1import io 2import itertools 3import shlex 4import string 5import unittest 6 7 8 9# The original test data set was from shellwords, by Hartmut Goebel. 10 11data = r"""x|x| 12foo bar|foo|bar| 13 foo bar|foo|bar| 14 foo bar |foo|bar| 15foo bar bla fasel|foo|bar|bla|fasel| 16x y z xxxx|x|y|z|xxxx| 17\x bar|\|x|bar| 18\ x bar|\|x|bar| 19\ bar|\|bar| 20foo \x bar|foo|\|x|bar| 21foo \ x bar|foo|\|x|bar| 22foo \ bar|foo|\|bar| 23foo "bar" bla|foo|"bar"|bla| 24"foo" "bar" "bla"|"foo"|"bar"|"bla"| 25"foo" bar "bla"|"foo"|bar|"bla"| 26"foo" bar bla|"foo"|bar|bla| 27foo 'bar' bla|foo|'bar'|bla| 28'foo' 'bar' 'bla'|'foo'|'bar'|'bla'| 29'foo' bar 'bla'|'foo'|bar|'bla'| 30'foo' bar bla|'foo'|bar|bla| 31blurb foo"bar"bar"fasel" baz|blurb|foo"bar"bar"fasel"|baz| 32blurb foo'bar'bar'fasel' baz|blurb|foo'bar'bar'fasel'|baz| 33""|""| 34''|''| 35foo "" bar|foo|""|bar| 36foo '' bar|foo|''|bar| 37foo "" "" "" bar|foo|""|""|""|bar| 38foo '' '' '' bar|foo|''|''|''|bar| 39\""|\|""| 40"\"|"\"| 41"foo\ bar"|"foo\ bar"| 42"foo\\ bar"|"foo\\ bar"| 43"foo\\ bar\"|"foo\\ bar\"| 44"foo\\" bar\""|"foo\\"|bar|\|""| 45"foo\\ bar\" dfadf"|"foo\\ bar\"|dfadf"| 46"foo\\\ bar\" dfadf"|"foo\\\ bar\"|dfadf"| 47"foo\\\x bar\" dfadf"|"foo\\\x bar\"|dfadf"| 48"foo\x bar\" dfadf"|"foo\x bar\"|dfadf"| 49\''|\|''| 50'foo\ bar'|'foo\ bar'| 51'foo\\ bar'|'foo\\ bar'| 52"foo\\\x bar\" df'a\ 'df'|"foo\\\x bar\"|df'a|\|'df'| 53\"foo"|\|"foo"| 54\"foo"\x|\|"foo"|\|x| 55"foo\x"|"foo\x"| 56"foo\ "|"foo\ "| 57foo\ xx|foo|\|xx| 58foo\ x\x|foo|\|x|\|x| 59foo\ x\x\""|foo|\|x|\|x|\|""| 60"foo\ x\x"|"foo\ x\x"| 61"foo\ x\x\\"|"foo\ x\x\\"| 62"foo\ x\x\\""foobar"|"foo\ x\x\\"|"foobar"| 63"foo\ x\x\\"\''"foobar"|"foo\ x\x\\"|\|''|"foobar"| 64"foo\ x\x\\"\'"fo'obar"|"foo\ x\x\\"|\|'"fo'|obar"| 65"foo\ x\x\\"\'"fo'obar" 'don'\''t'|"foo\ x\x\\"|\|'"fo'|obar"|'don'|\|''|t'| 66'foo\ bar'|'foo\ bar'| 67'foo\\ bar'|'foo\\ bar'| 68foo\ bar|foo|\|bar| 69foo#bar\nbaz|foobaz| 70:-) ;-)|:|-|)|;|-|)| 71áéíóú|á|é|í|ó|ú| 72""" 73 74posix_data = r"""x|x| 75foo bar|foo|bar| 76 foo bar|foo|bar| 77 foo bar |foo|bar| 78foo bar bla fasel|foo|bar|bla|fasel| 79x y z xxxx|x|y|z|xxxx| 80\x bar|x|bar| 81\ x bar| x|bar| 82\ bar| bar| 83foo \x bar|foo|x|bar| 84foo \ x bar|foo| x|bar| 85foo \ bar|foo| bar| 86foo "bar" bla|foo|bar|bla| 87"foo" "bar" "bla"|foo|bar|bla| 88"foo" bar "bla"|foo|bar|bla| 89"foo" bar bla|foo|bar|bla| 90foo 'bar' bla|foo|bar|bla| 91'foo' 'bar' 'bla'|foo|bar|bla| 92'foo' bar 'bla'|foo|bar|bla| 93'foo' bar bla|foo|bar|bla| 94blurb foo"bar"bar"fasel" baz|blurb|foobarbarfasel|baz| 95blurb foo'bar'bar'fasel' baz|blurb|foobarbarfasel|baz| 96""|| 97''|| 98foo "" bar|foo||bar| 99foo '' bar|foo||bar| 100foo "" "" "" bar|foo||||bar| 101foo '' '' '' bar|foo||||bar| 102\"|"| 103"\""|"| 104"foo\ bar"|foo\ bar| 105"foo\\ bar"|foo\ bar| 106"foo\\ bar\""|foo\ bar"| 107"foo\\" bar\"|foo\|bar"| 108"foo\\ bar\" dfadf"|foo\ bar" dfadf| 109"foo\\\ bar\" dfadf"|foo\\ bar" dfadf| 110"foo\\\x bar\" dfadf"|foo\\x bar" dfadf| 111"foo\x bar\" dfadf"|foo\x bar" dfadf| 112\'|'| 113'foo\ bar'|foo\ bar| 114'foo\\ bar'|foo\\ bar| 115"foo\\\x bar\" df'a\ 'df"|foo\\x bar" df'a\ 'df| 116\"foo|"foo| 117\"foo\x|"foox| 118"foo\x"|foo\x| 119"foo\ "|foo\ | 120foo\ xx|foo xx| 121foo\ x\x|foo xx| 122foo\ x\x\"|foo xx"| 123"foo\ x\x"|foo\ x\x| 124"foo\ x\x\\"|foo\ x\x\| 125"foo\ x\x\\""foobar"|foo\ x\x\foobar| 126"foo\ x\x\\"\'"foobar"|foo\ x\x\'foobar| 127"foo\ x\x\\"\'"fo'obar"|foo\ x\x\'fo'obar| 128"foo\ x\x\\"\'"fo'obar" 'don'\''t'|foo\ x\x\'fo'obar|don't| 129"foo\ x\x\\"\'"fo'obar" 'don'\''t' \\|foo\ x\x\'fo'obar|don't|\| 130'foo\ bar'|foo\ bar| 131'foo\\ bar'|foo\\ bar| 132foo\ bar|foo bar| 133foo#bar\nbaz|foo|baz| 134:-) ;-)|:-)|;-)| 135áéíóú|áéíóú| 136""" 137 138class ShlexTest(unittest.TestCase): 139 def setUp(self): 140 self.data = [x.split("|")[:-1] 141 for x in data.splitlines()] 142 self.posix_data = [x.split("|")[:-1] 143 for x in posix_data.splitlines()] 144 for item in self.data: 145 item[0] = item[0].replace(r"\n", "\n") 146 for item in self.posix_data: 147 item[0] = item[0].replace(r"\n", "\n") 148 149 def splitTest(self, data, comments): 150 for i in range(len(data)): 151 l = shlex.split(data[i][0], comments=comments) 152 self.assertEqual(l, data[i][1:], 153 "%s: %s != %s" % 154 (data[i][0], l, data[i][1:])) 155 156 def oldSplit(self, s): 157 ret = [] 158 lex = shlex.shlex(io.StringIO(s)) 159 tok = lex.get_token() 160 while tok: 161 ret.append(tok) 162 tok = lex.get_token() 163 return ret 164 165 def testSplitPosix(self): 166 """Test data splitting with posix parser""" 167 self.splitTest(self.posix_data, comments=True) 168 169 def testCompat(self): 170 """Test compatibility interface""" 171 for i in range(len(self.data)): 172 l = self.oldSplit(self.data[i][0]) 173 self.assertEqual(l, self.data[i][1:], 174 "%s: %s != %s" % 175 (self.data[i][0], l, self.data[i][1:])) 176 177 def testSyntaxSplitAmpersandAndPipe(self): 178 """Test handling of syntax splitting of &, |""" 179 # Could take these forms: &&, &, |&, ;&, ;;& 180 # of course, the same applies to | and || 181 # these should all parse to the same output 182 for delimiter in ('&&', '&', '|&', ';&', ';;&', 183 '||', '|', '&|', ';|', ';;|'): 184 src = ['echo hi %s echo bye' % delimiter, 185 'echo hi%secho bye' % delimiter] 186 ref = ['echo', 'hi', delimiter, 'echo', 'bye'] 187 for ss, ws in itertools.product(src, (False, True)): 188 s = shlex.shlex(ss, punctuation_chars=True) 189 s.whitespace_split = ws 190 result = list(s) 191 self.assertEqual(ref, result, 192 "While splitting '%s' [ws=%s]" % (ss, ws)) 193 194 def testSyntaxSplitSemicolon(self): 195 """Test handling of syntax splitting of ;""" 196 # Could take these forms: ;, ;;, ;&, ;;& 197 # these should all parse to the same output 198 for delimiter in (';', ';;', ';&', ';;&'): 199 src = ['echo hi %s echo bye' % delimiter, 200 'echo hi%s echo bye' % delimiter, 201 'echo hi%secho bye' % delimiter] 202 ref = ['echo', 'hi', delimiter, 'echo', 'bye'] 203 for ss, ws in itertools.product(src, (False, True)): 204 s = shlex.shlex(ss, punctuation_chars=True) 205 s.whitespace_split = ws 206 result = list(s) 207 self.assertEqual(ref, result, 208 "While splitting '%s' [ws=%s]" % (ss, ws)) 209 210 def testSyntaxSplitRedirect(self): 211 """Test handling of syntax splitting of >""" 212 # of course, the same applies to <, | 213 # these should all parse to the same output 214 for delimiter in ('<', '|'): 215 src = ['echo hi %s out' % delimiter, 216 'echo hi%s out' % delimiter, 217 'echo hi%sout' % delimiter] 218 ref = ['echo', 'hi', delimiter, 'out'] 219 for ss, ws in itertools.product(src, (False, True)): 220 s = shlex.shlex(ss, punctuation_chars=True) 221 result = list(s) 222 self.assertEqual(ref, result, 223 "While splitting '%s' [ws=%s]" % (ss, ws)) 224 225 def testSyntaxSplitParen(self): 226 """Test handling of syntax splitting of ()""" 227 # these should all parse to the same output 228 src = ['( echo hi )', 229 '(echo hi)'] 230 ref = ['(', 'echo', 'hi', ')'] 231 for ss, ws in itertools.product(src, (False, True)): 232 s = shlex.shlex(ss, punctuation_chars=True) 233 s.whitespace_split = ws 234 result = list(s) 235 self.assertEqual(ref, result, 236 "While splitting '%s' [ws=%s]" % (ss, ws)) 237 238 def testSyntaxSplitCustom(self): 239 """Test handling of syntax splitting with custom chars""" 240 ss = "~/a&&b-c --color=auto||d *.py?" 241 ref = ['~/a', '&', '&', 'b-c', '--color=auto', '||', 'd', '*.py?'] 242 s = shlex.shlex(ss, punctuation_chars="|") 243 result = list(s) 244 self.assertEqual(ref, result, "While splitting '%s' [ws=False]" % ss) 245 ref = ['~/a&&b-c', '--color=auto', '||', 'd', '*.py?'] 246 s = shlex.shlex(ss, punctuation_chars="|") 247 s.whitespace_split = True 248 result = list(s) 249 self.assertEqual(ref, result, "While splitting '%s' [ws=True]" % ss) 250 251 def testTokenTypes(self): 252 """Test that tokens are split with types as expected.""" 253 for source, expected in ( 254 ('a && b || c', 255 [('a', 'a'), ('&&', 'c'), ('b', 'a'), 256 ('||', 'c'), ('c', 'a')]), 257 ): 258 s = shlex.shlex(source, punctuation_chars=True) 259 observed = [] 260 while True: 261 t = s.get_token() 262 if t == s.eof: 263 break 264 if t[0] in s.punctuation_chars: 265 tt = 'c' 266 else: 267 tt = 'a' 268 observed.append((t, tt)) 269 self.assertEqual(observed, expected) 270 271 def testPunctuationInWordChars(self): 272 """Test that any punctuation chars are removed from wordchars""" 273 s = shlex.shlex('a_b__c', punctuation_chars='_') 274 self.assertNotIn('_', s.wordchars) 275 self.assertEqual(list(s), ['a', '_', 'b', '__', 'c']) 276 277 def testPunctuationWithWhitespaceSplit(self): 278 """Test that with whitespace_split, behaviour is as expected""" 279 s = shlex.shlex('a && b || c', punctuation_chars='&') 280 # whitespace_split is False, so splitting will be based on 281 # punctuation_chars 282 self.assertEqual(list(s), ['a', '&&', 'b', '|', '|', 'c']) 283 s = shlex.shlex('a && b || c', punctuation_chars='&') 284 s.whitespace_split = True 285 # whitespace_split is True, so splitting will be based on 286 # white space 287 self.assertEqual(list(s), ['a', '&&', 'b', '||', 'c']) 288 289 def testPunctuationWithPosix(self): 290 """Test that punctuation_chars and posix behave correctly together.""" 291 # see Issue #29132 292 s = shlex.shlex('f >"abc"', posix=True, punctuation_chars=True) 293 self.assertEqual(list(s), ['f', '>', 'abc']) 294 s = shlex.shlex('f >\\"abc\\"', posix=True, punctuation_chars=True) 295 self.assertEqual(list(s), ['f', '>', '"abc"']) 296 297 def testEmptyStringHandling(self): 298 """Test that parsing of empty strings is correctly handled.""" 299 # see Issue #21999 300 expected = ['', ')', 'abc'] 301 for punct in (False, True): 302 s = shlex.shlex("'')abc", posix=True, punctuation_chars=punct) 303 slist = list(s) 304 self.assertEqual(slist, expected) 305 expected = ["''", ')', 'abc'] 306 s = shlex.shlex("'')abc", punctuation_chars=True) 307 self.assertEqual(list(s), expected) 308 309 def testUnicodeHandling(self): 310 """Test punctuation_chars and whitespace_split handle unicode.""" 311 ss = "\u2119\u01b4\u2602\u210c\u00f8\u1f24" 312 # Should be parsed as one complete token (whitespace_split=True). 313 ref = ['\u2119\u01b4\u2602\u210c\u00f8\u1f24'] 314 s = shlex.shlex(ss, punctuation_chars=True) 315 s.whitespace_split = True 316 self.assertEqual(list(s), ref) 317 # Without whitespace_split, uses wordchars and splits on all. 318 ref = ['\u2119', '\u01b4', '\u2602', '\u210c', '\u00f8', '\u1f24'] 319 s = shlex.shlex(ss, punctuation_chars=True) 320 self.assertEqual(list(s), ref) 321 322 def testQuote(self): 323 safeunquoted = string.ascii_letters + string.digits + '@%_-+=:,./' 324 unicode_sample = '\xe9\xe0\xdf' # e + acute accent, a + grave, sharp s 325 unsafe = '"`$\\!' + unicode_sample 326 327 self.assertEqual(shlex.quote(''), "''") 328 self.assertEqual(shlex.quote(safeunquoted), safeunquoted) 329 self.assertEqual(shlex.quote('test file name'), "'test file name'") 330 for u in unsafe: 331 self.assertEqual(shlex.quote('test%sname' % u), 332 "'test%sname'" % u) 333 for u in unsafe: 334 self.assertEqual(shlex.quote("test%s'name'" % u), 335 "'test%s'\"'\"'name'\"'\"''" % u) 336 337 def testJoin(self): 338 for split_command, command in [ 339 (['a ', 'b'], "'a ' b"), 340 (['a', ' b'], "a ' b'"), 341 (['a', ' ', 'b'], "a ' ' b"), 342 (['"a', 'b"'], '\'"a\' \'b"\''), 343 ]: 344 with self.subTest(command=command): 345 joined = shlex.join(split_command) 346 self.assertEqual(joined, command) 347 348 def testJoinRoundtrip(self): 349 all_data = self.data + self.posix_data 350 for command, *split_command in all_data: 351 with self.subTest(command=command): 352 joined = shlex.join(split_command) 353 resplit = shlex.split(joined) 354 self.assertEqual(split_command, resplit) 355 356 def testPunctuationCharsReadOnly(self): 357 punctuation_chars = "/|$%^" 358 shlex_instance = shlex.shlex(punctuation_chars=punctuation_chars) 359 self.assertEqual(shlex_instance.punctuation_chars, punctuation_chars) 360 with self.assertRaises(AttributeError): 361 shlex_instance.punctuation_chars = False 362 363 364# Allow this test to be used with old shlex.py 365if not getattr(shlex, "split", None): 366 for methname in dir(ShlexTest): 367 if methname.startswith("test") and methname != "testCompat": 368 delattr(ShlexTest, methname) 369 370if __name__ == "__main__": 371 unittest.main() 372