Show More
@@ -1,602 +1,599 b'' | |||
|
1 | 1 | #!/usr/bin/env python |
|
2 | 2 | # |
|
3 | 3 | # check-code - a style and portability checker for Mercurial |
|
4 | 4 | # |
|
5 | 5 | # Copyright 2010 Matt Mackall <mpm@selenic.com> |
|
6 | 6 | # |
|
7 | 7 | # This software may be used and distributed according to the terms of the |
|
8 | 8 | # GNU General Public License version 2 or any later version. |
|
9 | 9 | |
|
10 | 10 | """style and portability checker for Mercurial |
|
11 | 11 | |
|
12 | 12 | when a rule triggers wrong, do one of the following (prefer one from top): |
|
13 | 13 | * do the work-around the rule suggests |
|
14 | 14 | * doublecheck that it is a false match |
|
15 | 15 | * improve the rule pattern |
|
16 | 16 | * add an ignore pattern to the rule (3rd arg) which matches your good line |
|
17 | 17 | (you can append a short comment and match this, like: #re-raises, # no-py24) |
|
18 | 18 | * change the pattern to a warning and list the exception in test-check-code-hg |
|
19 | 19 | * ONLY use no--check-code for skipping entire files from external sources |
|
20 | 20 | """ |
|
21 | 21 | |
|
22 | 22 | import re, glob, os, sys |
|
23 | 23 | import keyword |
|
24 | 24 | import optparse |
|
25 | 25 | try: |
|
26 | 26 | import re2 |
|
27 | 27 | except ImportError: |
|
28 | 28 | re2 = None |
|
29 | 29 | |
|
30 | 30 | def compilere(pat, multiline=False): |
|
31 | 31 | if multiline: |
|
32 | 32 | pat = '(?m)' + pat |
|
33 | 33 | if re2: |
|
34 | 34 | try: |
|
35 | 35 | return re2.compile(pat) |
|
36 | 36 | except re2.error: |
|
37 | 37 | pass |
|
38 | 38 | return re.compile(pat) |
|
39 | 39 | |
|
40 | 40 | def repquote(m): |
|
41 | 41 | fromc = '.:' |
|
42 | 42 | tochr = 'pq' |
|
43 | 43 | def encodechr(i): |
|
44 | 44 | if i > 255: |
|
45 | 45 | return 'u' |
|
46 | 46 | c = chr(i) |
|
47 | 47 | if c in ' \n': |
|
48 | 48 | return c |
|
49 | 49 | if c.isalpha(): |
|
50 | 50 | return 'x' |
|
51 | 51 | if c.isdigit(): |
|
52 | 52 | return 'n' |
|
53 | 53 | try: |
|
54 | 54 | return tochr[fromc.find(c)] |
|
55 | 55 | except (ValueError, IndexError): |
|
56 | 56 | return 'o' |
|
57 | 57 | t = m.group('text') |
|
58 | 58 | tt = ''.join(encodechr(i) for i in xrange(256)) |
|
59 | 59 | t = t.translate(tt) |
|
60 | 60 | return m.group('quote') + t + m.group('quote') |
|
61 | 61 | |
|
62 | 62 | def reppython(m): |
|
63 | 63 | comment = m.group('comment') |
|
64 | 64 | if comment: |
|
65 | 65 | l = len(comment.rstrip()) |
|
66 | 66 | return "#" * l + comment[l:] |
|
67 | 67 | return repquote(m) |
|
68 | 68 | |
|
69 | 69 | def repcomment(m): |
|
70 | 70 | return m.group(1) + "#" * len(m.group(2)) |
|
71 | 71 | |
|
72 | 72 | def repccomment(m): |
|
73 | 73 | t = re.sub(r"((?<=\n) )|\S", "x", m.group(2)) |
|
74 | 74 | return m.group(1) + t + "*/" |
|
75 | 75 | |
|
76 | 76 | def repcallspaces(m): |
|
77 | 77 | t = re.sub(r"\n\s+", "\n", m.group(2)) |
|
78 | 78 | return m.group(1) + t |
|
79 | 79 | |
|
80 | 80 | def repinclude(m): |
|
81 | 81 | return m.group(1) + "<foo>" |
|
82 | 82 | |
|
83 | 83 | def rephere(m): |
|
84 | 84 | t = re.sub(r"\S", "x", m.group(2)) |
|
85 | 85 | return m.group(1) + t |
|
86 | 86 | |
|
87 | 87 | |
|
88 | 88 | testpats = [ |
|
89 | 89 | [ |
|
90 | 90 | (r'pushd|popd', "don't use 'pushd' or 'popd', use 'cd'"), |
|
91 | 91 | (r'\W\$?\(\([^\)\n]*\)\)', "don't use (()) or $(()), use 'expr'"), |
|
92 | 92 | (r'grep.*-q', "don't use 'grep -q', redirect to /dev/null"), |
|
93 | 93 | (r'(?<!hg )grep.* -a', "don't use 'grep -a', use in-line python"), |
|
94 | 94 | (r'sed.*-i', "don't use 'sed -i', use a temporary file"), |
|
95 | 95 | (r'\becho\b.*\\n', "don't use 'echo \\n', use printf"), |
|
96 | 96 | (r'echo -n', "don't use 'echo -n', use printf"), |
|
97 | 97 | (r'(^|\|\s*)\bwc\b[^|]*$\n(?!.*\(re\))', "filter wc output"), |
|
98 | 98 | (r'head -c', "don't use 'head -c', use 'dd'"), |
|
99 | 99 | (r'tail -n', "don't use the '-n' option to tail, just use '-<num>'"), |
|
100 | 100 | (r'sha1sum', "don't use sha1sum, use $TESTDIR/md5sum.py"), |
|
101 | 101 | (r'ls.*-\w*R', "don't use 'ls -R', use 'find'"), |
|
102 | 102 | (r'printf.*[^\\]\\([1-9]|0\d)', "don't use 'printf \NNN', use Python"), |
|
103 | 103 | (r'printf.*[^\\]\\x', "don't use printf \\x, use Python"), |
|
104 | 104 | (r'\$\(.*\)', "don't use $(expr), use `expr`"), |
|
105 | 105 | (r'rm -rf \*', "don't use naked rm -rf, target a directory"), |
|
106 | 106 | (r'(^|\|\s*)grep (-\w\s+)*[^|]*[(|]\w', |
|
107 | 107 | "use egrep for extended grep syntax"), |
|
108 | 108 | (r'/bin/', "don't use explicit paths for tools"), |
|
109 | 109 | (r'[^\n]\Z', "no trailing newline"), |
|
110 | 110 | (r'export .*=', "don't export and assign at once"), |
|
111 | 111 | (r'^source\b', "don't use 'source', use '.'"), |
|
112 | 112 | (r'touch -d', "don't use 'touch -d', use 'touch -t' instead"), |
|
113 | 113 | (r'ls +[^|\n-]+ +-', "options to 'ls' must come before filenames"), |
|
114 | 114 | (r'[^>\n]>\s*\$HGRCPATH', "don't overwrite $HGRCPATH, append to it"), |
|
115 | 115 | (r'^stop\(\)', "don't use 'stop' as a shell function name"), |
|
116 | 116 | (r'(\[|\btest\b).*-e ', "don't use 'test -e', use 'test -f'"), |
|
117 | 117 | (r'\[\[\s+[^\]]*\]\]', "don't use '[[ ]]', use '[ ]'"), |
|
118 | 118 | (r'^alias\b.*=', "don't use alias, use a function"), |
|
119 | 119 | (r'if\s*!', "don't use '!' to negate exit status"), |
|
120 | 120 | (r'/dev/u?random', "don't use entropy, use /dev/zero"), |
|
121 | 121 | (r'do\s*true;\s*done', "don't use true as loop body, use sleep 0"), |
|
122 | 122 | (r'^( *)\t', "don't use tabs to indent"), |
|
123 | 123 | (r'sed (-e )?\'(\d+|/[^/]*/)i(?!\\\n)', |
|
124 | 124 | "put a backslash-escaped newline after sed 'i' command"), |
|
125 | 125 | (r'^diff *-\w*[uU].*$\n(^ \$ |^$)', "prefix diff -u/-U with cmp"), |
|
126 | 126 | (r'^\s+(if)? diff *-\w*[uU]', "prefix diff -u/-U with cmp"), |
|
127 | 127 | (r'seq ', "don't use 'seq', use $TESTDIR/seq.py"), |
|
128 | 128 | (r'\butil\.Abort\b', "directly use error.Abort"), |
|
129 | 129 | (r'\|&', "don't use |&, use 2>&1"), |
|
130 | 130 | (r'\w = +\w', "only one space after = allowed"), |
|
131 | 131 | ], |
|
132 | 132 | # warnings |
|
133 | 133 | [ |
|
134 | 134 | (r'^function', "don't use 'function', use old style"), |
|
135 | 135 | (r'^diff.*-\w*N', "don't use 'diff -N'"), |
|
136 | 136 | (r'\$PWD|\${PWD}', "don't use $PWD, use `pwd`"), |
|
137 | 137 | (r'^([^"\'\n]|("[^"\n]*")|(\'[^\'\n]*\'))*\^', "^ must be quoted"), |
|
138 | 138 | (r'kill (`|\$\()', "don't use kill, use killdaemons.py") |
|
139 | 139 | ] |
|
140 | 140 | ] |
|
141 | 141 | |
|
142 | 142 | testfilters = [ |
|
143 | 143 | (r"( *)(#([^\n]*\S)?)", repcomment), |
|
144 | 144 | (r"<<(\S+)((.|\n)*?\n\1)", rephere), |
|
145 | 145 | ] |
|
146 | 146 | |
|
147 | 147 | winglobmsg = "use (glob) to match Windows paths too" |
|
148 | 148 | uprefix = r"^ \$ " |
|
149 | 149 | utestpats = [ |
|
150 | 150 | [ |
|
151 | 151 | (r'^(\S.*|| [$>] \S.*)[ \t]\n', "trailing whitespace on non-output"), |
|
152 | 152 | (uprefix + r'.*\|\s*sed[^|>\n]*\n', |
|
153 | 153 | "use regex test output patterns instead of sed"), |
|
154 | 154 | (uprefix + r'(true|exit 0)', "explicit zero exit unnecessary"), |
|
155 | 155 | (uprefix + r'.*(?<!\[)\$\?', "explicit exit code checks unnecessary"), |
|
156 | 156 | (uprefix + r'.*\|\| echo.*(fail|error)', |
|
157 | 157 | "explicit exit code checks unnecessary"), |
|
158 | 158 | (uprefix + r'set -e', "don't use set -e"), |
|
159 | 159 | (uprefix + r'(\s|fi\b|done\b)', "use > for continued lines"), |
|
160 | 160 | (uprefix + r'.*:\.\S*/', "x:.y in a path does not work on msys, rewrite " |
|
161 | 161 | "as x://.y, or see `hg log -k msys` for alternatives", r'-\S+:\.|' #-Rxxx |
|
162 | 162 | '# no-msys'), # in test-pull.t which is skipped on windows |
|
163 | 163 | (r'^ saved backup bundle to \$TESTTMP.*\.hg$', winglobmsg), |
|
164 | 164 | (r'^ changeset .* references (corrupted|missing) \$TESTTMP/.*[^)]$', |
|
165 | 165 | winglobmsg), |
|
166 | 166 | (r'^ pulling from \$TESTTMP/.*[^)]$', winglobmsg, |
|
167 | 167 | '\$TESTTMP/unix-repo$'), # in test-issue1802.t which skipped on windows |
|
168 | 168 | (r'^ reverting (?!subrepo ).*/.*[^)]$', winglobmsg), |
|
169 | 169 | (r'^ cloning subrepo \S+/.*[^)]$', winglobmsg), |
|
170 | 170 | (r'^ pushing to \$TESTTMP/.*[^)]$', winglobmsg), |
|
171 | 171 | (r'^ pushing subrepo \S+/\S+ to.*[^)]$', winglobmsg), |
|
172 | 172 | (r'^ moving \S+/.*[^)]$', winglobmsg), |
|
173 | 173 | (r'^ no changes made to subrepo since.*/.*[^)]$', winglobmsg), |
|
174 | 174 | (r'^ .*: largefile \S+ not available from file:.*/.*[^)]$', winglobmsg), |
|
175 | 175 | (r'^ .*file://\$TESTTMP', |
|
176 | 176 | 'write "file:/*/$TESTTMP" + (glob) to match on windows too'), |
|
177 | 177 | (r'^ (cat|find): .*: No such file or directory', |
|
178 | 178 | 'use test -f to test for file existence'), |
|
179 | 179 | (r'^ diff -[^ -]*p', |
|
180 | 180 | "don't use (external) diff with -p for portability"), |
|
181 | 181 | (r'^ [-+][-+][-+] .* [-+]0000 \(glob\)', |
|
182 | 182 | "glob timezone field in diff output for portability"), |
|
183 | 183 | (r'^ @@ -[0-9]+ [+][0-9]+,[0-9]+ @@', |
|
184 | 184 | "use '@@ -N* +N,n @@ (glob)' style chunk header for portability"), |
|
185 | 185 | (r'^ @@ -[0-9]+,[0-9]+ [+][0-9]+ @@', |
|
186 | 186 | "use '@@ -N,n +N* @@ (glob)' style chunk header for portability"), |
|
187 | 187 | (r'^ @@ -[0-9]+ [+][0-9]+ @@', |
|
188 | 188 | "use '@@ -N* +N* @@ (glob)' style chunk header for portability"), |
|
189 | 189 | (uprefix + r'hg( +-[^ ]+( +[^ ]+)?)* +extdiff' |
|
190 | 190 | r'( +(-[^ po-]+|--(?!program|option)[^ ]+|[^-][^ ]*))*$', |
|
191 | 191 | "use $RUNTESTDIR/pdiff via extdiff (or -o/-p for false-positives)"), |
|
192 | 192 | ], |
|
193 | 193 | # warnings |
|
194 | 194 | [ |
|
195 | 195 | (r'^ [^*?/\n]* \(glob\)$', |
|
196 | 196 | "glob match with no glob character (?*/)"), |
|
197 | 197 | ] |
|
198 | 198 | ] |
|
199 | 199 | |
|
200 | 200 | for i in [0, 1]: |
|
201 | 201 | for tp in testpats[i]: |
|
202 | 202 | p = tp[0] |
|
203 | 203 | m = tp[1] |
|
204 | 204 | if p.startswith(r'^'): |
|
205 | 205 | p = r"^ [$>] (%s)" % p[1:] |
|
206 | 206 | else: |
|
207 | 207 | p = r"^ [$>] .*(%s)" % p |
|
208 | 208 | utestpats[i].append((p, m) + tp[2:]) |
|
209 | 209 | |
|
210 | 210 | utestfilters = [ |
|
211 | 211 | (r"<<(\S+)((.|\n)*?\n > \1)", rephere), |
|
212 | 212 | (r"( *)(#([^\n]*\S)?)", repcomment), |
|
213 | 213 | ] |
|
214 | 214 | |
|
215 | 215 | pypats = [ |
|
216 | 216 | [ |
|
217 | 217 | (r'^\s*def\s*\w+\s*\(.*,\s*\(', |
|
218 | 218 | "tuple parameter unpacking not available in Python 3+"), |
|
219 | 219 | (r'lambda\s*\(.*,.*\)', |
|
220 | 220 | "tuple parameter unpacking not available in Python 3+"), |
|
221 | (r'import (.+,[^.]+\.[^.]+|[^.]+\.[^.]+,)', | |
|
222 | '2to3 can\'t always rewrite "import qux, foo.bar", ' | |
|
223 | 'use "import foo.bar" on its own line instead.'), | |
|
224 | 221 | (r'(?<!def)\s+(cmp)\(', "cmp is not available in Python 3+"), |
|
225 | 222 | (r'\breduce\s*\(.*', "reduce is not available in Python 3+"), |
|
226 | 223 | (r'dict\(.*=', 'dict() is different in Py2 and 3 and is slower than {}', |
|
227 | 224 | 'dict-from-generator'), |
|
228 | 225 | (r'\.has_key\b', "dict.has_key is not available in Python 3+"), |
|
229 | 226 | (r'\s<>\s', '<> operator is not available in Python 3+, use !='), |
|
230 | 227 | (r'^\s*\t', "don't use tabs"), |
|
231 | 228 | (r'\S;\s*\n', "semicolon"), |
|
232 | 229 | (r'[^_]_\([ \t\n]*(?:"[^"]+"[ \t\n+]*)+%', "don't use % inside _()"), |
|
233 | 230 | (r"[^_]_\([ \t\n]*(?:'[^']+'[ \t\n+]*)+%", "don't use % inside _()"), |
|
234 | 231 | (r'(\w|\)),\w', "missing whitespace after ,"), |
|
235 | 232 | (r'(\w|\))[+/*\-<>]\w', "missing whitespace in expression"), |
|
236 | 233 | (r'^\s+(\w|\.)+=\w[^,()\n]*$', "missing whitespace in assignment"), |
|
237 | 234 | (r'\w\s=\s\s+\w', "gratuitous whitespace after ="), |
|
238 | 235 | (r'.{81}', "line too long"), |
|
239 | 236 | (r' x+[xo][\'"]\n\s+[\'"]x', 'string join across lines with no space'), |
|
240 | 237 | (r'[^\n]\Z', "no trailing newline"), |
|
241 | 238 | (r'(\S[ \t]+|^[ \t]+)\n', "trailing whitespace"), |
|
242 | 239 | # (r'^\s+[^_ \n][^_. \n]+_[^_\n]+\s*=', |
|
243 | 240 | # "don't use underbars in identifiers"), |
|
244 | 241 | (r'^\s+(self\.)?[A-za-z][a-z0-9]+[A-Z]\w* = ', |
|
245 | 242 | "don't use camelcase in identifiers"), |
|
246 | 243 | (r'^\s*(if|while|def|class|except|try)\s[^[\n]*:\s*[^\\n]#\s]+', |
|
247 | 244 | "linebreak after :"), |
|
248 | 245 | (r'class\s[^( \n]+:', "old-style class, use class foo(object)", |
|
249 | 246 | r'#.*old-style'), |
|
250 | 247 | (r'class\s[^( \n]+\(\):', |
|
251 | 248 | "class foo() creates old style object, use class foo(object)", |
|
252 | 249 | r'#.*old-style'), |
|
253 | 250 | (r'\b(%s)\(' % '|'.join(k for k in keyword.kwlist |
|
254 | 251 | if k not in ('print', 'exec')), |
|
255 | 252 | "Python keyword is not a function"), |
|
256 | 253 | (r',]', "unneeded trailing ',' in list"), |
|
257 | 254 | # (r'class\s[A-Z][^\(]*\((?!Exception)', |
|
258 | 255 | # "don't capitalize non-exception classes"), |
|
259 | 256 | # (r'in range\(', "use xrange"), |
|
260 | 257 | # (r'^\s*print\s+', "avoid using print in core and extensions"), |
|
261 | 258 | (r'[\x80-\xff]', "non-ASCII character literal"), |
|
262 | 259 | (r'("\')\.format\(', "str.format() has no bytes counterpart, use %"), |
|
263 | 260 | (r'^\s*(%s)\s\s' % '|'.join(keyword.kwlist), |
|
264 | 261 | "gratuitous whitespace after Python keyword"), |
|
265 | 262 | (r'([\(\[][ \t]\S)|(\S[ \t][\)\]])', "gratuitous whitespace in () or []"), |
|
266 | 263 | # (r'\s\s=', "gratuitous whitespace before ="), |
|
267 | 264 | (r'[^>< ](\+=|-=|!=|<>|<=|>=|<<=|>>=|%=)\S', |
|
268 | 265 | "missing whitespace around operator"), |
|
269 | 266 | (r'[^>< ](\+=|-=|!=|<>|<=|>=|<<=|>>=|%=)\s', |
|
270 | 267 | "missing whitespace around operator"), |
|
271 | 268 | (r'\s(\+=|-=|!=|<>|<=|>=|<<=|>>=|%=)\S', |
|
272 | 269 | "missing whitespace around operator"), |
|
273 | 270 | (r'[^^+=*/!<>&| %-](\s=|=\s)[^= ]', |
|
274 | 271 | "wrong whitespace around ="), |
|
275 | 272 | (r'\([^()]*( =[^=]|[^<>!=]= )', |
|
276 | 273 | "no whitespace around = for named parameters"), |
|
277 | 274 | (r'raise Exception', "don't raise generic exceptions"), |
|
278 | 275 | (r'raise [^,(]+, (\([^\)]+\)|[^,\(\)]+)$', |
|
279 | 276 | "don't use old-style two-argument raise, use Exception(message)"), |
|
280 | 277 | (r' is\s+(not\s+)?["\'0-9-]', "object comparison with literal"), |
|
281 | 278 | (r' [=!]=\s+(True|False|None)', |
|
282 | 279 | "comparison with singleton, use 'is' or 'is not' instead"), |
|
283 | 280 | (r'^\s*(while|if) [01]:', |
|
284 | 281 | "use True/False for constant Boolean expression"), |
|
285 | 282 | (r'(?:(?<!def)\s+|\()hasattr', |
|
286 | 283 | 'hasattr(foo, bar) is broken, use util.safehasattr(foo, bar) instead'), |
|
287 | 284 | (r'opener\([^)]*\).read\(', |
|
288 | 285 | "use opener.read() instead"), |
|
289 | 286 | (r'opener\([^)]*\).write\(', |
|
290 | 287 | "use opener.write() instead"), |
|
291 | 288 | (r'[\s\(](open|file)\([^)]*\)\.read\(', |
|
292 | 289 | "use util.readfile() instead"), |
|
293 | 290 | (r'[\s\(](open|file)\([^)]*\)\.write\(', |
|
294 | 291 | "use util.writefile() instead"), |
|
295 | 292 | (r'^[\s\(]*(open(er)?|file)\([^)]*\)', |
|
296 | 293 | "always assign an opened file to a variable, and close it afterwards"), |
|
297 | 294 | (r'[\s\(](open|file)\([^)]*\)\.', |
|
298 | 295 | "always assign an opened file to a variable, and close it afterwards"), |
|
299 | 296 | (r'(?i)descend[e]nt', "the proper spelling is descendAnt"), |
|
300 | 297 | (r'\.debug\(\_', "don't mark debug messages for translation"), |
|
301 | 298 | (r'\.strip\(\)\.split\(\)', "no need to strip before splitting"), |
|
302 | 299 | (r'^\s*except\s*:', "naked except clause", r'#.*re-raises'), |
|
303 | 300 | (r'^\s*except\s([^\(,]+|\([^\)]+\))\s*,', |
|
304 | 301 | 'legacy exception syntax; use "as" instead of ","'), |
|
305 | 302 | (r':\n( )*( ){1,3}[^ ]', "must indent 4 spaces"), |
|
306 | 303 | (r'ui\.(status|progress|write|note|warn)\([\'\"]x', |
|
307 | 304 | "missing _() in ui message (use () to hide false-positives)"), |
|
308 | 305 | (r'release\(.*wlock, .*lock\)', "wrong lock release order"), |
|
309 | 306 | (r'\b__bool__\b', "__bool__ should be __nonzero__ in Python 2"), |
|
310 | 307 | (r'os\.path\.join\(.*, *(""|\'\')\)', |
|
311 | 308 | "use pathutil.normasprefix(path) instead of os.path.join(path, '')"), |
|
312 | 309 | (r'\s0[0-7]+\b', 'legacy octal syntax; use "0o" prefix instead of "0"'), |
|
313 | 310 | # XXX only catch mutable arguments on the first line of the definition |
|
314 | 311 | (r'def.*[( ]\w+=\{\}', "don't use mutable default arguments"), |
|
315 | 312 | (r'\butil\.Abort\b', "directly use error.Abort"), |
|
316 | 313 | ], |
|
317 | 314 | # warnings |
|
318 | 315 | [ |
|
319 | 316 | (r'(^| )pp +xxxxqq[ \n][^\n]', "add two newlines after '.. note::'"), |
|
320 | 317 | ] |
|
321 | 318 | ] |
|
322 | 319 | |
|
323 | 320 | pyfilters = [ |
|
324 | 321 | (r"""(?msx)(?P<comment>\#.*?$)| |
|
325 | 322 | ((?P<quote>('''|\"\"\"|(?<!')'(?!')|(?<!")"(?!"))) |
|
326 | 323 | (?P<text>(([^\\]|\\.)*?)) |
|
327 | 324 | (?P=quote))""", reppython), |
|
328 | 325 | ] |
|
329 | 326 | |
|
330 | 327 | txtfilters = [] |
|
331 | 328 | |
|
332 | 329 | txtpats = [ |
|
333 | 330 | [ |
|
334 | 331 | ('\s$', 'trailing whitespace'), |
|
335 | 332 | ('.. note::[ \n][^\n]', 'add two newlines after note::') |
|
336 | 333 | ], |
|
337 | 334 | [] |
|
338 | 335 | ] |
|
339 | 336 | |
|
340 | 337 | cpats = [ |
|
341 | 338 | [ |
|
342 | 339 | (r'//', "don't use //-style comments"), |
|
343 | 340 | (r'^ ', "don't use spaces to indent"), |
|
344 | 341 | (r'\S\t', "don't use tabs except for indent"), |
|
345 | 342 | (r'(\S[ \t]+|^[ \t]+)\n', "trailing whitespace"), |
|
346 | 343 | (r'.{81}', "line too long"), |
|
347 | 344 | (r'(while|if|do|for)\(', "use space after while/if/do/for"), |
|
348 | 345 | (r'return\(', "return is not a function"), |
|
349 | 346 | (r' ;', "no space before ;"), |
|
350 | 347 | (r'[^;] \)', "no space before )"), |
|
351 | 348 | (r'[)][{]', "space between ) and {"), |
|
352 | 349 | (r'\w+\* \w+', "use int *foo, not int* foo"), |
|
353 | 350 | (r'\W\([^\)]+\) \w+', "use (int)foo, not (int) foo"), |
|
354 | 351 | (r'\w+ (\+\+|--)', "use foo++, not foo ++"), |
|
355 | 352 | (r'\w,\w', "missing whitespace after ,"), |
|
356 | 353 | (r'^[^#]\w[+/*]\w', "missing whitespace in expression"), |
|
357 | 354 | (r'\w\s=\s\s+\w', "gratuitous whitespace after ="), |
|
358 | 355 | (r'^#\s+\w', "use #foo, not # foo"), |
|
359 | 356 | (r'[^\n]\Z', "no trailing newline"), |
|
360 | 357 | (r'^\s*#import\b', "use only #include in standard C code"), |
|
361 | 358 | ], |
|
362 | 359 | # warnings |
|
363 | 360 | [] |
|
364 | 361 | ] |
|
365 | 362 | |
|
366 | 363 | cfilters = [ |
|
367 | 364 | (r'(/\*)(((\*(?!/))|[^*])*)\*/', repccomment), |
|
368 | 365 | (r'''(?P<quote>(?<!")")(?P<text>([^"]|\\")+)"(?!")''', repquote), |
|
369 | 366 | (r'''(#\s*include\s+<)([^>]+)>''', repinclude), |
|
370 | 367 | (r'(\()([^)]+\))', repcallspaces), |
|
371 | 368 | ] |
|
372 | 369 | |
|
373 | 370 | inutilpats = [ |
|
374 | 371 | [ |
|
375 | 372 | (r'\bui\.', "don't use ui in util"), |
|
376 | 373 | ], |
|
377 | 374 | # warnings |
|
378 | 375 | [] |
|
379 | 376 | ] |
|
380 | 377 | |
|
381 | 378 | inrevlogpats = [ |
|
382 | 379 | [ |
|
383 | 380 | (r'\brepo\.', "don't use repo in revlog"), |
|
384 | 381 | ], |
|
385 | 382 | # warnings |
|
386 | 383 | [] |
|
387 | 384 | ] |
|
388 | 385 | |
|
389 | 386 | webtemplatefilters = [] |
|
390 | 387 | |
|
391 | 388 | webtemplatepats = [ |
|
392 | 389 | [], |
|
393 | 390 | [ |
|
394 | 391 | (r'{desc(\|(?!websub|firstline)[^\|]*)+}', |
|
395 | 392 | 'follow desc keyword with either firstline or websub'), |
|
396 | 393 | ] |
|
397 | 394 | ] |
|
398 | 395 | |
|
399 | 396 | checks = [ |
|
400 | 397 | ('python', r'.*\.(py|cgi)$', r'^#!.*python', pyfilters, pypats), |
|
401 | 398 | ('test script', r'(.*/)?test-[^.~]*$', '', testfilters, testpats), |
|
402 | 399 | ('c', r'.*\.[ch]$', '', cfilters, cpats), |
|
403 | 400 | ('unified test', r'.*\.t$', '', utestfilters, utestpats), |
|
404 | 401 | ('layering violation repo in revlog', r'mercurial/revlog\.py', '', |
|
405 | 402 | pyfilters, inrevlogpats), |
|
406 | 403 | ('layering violation ui in util', r'mercurial/util\.py', '', pyfilters, |
|
407 | 404 | inutilpats), |
|
408 | 405 | ('txt', r'.*\.txt$', '', txtfilters, txtpats), |
|
409 | 406 | ('web template', r'mercurial/templates/.*\.tmpl', '', |
|
410 | 407 | webtemplatefilters, webtemplatepats), |
|
411 | 408 | ] |
|
412 | 409 | |
|
413 | 410 | def _preparepats(): |
|
414 | 411 | for c in checks: |
|
415 | 412 | failandwarn = c[-1] |
|
416 | 413 | for pats in failandwarn: |
|
417 | 414 | for i, pseq in enumerate(pats): |
|
418 | 415 | # fix-up regexes for multi-line searches |
|
419 | 416 | p = pseq[0] |
|
420 | 417 | # \s doesn't match \n |
|
421 | 418 | p = re.sub(r'(?<!\\)\\s', r'[ \\t]', p) |
|
422 | 419 | # [^...] doesn't match newline |
|
423 | 420 | p = re.sub(r'(?<!\\)\[\^', r'[^\\n', p) |
|
424 | 421 | |
|
425 | 422 | pats[i] = (re.compile(p, re.MULTILINE),) + pseq[1:] |
|
426 | 423 | filters = c[3] |
|
427 | 424 | for i, flt in enumerate(filters): |
|
428 | 425 | filters[i] = re.compile(flt[0]), flt[1] |
|
429 | 426 | _preparepats() |
|
430 | 427 | |
|
431 | 428 | class norepeatlogger(object): |
|
432 | 429 | def __init__(self): |
|
433 | 430 | self._lastseen = None |
|
434 | 431 | |
|
435 | 432 | def log(self, fname, lineno, line, msg, blame): |
|
436 | 433 | """print error related a to given line of a given file. |
|
437 | 434 | |
|
438 | 435 | The faulty line will also be printed but only once in the case |
|
439 | 436 | of multiple errors. |
|
440 | 437 | |
|
441 | 438 | :fname: filename |
|
442 | 439 | :lineno: line number |
|
443 | 440 | :line: actual content of the line |
|
444 | 441 | :msg: error message |
|
445 | 442 | """ |
|
446 | 443 | msgid = fname, lineno, line |
|
447 | 444 | if msgid != self._lastseen: |
|
448 | 445 | if blame: |
|
449 | 446 | print "%s:%d (%s):" % (fname, lineno, blame) |
|
450 | 447 | else: |
|
451 | 448 | print "%s:%d:" % (fname, lineno) |
|
452 | 449 | print " > %s" % line |
|
453 | 450 | self._lastseen = msgid |
|
454 | 451 | print " " + msg |
|
455 | 452 | |
|
456 | 453 | _defaultlogger = norepeatlogger() |
|
457 | 454 | |
|
458 | 455 | def getblame(f): |
|
459 | 456 | lines = [] |
|
460 | 457 | for l in os.popen('hg annotate -un %s' % f): |
|
461 | 458 | start, line = l.split(':', 1) |
|
462 | 459 | user, rev = start.split() |
|
463 | 460 | lines.append((line[1:-1], user, rev)) |
|
464 | 461 | return lines |
|
465 | 462 | |
|
466 | 463 | def checkfile(f, logfunc=_defaultlogger.log, maxerr=None, warnings=False, |
|
467 | 464 | blame=False, debug=False, lineno=True): |
|
468 | 465 | """checks style and portability of a given file |
|
469 | 466 | |
|
470 | 467 | :f: filepath |
|
471 | 468 | :logfunc: function used to report error |
|
472 | 469 | logfunc(filename, linenumber, linecontent, errormessage) |
|
473 | 470 | :maxerr: number of error to display before aborting. |
|
474 | 471 | Set to false (default) to report all errors |
|
475 | 472 | |
|
476 | 473 | return True if no error is found, False otherwise. |
|
477 | 474 | """ |
|
478 | 475 | blamecache = None |
|
479 | 476 | result = True |
|
480 | 477 | |
|
481 | 478 | try: |
|
482 | 479 | fp = open(f) |
|
483 | 480 | except IOError as e: |
|
484 | 481 | print "Skipping %s, %s" % (f, str(e).split(':', 1)[0]) |
|
485 | 482 | return result |
|
486 | 483 | pre = post = fp.read() |
|
487 | 484 | fp.close() |
|
488 | 485 | |
|
489 | 486 | for name, match, magic, filters, pats in checks: |
|
490 | 487 | if debug: |
|
491 | 488 | print name, f |
|
492 | 489 | fc = 0 |
|
493 | 490 | if not (re.match(match, f) or (magic and re.search(magic, pre))): |
|
494 | 491 | if debug: |
|
495 | 492 | print "Skipping %s for %s it doesn't match %s" % ( |
|
496 | 493 | name, match, f) |
|
497 | 494 | continue |
|
498 | 495 | if "no-" "check-code" in pre: |
|
499 | 496 | # If you're looking at this line, it's because a file has: |
|
500 | 497 | # no- check- code |
|
501 | 498 | # but the reason to output skipping is to make life for |
|
502 | 499 | # tests easier. So, instead of writing it with a normal |
|
503 | 500 | # spelling, we write it with the expected spelling from |
|
504 | 501 | # tests/test-check-code.t |
|
505 | 502 | print "Skipping %s it has no-che?k-code (glob)" % f |
|
506 | 503 | return "Skip" # skip checking this file |
|
507 | 504 | for p, r in filters: |
|
508 | 505 | post = re.sub(p, r, post) |
|
509 | 506 | nerrs = len(pats[0]) # nerr elements are errors |
|
510 | 507 | if warnings: |
|
511 | 508 | pats = pats[0] + pats[1] |
|
512 | 509 | else: |
|
513 | 510 | pats = pats[0] |
|
514 | 511 | # print post # uncomment to show filtered version |
|
515 | 512 | |
|
516 | 513 | if debug: |
|
517 | 514 | print "Checking %s for %s" % (name, f) |
|
518 | 515 | |
|
519 | 516 | prelines = None |
|
520 | 517 | errors = [] |
|
521 | 518 | for i, pat in enumerate(pats): |
|
522 | 519 | if len(pat) == 3: |
|
523 | 520 | p, msg, ignore = pat |
|
524 | 521 | else: |
|
525 | 522 | p, msg = pat |
|
526 | 523 | ignore = None |
|
527 | 524 | if i >= nerrs: |
|
528 | 525 | msg = "warning: " + msg |
|
529 | 526 | |
|
530 | 527 | pos = 0 |
|
531 | 528 | n = 0 |
|
532 | 529 | for m in p.finditer(post): |
|
533 | 530 | if prelines is None: |
|
534 | 531 | prelines = pre.splitlines() |
|
535 | 532 | postlines = post.splitlines(True) |
|
536 | 533 | |
|
537 | 534 | start = m.start() |
|
538 | 535 | while n < len(postlines): |
|
539 | 536 | step = len(postlines[n]) |
|
540 | 537 | if pos + step > start: |
|
541 | 538 | break |
|
542 | 539 | pos += step |
|
543 | 540 | n += 1 |
|
544 | 541 | l = prelines[n] |
|
545 | 542 | |
|
546 | 543 | if ignore and re.search(ignore, l, re.MULTILINE): |
|
547 | 544 | if debug: |
|
548 | 545 | print "Skipping %s for %s:%s (ignore pattern)" % ( |
|
549 | 546 | name, f, n) |
|
550 | 547 | continue |
|
551 | 548 | bd = "" |
|
552 | 549 | if blame: |
|
553 | 550 | bd = 'working directory' |
|
554 | 551 | if not blamecache: |
|
555 | 552 | blamecache = getblame(f) |
|
556 | 553 | if n < len(blamecache): |
|
557 | 554 | bl, bu, br = blamecache[n] |
|
558 | 555 | if bl == l: |
|
559 | 556 | bd = '%s@%s' % (bu, br) |
|
560 | 557 | |
|
561 | 558 | errors.append((f, lineno and n + 1, l, msg, bd)) |
|
562 | 559 | result = False |
|
563 | 560 | |
|
564 | 561 | errors.sort() |
|
565 | 562 | for e in errors: |
|
566 | 563 | logfunc(*e) |
|
567 | 564 | fc += 1 |
|
568 | 565 | if maxerr and fc >= maxerr: |
|
569 | 566 | print " (too many errors, giving up)" |
|
570 | 567 | break |
|
571 | 568 | |
|
572 | 569 | return result |
|
573 | 570 | |
|
574 | 571 | if __name__ == "__main__": |
|
575 | 572 | parser = optparse.OptionParser("%prog [options] [files]") |
|
576 | 573 | parser.add_option("-w", "--warnings", action="store_true", |
|
577 | 574 | help="include warning-level checks") |
|
578 | 575 | parser.add_option("-p", "--per-file", type="int", |
|
579 | 576 | help="max warnings per file") |
|
580 | 577 | parser.add_option("-b", "--blame", action="store_true", |
|
581 | 578 | help="use annotate to generate blame info") |
|
582 | 579 | parser.add_option("", "--debug", action="store_true", |
|
583 | 580 | help="show debug information") |
|
584 | 581 | parser.add_option("", "--nolineno", action="store_false", |
|
585 | 582 | dest='lineno', help="don't show line numbers") |
|
586 | 583 | |
|
587 | 584 | parser.set_defaults(per_file=15, warnings=False, blame=False, debug=False, |
|
588 | 585 | lineno=True) |
|
589 | 586 | (options, args) = parser.parse_args() |
|
590 | 587 | |
|
591 | 588 | if len(args) == 0: |
|
592 | 589 | check = glob.glob("*") |
|
593 | 590 | else: |
|
594 | 591 | check = args |
|
595 | 592 | |
|
596 | 593 | ret = 0 |
|
597 | 594 | for f in check: |
|
598 | 595 | if not checkfile(f, maxerr=options.per_file, warnings=options.warnings, |
|
599 | 596 | blame=options.blame, debug=options.debug, |
|
600 | 597 | lineno=options.lineno): |
|
601 | 598 | ret = 1 |
|
602 | 599 | sys.exit(ret) |
General Comments 0
You need to be logged in to leave comments.
Login now