{{ message }}
-
-
Notifications
You must be signed in to change notification settings - Fork 35.4k
Expand file tree
/
Copy pathHyperParser.py
More file actions
313 lines (268 loc) · 12.6 KB
/
Copy pathHyperParser.py
File metadata and controls
313 lines (268 loc) · 12.6 KB
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
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
"""Provide advanced parsing abilities for ParenMatch and other extensions.
HyperParser uses PyParser. PyParser mostly gives information on the
proper indentation of code. HyperParser gives additional information on
the structure of code.
"""
import string
from keyword import iskeyword
from idlelib import PyParse
# all ASCII chars that may be in an identifier
_ASCII_ID_CHARS = frozenset(string.ascii_letters + string.digits + "_")
# all ASCII chars that may be the first char of an identifier
_ASCII_ID_FIRST_CHARS = frozenset(string.ascii_letters + "_")
# lookup table for whether 7-bit ASCII chars are valid in a Python identifier
_IS_ASCII_ID_CHAR = [(chr(x) in _ASCII_ID_CHARS) for x in range(128)]
# lookup table for whether 7-bit ASCII chars are valid as the first
# char in a Python identifier
_IS_ASCII_ID_FIRST_CHAR = \
[(chr(x) in _ASCII_ID_FIRST_CHARS) for x in range(128)]
class HyperParser:
def __init__(self, editwin, index):
"To initialize, analyze the surroundings of the given index."
self.editwin = editwin
self.text = text = editwin.text
parser = PyParse.Parser(editwin.indentwidth, editwin.tabwidth)
def index2line(index):
return int(float(index))
lno = index2line(text.index(index))
if not editwin.context_use_ps1:
for context in editwin.num_context_lines:
startat = max(lno - context, 1)
startatindex = repr(startat) + ".0"
stopatindex = "%d.end" % lno
# We add the newline because PyParse requires a newline
# at end. We add a space so that index won't be at end
# of line, so that its status will be the same as the
# char before it, if should.
parser.set_str(text.get(startatindex, stopatindex)+' \n')
bod = parser.find_good_parse_start(
editwin._build_char_in_string_func(startatindex))
if bod is not None or startat == 1:
break
parser.set_lo(bod or 0)
else:
r = text.tag_prevrange("console", index)
if r:
startatindex = r[1]
else:
startatindex = "1.0"
stopatindex = "%d.end" % lno
# We add the newline because PyParse requires it. We add a
# space so that index won't be at end of line, so that its
# status will be the same as the char before it, if should.
parser.set_str(text.get(startatindex, stopatindex)+' \n')
parser.set_lo(0)
# We want what the parser has, minus the last newline and space.
self.rawtext = parser.str[:-2]
# Parser.str apparently preserves the statement we are in, so
# that stopatindex can be used to synchronize the string with
# the text box indices.
self.stopatindex = stopatindex
self.bracketing = parser.get_last_stmt_bracketing()
# find which pairs of bracketing are openers. These always
# correspond to a character of rawtext.
self.isopener = [i>0 and self.bracketing[i][1] >
self.bracketing[i-1][1]
for i in range(len(self.bracketing))]
self.set_index(index)
def set_index(self, index):
"""Set the index to which the functions relate.
The index must be in the same statement.
"""
indexinrawtext = (len(self.rawtext) -
len(self.text.get(index, self.stopatindex)))
if indexinrawtext < 0:
raise ValueError("Index %s precedes the analyzed statement"
% index)
self.indexinrawtext = indexinrawtext
# find the rightmost bracket to which index belongs
self.indexbracket = 0
while (self.indexbracket < len(self.bracketing)-1 and
self.bracketing[self.indexbracket+1][0] < self.indexinrawtext):
self.indexbracket += 1
if (self.indexbracket < len(self.bracketing)-1 and
self.bracketing[self.indexbracket+1][0] == self.indexinrawtext and
not self.isopener[self.indexbracket+1]):
self.indexbracket += 1
def is_in_string(self):
"""Is the index given to the HyperParser in a string?"""
# The bracket to which we belong should be an opener.
# If it's an opener, it has to have a character.
return (self.isopener[self.indexbracket] and
self.rawtext[self.bracketing[self.indexbracket][0]]
in ('"', "'"))
def is_in_code(self):
"""Is the index given to the HyperParser in normal code?"""
return (not self.isopener[self.indexbracket] or
self.rawtext[self.bracketing[self.indexbracket][0]]
not in ('#', '"', "'"))
def get_surrounding_brackets(self, openers='([{', mustclose=False):
"""Return bracket indexes or None.
If the index given to the HyperParser is surrounded by a
bracket defined in openers (or at least has one before it),
return the indices of the opening bracket and the closing
bracket (or the end of line, whichever comes first).
If it is not surrounded by brackets, or the end of line comes
before the closing bracket and mustclose is True, returns None.
"""
bracketinglevel = self.bracketing[self.indexbracket][1]
before = self.indexbracket
while (not self.isopener[before] or
self.rawtext[self.bracketing[before][0]] not in openers or
self.bracketing[before][1] > bracketinglevel):
before -= 1
if before < 0:
return None
bracketinglevel = min(bracketinglevel, self.bracketing[before][1])
after = self.indexbracket + 1
while (after < len(self.bracketing) and
self.bracketing[after][1] >= bracketinglevel):
after += 1
beforeindex = self.text.index("%s-%dc" %
(self.stopatindex, len(self.rawtext)-self.bracketing[before][0]))
if (after >= len(self.bracketing) or
self.bracketing[after][0] > len(self.rawtext)):
if mustclose:
return None
afterindex = self.stopatindex
else:
# We are after a real char, so it is a ')' and we give the
# index before it.
afterindex = self.text.index(
"%s-%dc" % (self.stopatindex,
len(self.rawtext)-(self.bracketing[after][0]-1)))
return beforeindex, afterindex
# the set of built-in identifiers which are also keywords,
# i.e. keyword.iskeyword() returns True for them
_ID_KEYWORDS = frozenset({"True", "False", "None"})
@classmethod
def _eat_identifier(cls, str, limit, pos):
"""Given a string and pos, return the number of chars in the
identifier which ends at pos, or 0 if there is no such one.
This ignores non-identifier eywords are not identifiers.
"""
is_ascii_id_char = _IS_ASCII_ID_CHAR
# Start at the end (pos) and work backwards.
i = pos
# Go backwards as long as the characters are valid ASCII
# identifier characters. This is an optimization, since it
# is faster in the common case where most of the characters
# are ASCII.
while i > limit and (
ord(str[i - 1]) < 128 and
is_ascii_id_char[ord(str[i - 1])]
):
i -= 1
# If the above loop ended due to reaching a non-ASCII
# character, continue going backwards using the most generic
# test for whether a string contains only valid identifier
# characters.
if i > limit and ord(str[i - 1]) >= 128:
while i - 4 >= limit and ('a' + str[i - 4:pos]).isidentifier():
i -= 4
if i - 2 >= limit and ('a' + str[i - 2:pos]).isidentifier():
i -= 2
if i - 1 >= limit and ('a' + str[i - 1:pos]).isidentifier():
i -= 1
# The identifier candidate starts here. If it isn't a valid
# identifier, don't eat anything. At this point that is only
# possible if the first character isn't a valid first
# character for an identifier.
if not str[i:pos].isidentifier():
return 0
elif i < pos:
# All characters in str[i:pos] are valid ASCII identifier
# characters, so it is enough to check that the first is
# valid as the first character of an identifier.
if not _IS_ASCII_ID_FIRST_CHAR[ord(str[i])]:
return 0
# All keywords are valid identifiers, but should not be
# considered identifiers here, except for True, False and None.
if i < pos and (
iskeyword(str[i:pos]) and
str[i:pos] not in cls._ID_KEYWORDS
):
return 0
return pos - i
# This string includes all chars that may be in a white space
_whitespace_chars = " \t\n\\"
def get_expression(self):
"""Return a string with the Python expression which ends at the
given index, which is empty if there is no real one.
"""
if not self.is_in_code():
raise ValueError("get_expression should only be called"
"if index is inside a code.")
rawtext = self.rawtext
bracketing = self.bracketing
brck_index = self.indexbracket
brck_limit = bracketing[brck_index][0]
pos = self.indexinrawtext
last_identifier_pos = pos
postdot_phase = True
while 1:
# Eat whitespaces, comments, and if postdot_phase is False - a dot
while 1:
if pos>brck_limit and rawtext[pos-1] in self._whitespace_chars:
# Eat a whitespace
pos -= 1
elif (not postdot_phase and
pos > brck_limit and rawtext[pos-1] == '.'):
# Eat a dot
pos -= 1
postdot_phase = True
# The next line will fail if we are *inside* a comment,
# but we shouldn't be.
elif (pos == brck_limit and brck_index > 0 and
rawtext[bracketing[brck_index-1][0]] == '#'):
# Eat a comment
brck_index -= 2
brck_limit = bracketing[brck_index][0]
pos = bracketing[brck_index+1][0]
else:
# If we didn't eat anything, quit.
break
if not postdot_phase:
# We didn't find a dot, so the expression end at the
# last identifier pos.
break
ret = self._eat_identifier(rawtext, brck_limit, pos)
if ret:
# There is an identifier to eat
pos = pos - ret
last_identifier_pos = pos
# Now, to continue the search, we must find a dot.
postdot_phase = False
# (the loop continues now)
elif pos == brck_limit:
# We are at a bracketing limit. If it is a closing
# bracket, eat the bracket, otherwise, stop the search.
level = bracketing[brck_index][1]
while brck_index > 0 and bracketing[brck_index-1][1] > level:
brck_index -= 1
if bracketing[brck_index][0] == brck_limit:
# We were not at the end of a closing bracket
break
pos = bracketing[brck_index][0]
brck_index -= 1
brck_limit = bracketing[brck_index][0]
last_identifier_pos = pos
if rawtext[pos] in "([":
# [] and () may be used after an identifier, so we
# continue. postdot_phase is True, so we don't allow a dot.
pass
else:
# We can't continue after other types of brackets
if rawtext[pos] in "'\"":
# Scan a string prefix
while pos > 0 and rawtext[pos - 1] in "rRbBuU":
pos -= 1
last_identifier_pos = pos
break
else:
# We've found an operator or something.
break
return rawtext[last_identifier_pos:self.indexinrawtext]
if __name__ == '__main__':
import unittest
unittest.main('idlelib.idle_test.test_hyperparser', verbosity=2)
You can’t perform that action at this time.