| 1 |
import difflib |
| 2 |
from test.support import findfile |
| 3 |
import unittest |
| 4 |
import doctest |
| 5 |
import sys |
| 6 |
|
| 7 |
|
| 8 |
class TestWithAscii(unittest.TestCase): |
| 9 |
def test_one_insert(self): |
| 10 |
sm = difflib.SequenceMatcher(None, 'b' * 100, 'a' + 'b' * 100) |
| 11 |
self.assertAlmostEqual(sm.ratio(), 0.995, places=3) |
| 12 |
self.assertEqual(list(sm.get_opcodes()), |
| 13 |
[ ('insert', 0, 0, 0, 1), |
| 14 |
('equal', 0, 100, 1, 101)]) |
| 15 |
self.assertEqual(sm.bpopular, set()) |
| 16 |
sm = difflib.SequenceMatcher(None, 'b' * 100, 'b' * 50 + 'a' + 'b' * 50) |
| 17 |
self.assertAlmostEqual(sm.ratio(), 0.995, places=3) |
| 18 |
self.assertEqual(list(sm.get_opcodes()), |
| 19 |
[ ('equal', 0, 50, 0, 50), |
| 20 |
('insert', 50, 50, 50, 51), |
| 21 |
('equal', 50, 100, 51, 101)]) |
| 22 |
self.assertEqual(sm.bpopular, set()) |
| 23 |
|
| 24 |
def test_one_delete(self): |
| 25 |
sm = difflib.SequenceMatcher(None, 'a' * 40 + 'c' + 'b' * 40, 'a' * 40 + 'b' * 40) |
| 26 |
self.assertAlmostEqual(sm.ratio(), 0.994, places=3) |
| 27 |
self.assertEqual(list(sm.get_opcodes()), |
| 28 |
[ ('equal', 0, 40, 0, 40), |
| 29 |
('delete', 40, 41, 40, 40), |
| 30 |
('equal', 41, 81, 40, 80)]) |
| 31 |
|
| 32 |
def test_opcode_caching(self): |
| 33 |
sm = difflib.SequenceMatcher(None, 'b' * 100, 'a' + 'b' * 100) |
| 34 |
opcode = sm.get_opcodes() |
| 35 |
self.assertEqual(opcode, |
| 36 |
[ ('insert', 0, 0, 0, 1), |
| 37 |
('equal', 0, 100, 1, 101)]) |
| 38 |
# Implementation detail: opcodes are cached; |
| 39 |
# `get_opcodes()` returns the same object |
| 40 |
self.assertIs(opcode, sm.get_opcodes()) |
| 41 |
|
| 42 |
def test_bjunk(self): |
| 43 |
sm = difflib.SequenceMatcher(isjunk=lambda x: x == ' ', |
| 44 |
a='a' * 40 + 'b' * 40, b='a' * 44 + 'b' * 40) |
| 45 |
self.assertEqual(sm.bjunk, set()) |
| 46 |
|
| 47 |
sm = difflib.SequenceMatcher(isjunk=lambda x: x == ' ', |
| 48 |
a='a' * 40 + 'b' * 40, b='a' * 44 + 'b' * 40 + ' ' * 20) |
| 49 |
self.assertEqual(sm.bjunk, {' '}) |
| 50 |
|
| 51 |
sm = difflib.SequenceMatcher(isjunk=lambda x: x in [' ', 'b'], |
| 52 |
a='a' * 40 + 'b' * 40, b='a' * 44 + 'b' * 40 + ' ' * 20) |
| 53 |
self.assertEqual(sm.bjunk, {' ', 'b'}) |
| 54 |
|
| 55 |
|
| 56 |
class TestAutojunk(unittest.TestCase): |
| 57 |
"""Tests for the autojunk parameter added in 2.7""" |
| 58 |
def test_one_insert_homogenous_sequence(self): |
| 59 |
# By default autojunk=True and the heuristic kicks in for a sequence |
| 60 |
# of length 200+ |
| 61 |
seq1 = 'b' * 200 |
| 62 |
seq2 = 'a' + 'b' * 200 |
| 63 |
|
| 64 |
sm = difflib.SequenceMatcher(None, seq1, seq2) |
| 65 |
self.assertAlmostEqual(sm.ratio(), 0, places=3) |
| 66 |
self.assertEqual(sm.bpopular, {'b'}) |
| 67 |
|
| 68 |
# Now turn the heuristic off |
| 69 |
sm = difflib.SequenceMatcher(None, seq1, seq2, autojunk=False) |
| 70 |
self.assertAlmostEqual(sm.ratio(), 0.9975, places=3) |
| 71 |
self.assertEqual(sm.bpopular, set()) |
| 72 |
|
| 73 |
|
| 74 |
class TestSFbugs(unittest.TestCase): |
| 75 |
def test_ratio_for_null_seqn(self): |
| 76 |
# Check clearing of SF bug 763023 |
| 77 |
s = difflib.SequenceMatcher(None, [], []) |
| 78 |
self.assertEqual(s.ratio(), 1) |
| 79 |
self.assertEqual(s.quick_ratio(), 1) |
| 80 |
self.assertEqual(s.real_quick_ratio(), 1) |
| 81 |
|
| 82 |
def test_comparing_empty_lists(self): |
| 83 |
# Check fix for bug #979794 |
| 84 |
group_gen = difflib.SequenceMatcher(None, [], []).get_grouped_opcodes() |
| 85 |
self.assertRaises(StopIteration, next, group_gen) |
| 86 |
diff_gen = difflib.unified_diff([], []) |
| 87 |
self.assertRaises(StopIteration, next, diff_gen) |
| 88 |
|
| 89 |
def test_matching_blocks_cache(self): |
| 90 |
# Issue #21635 |
| 91 |
s = difflib.SequenceMatcher(None, "abxcd", "abcd") |
| 92 |
first = s.get_matching_blocks() |
| 93 |
second = s.get_matching_blocks() |
| 94 |
self.assertEqual(second[0].size, 2) |
| 95 |
self.assertEqual(second[1].size, 2) |
| 96 |
self.assertEqual(second[2].size, 0) |
| 97 |
|
| 98 |
def test_added_tab_hint(self): |
| 99 |
# Check fix for bug #1488943 |
| 100 |
diff = list(difflib.Differ().compare(["\tI am a buggy"],["\t\tI am a bug"])) |
| 101 |
self.assertEqual("- \tI am a buggy", diff[0]) |
| 102 |
self.assertEqual("? \t --\n", diff[1]) |
| 103 |
self.assertEqual("+ \t\tI am a bug", diff[2]) |
| 104 |
self.assertEqual("? +\n", diff[3]) |
| 105 |
|
| 106 |
def test_hint_indented_properly_with_tabs(self): |
| 107 |
diff = list(difflib.Differ().compare(["\t \t \t^"], ["\t \t \t^\n"])) |
| 108 |
self.assertEqual("- \t \t \t^", diff[0]) |
| 109 |
self.assertEqual("+ \t \t \t^\n", diff[1]) |
| 110 |
self.assertEqual("? \t \t \t +\n", diff[2]) |
| 111 |
|
| 112 |
def test_mdiff_catch_stop_iteration(self): |
| 113 |
# Issue #33224 |
| 114 |
self.assertEqual( |
| 115 |
list(difflib._mdiff(["2"], ["3"], 1)), |
| 116 |
[((1, '\x00-2\x01'), (1, '\x00+3\x01'), True)], |
| 117 |
) |
| 118 |
|
| 119 |
|
| 120 |
patch914575_from1 = """ |
| 121 |
1. Beautiful is beTTer than ugly. |
| 122 |
2. Explicit is better than implicit. |
| 123 |
3. Simple is better than complex. |
| 124 |
4. Complex is better than complicated. |
| 125 |
""" |
| 126 |
|
| 127 |
patch914575_to1 = """ |
| 128 |
1. Beautiful is better than ugly. |
| 129 |
3. Simple is better than complex. |
| 130 |
4. Complicated is better than complex. |
| 131 |
5. Flat is better than nested. |
| 132 |
""" |
| 133 |
|
| 134 |
patch914575_nonascii_from1 = """ |
| 135 |
1. Beautiful is beTTer than ugly. |
| 136 |
2. Explicit is better than ımplıcıt. |
| 137 |
3. Simple is better than complex. |
| 138 |
4. Complex is better than complicated. |
| 139 |
""" |
| 140 |
|
| 141 |
patch914575_nonascii_to1 = """ |
| 142 |
1. Beautiful is better than ügly. |
| 143 |
3. Sımple is better than complex. |
| 144 |
4. Complicated is better than cömplex. |
| 145 |
5. Flat is better than nested. |
| 146 |
""" |
| 147 |
|
| 148 |
patch914575_from2 = """ |
| 149 |
\t\tLine 1: preceded by from:[tt] to:[ssss] |
| 150 |
\t\tLine 2: preceded by from:[sstt] to:[sssst] |
| 151 |
\t \tLine 3: preceded by from:[sstst] to:[ssssss] |
| 152 |
Line 4: \thas from:[sst] to:[sss] after : |
| 153 |
Line 5: has from:[t] to:[ss] at end\t |
| 154 |
""" |
| 155 |
|
| 156 |
patch914575_to2 = """ |
| 157 |
Line 1: preceded by from:[tt] to:[ssss] |
| 158 |
\tLine 2: preceded by from:[sstt] to:[sssst] |
| 159 |
Line 3: preceded by from:[sstst] to:[ssssss] |
| 160 |
Line 4: has from:[sst] to:[sss] after : |
| 161 |
Line 5: has from:[t] to:[ss] at end |
| 162 |
""" |
| 163 |
|
| 164 |
patch914575_from3 = """line 0 |
| 165 |
1234567890123456789012345689012345 |
| 166 |
line 1 |
| 167 |
line 2 |
| 168 |
line 3 |
| 169 |
line 4 changed |
| 170 |
line 5 changed |
| 171 |
line 6 changed |
| 172 |
line 7 |
| 173 |
line 8 subtracted |
| 174 |
line 9 |
| 175 |
1234567890123456789012345689012345 |
| 176 |
short line |
| 177 |
just fits in!! |
| 178 |
just fits in two lines yup!! |
| 179 |
the end""" |
| 180 |
|
| 181 |
patch914575_to3 = """line 0 |
| 182 |
1234567890123456789012345689012345 |
| 183 |
line 1 |
| 184 |
line 2 added |
| 185 |
line 3 |
| 186 |
line 4 chanGEd |
| 187 |
line 5a chanGed |
| 188 |
line 6a changEd |
| 189 |
line 7 |
| 190 |
line 8 |
| 191 |
line 9 |
| 192 |
1234567890 |
| 193 |
another long line that needs to be wrapped |
| 194 |
just fitS in!! |
| 195 |
just fits in two lineS yup!! |
| 196 |
the end""" |
| 197 |
|
| 198 |
class TestSFpatches(unittest.TestCase): |
| 199 |
|
| 200 |
def test_html_diff(self): |
| 201 |
# Check SF patch 914575 for generating HTML differences |
| 202 |
f1a = ((patch914575_from1 + '123\n'*10)*3) |
| 203 |
t1a = (patch914575_to1 + '123\n'*10)*3 |
| 204 |
f1b = '456\n'*10 + f1a |
| 205 |
t1b = '456\n'*10 + t1a |
| 206 |
f1a = f1a.splitlines() |
| 207 |
t1a = t1a.splitlines() |
| 208 |
f1b = f1b.splitlines() |
| 209 |
t1b = t1b.splitlines() |
| 210 |
f2 = patch914575_from2.splitlines() |
| 211 |
t2 = patch914575_to2.splitlines() |
| 212 |
f3 = patch914575_from3 |
| 213 |
t3 = patch914575_to3 |
| 214 |
i = difflib.HtmlDiff() |
| 215 |
j = difflib.HtmlDiff(tabsize=2) |
| 216 |
k = difflib.HtmlDiff(wrapcolumn=14) |
| 217 |
|
| 218 |
full = i.make_file(f1a,t1a,'from','to',context=False,numlines=5) |
| 219 |
tables = '\n'.join( |
| 220 |
[ |
| 221 |
'<h2>Context (first diff within numlines=5(default))</h2>', |
| 222 |
i.make_table(f1a,t1a,'from','to',context=True), |
| 223 |
'<h2>Context (first diff after numlines=5(default))</h2>', |
| 224 |
i.make_table(f1b,t1b,'from','to',context=True), |
| 225 |
'<h2>Context (numlines=6)</h2>', |
| 226 |
i.make_table(f1a,t1a,'from','to',context=True,numlines=6), |
| 227 |
'<h2>Context (numlines=0)</h2>', |
| 228 |
i.make_table(f1a,t1a,'from','to',context=True,numlines=0), |
| 229 |
'<h2>Same Context</h2>', |
| 230 |
i.make_table(f1a,f1a,'from','to',context=True), |
| 231 |
'<h2>Same Full</h2>', |
| 232 |
i.make_table(f1a,f1a,'from','to',context=False), |
| 233 |
'<h2>Empty Context</h2>', |
| 234 |
i.make_table([],[],'from','to',context=True), |
| 235 |
'<h2>Empty Full</h2>', |
| 236 |
i.make_table([],[],'from','to',context=False), |
| 237 |
'<h2>tabsize=2</h2>', |
| 238 |
j.make_table(f2,t2), |
| 239 |
'<h2>tabsize=default</h2>', |
| 240 |
i.make_table(f2,t2), |
| 241 |
'<h2>Context (wrapcolumn=14,numlines=0)</h2>', |
| 242 |
k.make_table(f3.splitlines(),t3.splitlines(),context=True,numlines=0), |
| 243 |
'<h2>wrapcolumn=14,splitlines()</h2>', |
| 244 |
k.make_table(f3.splitlines(),t3.splitlines()), |
| 245 |
'<h2>wrapcolumn=14,splitlines(True)</h2>', |
| 246 |
k.make_table(f3.splitlines(True),t3.splitlines(True)), |
| 247 |
]) |
| 248 |
actual = full.replace('</body>','\n%s\n</body>' % tables) |
| 249 |
|
| 250 |
# temporarily uncomment next two lines to baseline this test |
| 251 |
#with open('test_difflib_expect.html','w') as fp: |
| 252 |
# fp.write(actual) |
| 253 |
|
| 254 |
with open(findfile('test_difflib_expect.html'), encoding="utf-8") as fp: |
| 255 |
self.assertEqual(actual, fp.read()) |
| 256 |
|
| 257 |
def test_recursion_limit(self): |
| 258 |
# Check if the problem described in patch #1413711 exists. |
| 259 |
limit = sys.getrecursionlimit() |
| 260 |
old = [(i%2 and "K:%d" or "V:A:%d") % i for i in range(limit*2)] |
| 261 |
new = [(i%2 and "K:%d" or "V:B:%d") % i for i in range(limit*2)] |
| 262 |
difflib.SequenceMatcher(None, old, new).get_opcodes() |
| 263 |
|
| 264 |
def test_make_file_default_charset(self): |
| 265 |
html_diff = difflib.HtmlDiff() |
| 266 |
output = html_diff.make_file(patch914575_from1.splitlines(), |
| 267 |
patch914575_to1.splitlines()) |
| 268 |
self.assertIn('content="text/html; charset=utf-8"', output) |
| 269 |
|
| 270 |
def test_make_file_iso88591_charset(self): |
| 271 |
html_diff = difflib.HtmlDiff() |
| 272 |
output = html_diff.make_file(patch914575_from1.splitlines(), |
| 273 |
patch914575_to1.splitlines(), |
| 274 |
charset='iso-8859-1') |
| 275 |
self.assertIn('content="text/html; charset=iso-8859-1"', output) |
| 276 |
|
| 277 |
def test_make_file_usascii_charset_with_nonascii_input(self): |
| 278 |
html_diff = difflib.HtmlDiff() |
| 279 |
output = html_diff.make_file(patch914575_nonascii_from1.splitlines(), |
| 280 |
patch914575_nonascii_to1.splitlines(), |
| 281 |
charset='us-ascii') |
| 282 |
self.assertIn('content="text/html; charset=us-ascii"', output) |
| 283 |
self.assertIn('ımplıcıt', output) |
| 284 |
|
| 285 |
class TestDiffer(unittest.TestCase): |
| 286 |
def test_close_matches_aligned(self): |
| 287 |
# Of the 4 closely matching pairs, we want 1 to match with 3, |
| 288 |
# and 2 with 4, to align with a "top to bottom" mental model. |
| 289 |
a = ["cat\n", "dog\n", "close match 1\n", "close match 2\n"] |
| 290 |
b = ["close match 3\n", "close match 4\n", "kitten\n", "puppy\n"] |
| 291 |
m = difflib.Differ().compare(a, b) |
| 292 |
self.assertEqual(list(m), |
| 293 |
['- cat\n', |
| 294 |
'- dog\n', |
| 295 |
'- close match 1\n', |
| 296 |
'? ^\n', |
| 297 |
'+ close match 3\n', |
| 298 |
'? ^\n', |
| 299 |
'- close match 2\n', |
| 300 |
'? ^\n', |
| 301 |
'+ close match 4\n', |
| 302 |
'? ^\n', |
| 303 |
'+ kitten\n', |
| 304 |
'+ puppy\n']) |
| 305 |
|
| 306 |
def test_one_insert(self): |
| 307 |
m = difflib.Differ().compare('b' * 2, 'a' + 'b' * 2) |
| 308 |
self.assertEqual(list(m), ['+ a', ' b', ' b']) |
| 309 |
|
| 310 |
def test_one_delete(self): |
| 311 |
m = difflib.Differ().compare('a' + 'b' * 2, 'b' * 2) |
| 312 |
self.assertEqual(list(m), ['- a', ' b', ' b']) |
| 313 |
|
| 314 |
|
| 315 |
class TestOutputFormat(unittest.TestCase): |
| 316 |
def test_tab_delimiter(self): |
| 317 |
args = [['one'], ['two'], 'Original', 'Current', |
| 318 |
'2005-01-26 23:30:50', '2010-04-02 10:20:52'] |
| 319 |
ud = difflib.unified_diff(*args, lineterm='') |
| 320 |
self.assertEqual(list(ud)[0:2], [ |
| 321 |
"--- Original\t2005-01-26 23:30:50", |
| 322 |
"+++ Current\t2010-04-02 10:20:52"]) |
| 323 |
cd = difflib.context_diff(*args, lineterm='') |
| 324 |
self.assertEqual(list(cd)[0:2], [ |
| 325 |
"*** Original\t2005-01-26 23:30:50", |
| 326 |
"--- Current\t2010-04-02 10:20:52"]) |
| 327 |
|
| 328 |
def test_no_trailing_tab_on_empty_filedate(self): |
| 329 |
args = [['one'], ['two'], 'Original', 'Current'] |
| 330 |
ud = difflib.unified_diff(*args, lineterm='') |
| 331 |
self.assertEqual(list(ud)[0:2], ["--- Original", "+++ Current"]) |
| 332 |
|
| 333 |
cd = difflib.context_diff(*args, lineterm='') |
| 334 |
self.assertEqual(list(cd)[0:2], ["*** Original", "--- Current"]) |
| 335 |
|
| 336 |
def test_range_format_unified(self): |
| 337 |
# Per the diff spec at http://www.unix.org/single_unix_specification/ |
| 338 |
spec = '''\ |
| 339 |
Each <range> field shall be of the form: |
| 340 |
%1d", <beginning line number> if the range contains exactly one line, |
| 341 |
and: |
| 342 |
"%1d,%1d", <beginning line number>, <number of lines> otherwise. |
| 343 |
If a range is empty, its beginning line number shall be the number of |
| 344 |
the line just before the range, or 0 if the empty range starts the file. |
| 345 |
''' |
| 346 |
fmt = difflib._format_range_unified |
| 347 |
self.assertEqual(fmt(3,3), '3,0') |
| 348 |
self.assertEqual(fmt(3,4), '4') |
| 349 |
self.assertEqual(fmt(3,5), '4,2') |
| 350 |
self.assertEqual(fmt(3,6), '4,3') |
| 351 |
self.assertEqual(fmt(0,0), '0,0') |
| 352 |
|
| 353 |
def test_range_format_context(self): |
| 354 |
# Per the diff spec at http://www.unix.org/single_unix_specification/ |
| 355 |
spec = '''\ |
| 356 |
The range of lines in file1 shall be written in the following format |
| 357 |
if the range contains two or more lines: |
| 358 |
"*** %d,%d ****\n", <beginning line number>, <ending line number> |
| 359 |
and the following format otherwise: |
| 360 |
"*** %d ****\n", <ending line number> |
| 361 |
The ending line number of an empty range shall be the number of the preceding line, |
| 362 |
or 0 if the range is at the start of the file. |
| 363 |
|
| 364 |
Next, the range of lines in file2 shall be written in the following format |
| 365 |
if the range contains two or more lines: |
| 366 |
"--- %d,%d ----\n", <beginning line number>, <ending line number> |
| 367 |
and the following format otherwise: |
| 368 |
"--- %d ----\n", <ending line number> |
| 369 |
''' |
| 370 |
fmt = difflib._format_range_context |
| 371 |
self.assertEqual(fmt(3,3), '3') |
| 372 |
self.assertEqual(fmt(3,4), '4') |
| 373 |
self.assertEqual(fmt(3,5), '4,5') |
| 374 |
self.assertEqual(fmt(3,6), '4,6') |
| 375 |
self.assertEqual(fmt(0,0), '0') |
| 376 |
|
| 377 |
|
| 378 |
class TestBytes(unittest.TestCase): |
| 379 |
# don't really care about the content of the output, just the fact |
| 380 |
# that it's bytes and we don't crash |
| 381 |
def check(self, diff): |
| 382 |
diff = list(diff) # trigger exceptions first |
| 383 |
for line in diff: |
| 384 |
self.assertIsInstance( |
| 385 |
line, bytes, |
| 386 |
"all lines of diff should be bytes, but got: %r" % line) |
| 387 |
|
| 388 |
def test_byte_content(self): |
| 389 |
# if we receive byte strings, we return byte strings |
| 390 |
a = [b'hello', b'andr\xe9'] # iso-8859-1 bytes |
| 391 |
b = [b'hello', b'andr\xc3\xa9'] # utf-8 bytes |
| 392 |
|
| 393 |
unified = difflib.unified_diff |
| 394 |
context = difflib.context_diff |
| 395 |
|
| 396 |
check = self.check |
| 397 |
check(difflib.diff_bytes(unified, a, a)) |
| 398 |
check(difflib.diff_bytes(unified, a, b)) |
| 399 |
|
| 400 |
# now with filenames (content and filenames are all bytes!) |
| 401 |
check(difflib.diff_bytes(unified, a, a, b'a', b'a')) |
| 402 |
check(difflib.diff_bytes(unified, a, b, b'a', b'b')) |
| 403 |
|
| 404 |
# and with filenames and dates |
| 405 |
check(difflib.diff_bytes(unified, a, a, b'a', b'a', b'2005', b'2013')) |
| 406 |
check(difflib.diff_bytes(unified, a, b, b'a', b'b', b'2005', b'2013')) |
| 407 |
|
| 408 |
# same all over again, with context diff |
| 409 |
check(difflib.diff_bytes(context, a, a)) |
| 410 |
check(difflib.diff_bytes(context, a, b)) |
| 411 |
check(difflib.diff_bytes(context, a, a, b'a', b'a')) |
| 412 |
check(difflib.diff_bytes(context, a, b, b'a', b'b')) |
| 413 |
check(difflib.diff_bytes(context, a, a, b'a', b'a', b'2005', b'2013')) |
| 414 |
check(difflib.diff_bytes(context, a, b, b'a', b'b', b'2005', b'2013')) |
| 415 |
|
| 416 |
def test_byte_filenames(self): |
| 417 |
# somebody renamed a file from ISO-8859-2 to UTF-8 |
| 418 |
fna = b'\xb3odz.txt' # "łodz.txt" |
| 419 |
fnb = b'\xc5\x82odz.txt' |
| 420 |
|
| 421 |
# they transcoded the content at the same time |
| 422 |
a = [b'\xa3odz is a city in Poland.'] |
| 423 |
b = [b'\xc5\x81odz is a city in Poland.'] |
| 424 |
|
| 425 |
check = self.check |
| 426 |
unified = difflib.unified_diff |
| 427 |
context = difflib.context_diff |
| 428 |
check(difflib.diff_bytes(unified, a, b, fna, fnb)) |
| 429 |
check(difflib.diff_bytes(context, a, b, fna, fnb)) |
| 430 |
|
| 431 |
def assertDiff(expect, actual): |
| 432 |
# do not compare expect and equal as lists, because unittest |
| 433 |
# uses difflib to report difference between lists |
| 434 |
actual = list(actual) |
| 435 |
self.assertEqual(len(expect), len(actual)) |
| 436 |
for e, a in zip(expect, actual): |
| 437 |
self.assertEqual(e, a) |
| 438 |
|
| 439 |
expect = [ |
| 440 |
b'--- \xb3odz.txt', |
| 441 |
b'+++ \xc5\x82odz.txt', |
| 442 |
b'@@ -1 +1 @@', |
| 443 |
b'-\xa3odz is a city in Poland.', |
| 444 |
b'+\xc5\x81odz is a city in Poland.', |
| 445 |
] |
| 446 |
actual = difflib.diff_bytes(unified, a, b, fna, fnb, lineterm=b'') |
| 447 |
assertDiff(expect, actual) |
| 448 |
|
| 449 |
# with dates (plain ASCII) |
| 450 |
datea = b'2005-03-18' |
| 451 |
dateb = b'2005-03-19' |
| 452 |
check(difflib.diff_bytes(unified, a, b, fna, fnb, datea, dateb)) |
| 453 |
check(difflib.diff_bytes(context, a, b, fna, fnb, datea, dateb)) |
| 454 |
|
| 455 |
expect = [ |
| 456 |
# note the mixed encodings here: this is deeply wrong by every |
| 457 |
# tenet of Unicode, but it doesn't crash, it's parseable by |
| 458 |
# patch, and it's how UNIX(tm) diff behaves |
| 459 |
b'--- \xb3odz.txt\t2005-03-18', |
| 460 |
b'+++ \xc5\x82odz.txt\t2005-03-19', |
| 461 |
b'@@ -1 +1 @@', |
| 462 |
b'-\xa3odz is a city in Poland.', |
| 463 |
b'+\xc5\x81odz is a city in Poland.', |
| 464 |
] |
| 465 |
actual = difflib.diff_bytes(unified, a, b, fna, fnb, datea, dateb, |
| 466 |
lineterm=b'') |
| 467 |
assertDiff(expect, actual) |
| 468 |
|
| 469 |
|
| 470 |
class TestInputTypes(unittest.TestCase): |
| 471 |
def _assert_type_error(self, msg, generator, *args): |
| 472 |
with self.assertRaises(TypeError) as ctx: |
| 473 |
list(generator(*args)) |
| 474 |
self.assertEqual(msg, str(ctx.exception)) |
| 475 |
|
| 476 |
def test_input_type_checks(self): |
| 477 |
unified = difflib.unified_diff |
| 478 |
context = difflib.context_diff |
| 479 |
|
| 480 |
expect = "input must be a sequence of strings, not str" |
| 481 |
self._assert_type_error(expect, unified, 'a', ['b']) |
| 482 |
self._assert_type_error(expect, context, 'a', ['b']) |
| 483 |
|
| 484 |
self._assert_type_error(expect, unified, ['a'], 'b') |
| 485 |
self._assert_type_error(expect, context, ['a'], 'b') |
| 486 |
|
| 487 |
expect = "lines to compare must be str, not NoneType (None)" |
| 488 |
self._assert_type_error(expect, unified, ['a'], [None]) |
| 489 |
self._assert_type_error(expect, context, ['a'], [None]) |
| 490 |
|
| 491 |
def test_mixed_types_content(self): |
| 492 |
# type of input content must be consistent: all str or all bytes |
| 493 |
a = [b'hello'] |
| 494 |
b = ['hello'] |
| 495 |
|
| 496 |
unified = difflib.unified_diff |
| 497 |
context = difflib.context_diff |
| 498 |
|
| 499 |
expect = "lines to compare must be str, not bytes (b'hello')" |
| 500 |
self._assert_type_error(expect, unified, a, b) |
| 501 |
self._assert_type_error(expect, unified, b, a) |
| 502 |
self._assert_type_error(expect, context, a, b) |
| 503 |
self._assert_type_error(expect, context, b, a) |
| 504 |
|
| 505 |
expect = "all arguments must be bytes, not str ('hello')" |
| 506 |
self._assert_type_error(expect, difflib.diff_bytes, unified, a, b) |
| 507 |
self._assert_type_error(expect, difflib.diff_bytes, unified, b, a) |
| 508 |
self._assert_type_error(expect, difflib.diff_bytes, context, a, b) |
| 509 |
self._assert_type_error(expect, difflib.diff_bytes, context, b, a) |
| 510 |
|
| 511 |
def test_mixed_types_filenames(self): |
| 512 |
# cannot pass filenames as bytes if content is str (this may not be |
| 513 |
# the right behaviour, but at least the test demonstrates how |
| 514 |
# things work) |
| 515 |
a = ['hello\n'] |
| 516 |
b = ['ohell\n'] |
| 517 |
fna = b'ol\xe9.txt' # filename transcoded from ISO-8859-1 |
| 518 |
fnb = b'ol\xc3a9.txt' # to UTF-8 |
| 519 |
self._assert_type_error( |
| 520 |
"all arguments must be str, not: b'ol\\xe9.txt'", |
| 521 |
difflib.unified_diff, a, b, fna, fnb) |
| 522 |
|
| 523 |
def test_mixed_types_dates(self): |
| 524 |
# type of dates must be consistent with type of contents |
| 525 |
a = [b'foo\n'] |
| 526 |
b = [b'bar\n'] |
| 527 |
datea = '1 fév' |
| 528 |
dateb = '3 fév' |
| 529 |
self._assert_type_error( |
| 530 |
"all arguments must be bytes, not str ('1 fév')", |
| 531 |
difflib.diff_bytes, difflib.unified_diff, |
| 532 |
a, b, b'a', b'b', datea, dateb) |
| 533 |
|
| 534 |
# if input is str, non-ASCII dates are fine |
| 535 |
a = ['foo\n'] |
| 536 |
b = ['bar\n'] |
| 537 |
list(difflib.unified_diff(a, b, 'a', 'b', datea, dateb)) |
| 538 |
|
| 539 |
|
| 540 |
class TestJunkAPIs(unittest.TestCase): |
| 541 |
def test_is_line_junk_true(self): |
| 542 |
for line in ['#', ' ', ' #', '# ', ' # ', '']: |
| 543 |
self.assertTrue(difflib.IS_LINE_JUNK(line), repr(line)) |
| 544 |
|
| 545 |
def test_is_line_junk_false(self): |
| 546 |
for line in ['##', ' ##', '## ', 'abc ', 'abc #', 'Mr. Moose is up!']: |
| 547 |
self.assertFalse(difflib.IS_LINE_JUNK(line), repr(line)) |
| 548 |
|
| 549 |
def test_is_line_junk_REDOS(self): |
| 550 |
evil_input = ('\t' * 1000000) + '##' |
| 551 |
self.assertFalse(difflib.IS_LINE_JUNK(evil_input)) |
| 552 |
|
| 553 |
def test_is_character_junk_true(self): |
| 554 |
for char in [' ', '\t']: |
| 555 |
self.assertTrue(difflib.IS_CHARACTER_JUNK(char), repr(char)) |
| 556 |
|
| 557 |
def test_is_character_junk_false(self): |
| 558 |
for char in ['a', '#', '\n', '\f', '\r', '\v']: |
| 559 |
self.assertFalse(difflib.IS_CHARACTER_JUNK(char), repr(char)) |
| 560 |
|
| 561 |
class TestFindLongest(unittest.TestCase): |
| 562 |
def longer_match_exists(self, a, b, n): |
| 563 |
return any(b_part in a for b_part in |
| 564 |
[b[i:i + n + 1] for i in range(0, len(b) - n - 1)]) |
| 565 |
|
| 566 |
def test_default_args(self): |
| 567 |
a = 'foo bar' |
| 568 |
b = 'foo baz bar' |
| 569 |
sm = difflib.SequenceMatcher(a=a, b=b) |
| 570 |
match = sm.find_longest_match() |
| 571 |
self.assertEqual(match.a, 0) |
| 572 |
self.assertEqual(match.b, 0) |
| 573 |
self.assertEqual(match.size, 6) |
| 574 |
self.assertEqual(a[match.a: match.a + match.size], |
| 575 |
b[match.b: match.b + match.size]) |
| 576 |
self.assertFalse(self.longer_match_exists(a, b, match.size)) |
| 577 |
|
| 578 |
match = sm.find_longest_match(alo=2, blo=4) |
| 579 |
self.assertEqual(match.a, 3) |
| 580 |
self.assertEqual(match.b, 7) |
| 581 |
self.assertEqual(match.size, 4) |
| 582 |
self.assertEqual(a[match.a: match.a + match.size], |
| 583 |
b[match.b: match.b + match.size]) |
| 584 |
self.assertFalse(self.longer_match_exists(a[2:], b[4:], match.size)) |
| 585 |
|
| 586 |
match = sm.find_longest_match(bhi=5, blo=1) |
| 587 |
self.assertEqual(match.a, 1) |
| 588 |
self.assertEqual(match.b, 1) |
| 589 |
self.assertEqual(match.size, 4) |
| 590 |
self.assertEqual(a[match.a: match.a + match.size], |
| 591 |
b[match.b: match.b + match.size]) |
| 592 |
self.assertFalse(self.longer_match_exists(a, b[1:5], match.size)) |
| 593 |
|
| 594 |
def test_longest_match_with_popular_chars(self): |
| 595 |
a = 'dabcd' |
| 596 |
b = 'd'*100 + 'abc' + 'd'*100 # length over 200 so popular used |
| 597 |
sm = difflib.SequenceMatcher(a=a, b=b) |
| 598 |
match = sm.find_longest_match(0, len(a), 0, len(b)) |
| 599 |
self.assertEqual(match.a, 0) |
| 600 |
self.assertEqual(match.b, 99) |
| 601 |
self.assertEqual(match.size, 5) |
| 602 |
self.assertEqual(a[match.a: match.a + match.size], |
| 603 |
b[match.b: match.b + match.size]) |
| 604 |
self.assertFalse(self.longer_match_exists(a, b, match.size)) |
| 605 |
|
| 606 |
|
| 607 |
class TestCloseMatches(unittest.TestCase): |
| 608 |
# Happy paths are tested in the doctests of `difflib.get_close_matches`. |
| 609 |
|
| 610 |
def test_invalid_inputs(self): |
| 611 |
self.assertRaises(ValueError, difflib.get_close_matches, "spam", ['egg'], n=0) |
| 612 |
self.assertRaises(ValueError, difflib.get_close_matches, "spam", ['egg'], n=-1) |
| 613 |
self.assertRaises(ValueError, difflib.get_close_matches, "spam", ['egg'], cutoff=1.1) |
| 614 |
self.assertRaises(ValueError, difflib.get_close_matches, "spam", ['egg'], cutoff=-0.1) |
| 615 |
|
| 616 |
|
| 617 |
class TestRestore(unittest.TestCase): |
| 618 |
# Happy paths are tested in the doctests of `difflib.restore`. |
| 619 |
|
| 620 |
def test_invalid_input(self): |
| 621 |
with self.assertRaises(ValueError): |
| 622 |
''.join(difflib.restore([], 0)) |
| 623 |
with self.assertRaises(ValueError): |
| 624 |
''.join(difflib.restore([], 3)) |
| 625 |
|
| 626 |
|
| 627 |
def setUpModule(): |
| 628 |
difflib.HtmlDiff._default_prefix = 0 |
| 629 |
|
| 630 |
|
| 631 |
def load_tests(loader, tests, pattern): |
| 632 |
tests.addTest(doctest.DocTestSuite(difflib)) |
| 633 |
return tests |
| 634 |
|
| 635 |
|
| 636 |
if __name__ == '__main__': |
| 637 |
unittest.main() |