s3 += u"[\u0130" + c + "]"
else:
s3 += "[" + c.upper() + c + "]"
elif lu.match(c) and c.islower() and state == 1 and s1[i+1:i+2] != "-":
if s1[i-1:i] == "-" and s1[i-2:i-1].islower(): # [a-z] -> [a-zA-Z]
s3 += c + s1[i-2:i-1].upper() + "-" + c.upper()
elif c=="i" and (language == "tr" or language == "az"):
s3 += u"\u0130" + c
else:
s3 += c.upper() + c
else:
s3 += c
if c == "\\":
state = 4
elif state == 4:
state = 0
s1 = s3
s1 = renum("[?]P<([^<_]*)>", s1, "?P<")
if exprep == 0:
s2 = re.sub("[{]([_a-zA-Z][_a-zA-Z0-9]*)}", r"\\g<\1>", s2)
s2 = renum(r"\\g<([^<_]*)>", s2, r"\\g<")
else:
s2 = prepare_for_eval(s2)
# check
if re.compile("[(][?]iu[)]").match(s1):
cap = True
sc = re.sub("[(][?]iu[)]", "(?u)", s1)
else:
cap = False
sc = s1
try:
compr = re.compile(sc)
if not condition:
comp += [[compr, s2, com, cap, line]]
except Exception as e:
raise Exception(str(e), oldline)
if debug:
return [s1, s2, com, condition, ngroup, oldline]
return [s1, s2, com, condition, ngroup]
# group renum (<groupname> -> <groupname_1> etc.)
def renum(regex, s1, beg):
j={}
mr = re.compile(regex)
m = mr.search(s1)
nl = s1.find("\\n")
while m:
# restart numbering in new lines
if nl > -1 and m.start(0) > (nl + 1):
j={}
nl = s1[m.start(0):].find("\\n")
if nl > -1:
nl = m.start(0) + nl
n = m.group(1)
if n in j:
j[n] += 1
else:
j[n] = 1
s1 = re.sub(mr, beg + n + "_" + str(j[n]) + ">", s1, 1)
m = mr.search(s1)
return s1
def cap(a, iscap):
global language
if iscap:
for i in range(0, len(a)):
if a[i][0:1] == "i":
if language == "tr" or language == "az":
a[i] = u"\u0130" + a[i][1:]
elif a[i][1:2] == "j" and language == "nl":
a[i] = "IJ" + a[i][2:]
else:
a[i] = "I" + a[i][1:]
else:
a[i] = a[i].capitalize()
return a
def c(rules, lang, debug = False):
global language
global code
global oldlinenums
language = lang
r = re.compile("[\n#]")
code = ""
dic = []
oldlinenums = {}
lines = rules.split("\n")
lines2 = []
result = {}
cm = 0
lnums = 1
for i in lines:
if i.strip() in modes:
if i.strip() == "[code]":
cm = 1
continue
else:
cm = 0
if cm == 1:
code = code + i + "\n"
elif len(i.strip()) > 0:
lines2 = lines2 + [i]
oldlinenums[i] = lnums
lnums = lnums + 1
lines = lines2
# concatenate multiline commands
# last tabulator + comment is the message
l = u""
comment = 0
for i in range(len(lines)-1, -1, -1):
if re.match("\t", lines[i]):
if not (comment and re.match("\t+#", lines[i])):
l = lines[i].strip() + " " + l
if re.search("#", lines[i]):
comment = 1
del lines[i]
elif l != "":
lnums = oldlinenums[lines[i]]
lines[i] = lines[i].strip() + " " + l
oldlinenums[lines[i]] = lnums
l = ""
comment = 0
# processing
for i in range(0, len(lines)):
if not r.match(lines[i]):
item = mysplit(lines[i].strip(), i + 1, oldlinenums[lines[i]], debug)
if item != None:
if type(item) == type(1):
raise Exception("Syntax error in line ", item)
dic = dic + [item]
result["rules"] = dic
code = re.sub(r"(?<![\\])\\n", "\n", code.encode("unicode-escape"))
result["code"] = re.sub(r"(?<![\\])\\t", "\t", code).replace(r"\\n", r"\n").replace(r"\\t", r"\t")
return result
|
