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
|
#!/usr/bin/env python3
# Copyright 2021 WebAssembly Community Group participants
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""A test case update script.
This script is a utility to update wasm-opt based lit tests with new FileCheck
patterns. It is based on LLVM's update_llc_test_checks.py script.
"""
import argparse
import glob
import os
import re
import subprocess
import sys
import tempfile
script_name = os.path.basename(__file__)
NOTICE = (';; NOTE: Assertions have been generated by {script} and should not' +
' be edited.')
RUN_LINE_RE = re.compile(r'^\s*;;\s*RUN:\s*(.*)$')
CHECK_PREFIX_RE = re.compile(r'.*--check-prefix[= ](\S+).*')
items = ['type', 'import', 'global', 'memory', 'data', 'table', 'elem', 'tag',
'export', 'start', 'func']
ITEM_RE = re.compile(r'(^\s*)\((' + '|'.join(items) + r')\s+(\$?[^\s()]*).*$',
re.MULTILINE)
def warn(msg):
print(f'WARNING: {msg}', file=sys.stderr)
def itertests(args):
"""
Yield (filename, lines) for each test specified in the command line args
"""
for pattern in args.tests:
tests = glob.glob(pattern, recursive=True)
if not tests:
warn(f'No tests matched {pattern}. Ignoring it.')
continue
for test in tests:
with open(test) as f:
lines = [line.rstrip() for line in f]
first_line = lines[0] if lines else ''
if script_name not in first_line and not args.force:
warn(f'Skipping test {test} which was not generated by '
f'{script_name}. Use -f to override.')
continue
yield test, lines
def find_run_lines(test, lines):
line_matches = [RUN_LINE_RE.match(l) for l in lines]
matches = [match.group(1) for match in line_matches if match]
if not matches:
warn(f'No RUN lines found in {test}. Ignoring.')
return []
run_lines = [matches[0]]
for line in matches[1:]:
if run_lines[-1].endswith('\\'):
run_lines[-1] = run_lines[-1].rstrip('\\') + ' ' + line
else:
run_lines.append(line)
return run_lines
def run_command(args, test, tmp, command):
env = dict(os.environ)
env['PATH'] = args.binaryen_bin + os.pathsep + env['PATH']
command = command.replace('%s', test)
command = command.replace('%t', tmp)
return subprocess.check_output(command, shell=True, env=env).decode('utf-8')
def find_end(module, start):
# Find the index one past the closing parenthesis corresponding to the first
# open parenthesis at `start`.
assert module[start] == '('
depth = 1
for end in range(start + 1, len(module)):
if depth == 0:
break
elif module[end] == '(':
depth += 1
elif module[end] == ')':
depth -= 1
return end
def split_output(module):
# Return a list of (name, [lines]) for module items
out = []
for match in ITEM_RE.finditer(module):
kind, name = match[2], match[3]
end = find_end(module, match.end(1))
lines = module[match.start():end].split('\n')
out.append(((kind, name), lines))
return out
def main():
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument(
'--binaryen-bin', dest='binaryen_bin', default='bin',
help=('Specifies the path to the Binaryen executables in the CMake build'
' directory. Default: bin/ of current directory (i.e. assume an'
' in-tree build).'))
parser.add_argument(
'--all-items', action='store_true',
help=('Emit checks for all module items, even those that do not appear'
' in the input.'))
parser.add_argument(
'-f', '--force', action='store_true',
help=('Generate FileCheck patterns even for test files whose existing '
'patterns were not generated by this script.'))
parser.add_argument(
'--dry-run', action='store_true',
help=('Print the updated test file contents instead of changing the '
'test files'))
parser.add_argument('tests', nargs='+', help='The test files to update')
args = parser.parse_args()
args.binaryen_bin = os.path.abspath(args.binaryen_bin)
tmp = tempfile.mktemp()
for test, lines in itertests(args):
all_items = args.all_items
if lines and script_name in lines[0]:
# Apply previously used options for this file
if '--all-items' in lines[0]:
all_items = True
# Skip the notice if it is already in the output
lines = lines[1:]
# List of (prefix, command)
run_list = []
for line in find_run_lines(test, lines):
commands = [cmd.strip() for cmd in line.rsplit('|', 1)]
filecheck_cmd = ''
if len(commands) > 1 and commands[1].startswith('filecheck '):
filecheck_cmd = commands[1]
commands = commands[:1]
check_prefix = ''
if filecheck_cmd.startswith('filecheck '):
prefix_match = CHECK_PREFIX_RE.match(filecheck_cmd)
if prefix_match:
check_prefix = prefix_match.group(1)
else:
check_prefix = 'CHECK'
run_list.append((check_prefix, commands[0]))
# Map check prefixes to lists of ((kind, name), [line])
output_modules = {}
for prefix, command, in run_list:
output = run_command(args, test, tmp, command)
if prefix:
output_modules[prefix] = split_output(output)
any_prefix = '|'.join(output_modules.keys())
check_line_re = re.compile(r'^\s*;;\s*(' + any_prefix +
r')(?:-NEXT|-LABEL|-NOT)?:.*$')
# Filter out whitespace between check blocks
if lines:
filtered = [lines[0]]
for i in range(1, len(lines) - 1):
if lines[i] or not check_line_re.match(lines[i - 1]) or \
not check_line_re.match(lines[i + 1]):
filtered.append(lines[i])
filtered.append(lines[-1])
lines = filtered
named_items = []
for line in lines:
match = ITEM_RE.match(line)
if match:
kind, name = match[2], match[3]
named_items.append((kind, name))
script = script_name
if all_items:
script += ' --all-items'
output_lines = [NOTICE.format(script=script)]
def emit_checks(indent, prefix, lines):
output_lines.append(f'{indent};; {prefix}: {lines[0]}')
for line in lines[1:]:
output_lines.append(f'{indent};; {prefix}-NEXT:{line}')
for line in lines:
# Skip pre-existing check lines; we will regenerate them.
if check_line_re.match(line):
continue
match = ITEM_RE.match(line)
if match:
indent, kind, name = match.groups()
for prefix, items in output_modules.items():
# If the output for this prefix contains an item with this
# name, emit all the items up to and including the matching
# item
has_item = False
for kind_name, lines in items:
if name and (kind, name) == kind_name:
has_item = True
break
if has_item:
first = True
while True:
kind_name, lines = items.pop(0)
if all_items or kind_name in named_items:
if not first:
output_lines.append('')
first = False
emit_checks(indent, prefix, lines)
if name and (kind, name) == kind_name:
break
output_lines.append(line)
# Output any remaining checks for each prefix
first = True
for prefix, items in output_modules.items():
for kind_name, lines in items:
if all_items or kind_name in named_items:
if not first:
output_lines.append('')
first = False
emit_checks('', prefix, lines)
if args.dry_run:
print('\n'.join(output_lines))
else:
with open(test, 'w') as f:
for line in output_lines:
f.write(line + '\n')
if __name__ == '__main__':
main()
|
