[BUG] test_predefined_character_classes failed with seed 4

[thirtiseven]

Changing seed to some other values like 4 will fail test_predefined_character_classes:

def test_predefined_character_classes():
    gen = mk_str_gen('[a-zA-Z]{0,2}[\r\n!"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~]{0,2}[0-9]{0,2}')
    assert_gpu_and_cpu_are_equal_collect(
        lambda spark: unary_op_df(spark, gen, seed=4).selectExpr(
            'regexp_replace(a, "\\\\p{Lower}", "x")',
            'regexp_replace(a, "\\\\p{Upper}", "x")',
            'regexp_replace(a, "\\\\p{ASCII}", "x")',
            'regexp_replace(a, "\\\\p{Alpha}", "x")',
            'regexp_replace(a, "\\\\p{Digit}", "x")',
            'regexp_replace(a, "\\\\p{Alnum}", "x")',
            'regexp_replace(a, "\\\\p{Punct}", "x")',
            'regexp_replace(a, "\\\\p{Graph}", "x")',
            'regexp_replace(a, "\\\\p{Print}", "x")',
            'regexp_replace(a, "\\\\p{Blank}", "x")',
            'regexp_replace(a, "\\\\p{Cntrl}", "x")',
            'regexp_replace(a, "\\\\p{XDigit}", "x")',
            'regexp_replace(a, "\\\\p{Space}", "x")',
            'regexp_replace(a, "\\\\P{Lower}", "x")',
            'regexp_replace(a, "\\\\P{Upper}", "x")',
            'regexp_replace(a, "\\\\P{ASCII}", "x")',
            'regexp_replace(a, "\\\\P{Alpha}", "x")',
            'regexp_replace(a, "\\\\P{Digit}", "x")',
            'regexp_replace(a, "\\\\P{Alnum}", "x")',
            'regexp_replace(a, "\\\\P{Punct}", "x")',
            'regexp_replace(a, "\\\\P{Graph}", "x")',
            'regexp_replace(a, "\\\\P{Print}", "x")',
            'regexp_replace(a, "\\\\P{Blank}", "x")',
            'regexp_replace(a, "\\\\P{Cntrl}", "x")',
            'regexp_replace(a, "\\\\P{XDigit}", "x")',
            'regexp_replace(a, "\\\\P{Space}", "x")',
        ),
        conf=_regexp_conf)

log:

______________________________________________________________ test_predefined_character_classes ______________________________________________________________
[gw0] linux -- Python 3.8.3 /home/haoyangl/.pyenv/versions/3.8.3/bin/python

    def test_predefined_character_classes():
        gen = mk_str_gen('[a-zA-Z]{0,2}[\r\n!"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~]{0,2}[0-9]{0,2}')
>       assert_gpu_and_cpu_are_equal_collect(
            lambda spark: unary_op_df(spark, gen, seed=4).selectExpr(
                'regexp_replace(a, "\\\\p{Lower}", "x")',
                'regexp_replace(a, "\\\\p{Upper}", "x")',
                'regexp_replace(a, "\\\\p{ASCII}", "x")',
                'regexp_replace(a, "\\\\p{Alpha}", "x")',
                'regexp_replace(a, "\\\\p{Digit}", "x")',
                'regexp_replace(a, "\\\\p{Alnum}", "x")',
                'regexp_replace(a, "\\\\p{Punct}", "x")',
                'regexp_replace(a, "\\\\p{Graph}", "x")',
                'regexp_replace(a, "\\\\p{Print}", "x")',
                'regexp_replace(a, "\\\\p{Blank}", "x")',
                'regexp_replace(a, "\\\\p{Cntrl}", "x")',
                'regexp_replace(a, "\\\\p{XDigit}", "x")',
                'regexp_replace(a, "\\\\p{Space}", "x")',
                'regexp_replace(a, "\\\\P{Lower}", "x")',
                'regexp_replace(a, "\\\\P{Upper}", "x")',
                'regexp_replace(a, "\\\\P{ASCII}", "x")',
                'regexp_replace(a, "\\\\P{Alpha}", "x")',
                'regexp_replace(a, "\\\\P{Digit}", "x")',
                'regexp_replace(a, "\\\\P{Alnum}", "x")',
                'regexp_replace(a, "\\\\P{Punct}", "x")',
                'regexp_replace(a, "\\\\P{Graph}", "x")',
                'regexp_replace(a, "\\\\P{Print}", "x")',
                'regexp_replace(a, "\\\\P{Blank}", "x")',
                'regexp_replace(a, "\\\\P{Cntrl}", "x")',
                'regexp_replace(a, "\\\\P{XDigit}", "x")',
                'regexp_replace(a, "\\\\P{Space}", "x")',
            ),
            conf=_regexp_conf)

../../src/main/python/regexp_test.py:673:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
../../src/main/python/asserts.py:566: in assert_gpu_and_cpu_are_equal_collect
    _assert_gpu_and_cpu_are_equal(func, 'COLLECT', conf=conf, is_cpu_first=is_cpu_first)
../../src/main/python/asserts.py:497: in _assert_gpu_and_cpu_are_equal
    assert_equal(from_cpu, from_gpu)
../../src/main/python/asserts.py:107: in assert_equal
    _assert_equal(cpu, gpu, float_check=get_float_check(), path=[])
../../src/main/python/asserts.py:43: in _assert_equal
    _assert_equal(cpu[index], gpu[index], float_check, path + [index])
../../src/main/python/asserts.py:36: in _assert_equal
    _assert_equal(cpu[field], gpu[field], float_check, path + [field])
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

cpu = 'ÝM½¨x\x97Ð\x0fxö', gpu = 'ÝM½¨\\\x97Ð\x0fxö', float_check = <function get_float_check.<locals>.<lambda> at 0x7fa0bb258820>
path = [1261, 'regexp_replace(a, \\p{Punct}, x, 1)']

    def _assert_equal(cpu, gpu, float_check, path):
        t = type(cpu)
        if (t is Row):
            assert len(cpu) == len(gpu), "CPU and GPU row have different lengths at {} CPU: {} GPU: {}".format(path, len(cpu), len(gpu))
            if hasattr(cpu, "__fields__") and hasattr(gpu, "__fields__"):
                assert cpu.__fields__ == gpu.__fields__, "CPU and GPU row have different fields at {} CPU: {} GPU: {}".format(path, cpu.__fields__, gpu.__fields__)
                for field in cpu.__fields__:
                    _assert_equal(cpu[field], gpu[field], float_check, path + [field])
            else:
                for index in range(len(cpu)):
                    _assert_equal(cpu[index], gpu[index], float_check, path + [index])
        elif (t is list):
            assert len(cpu) == len(gpu), "CPU and GPU list have different lengths at {} CPU: {} GPU: {}".format(path, len(cpu), len(gpu))
            for index in range(len(cpu)):
                _assert_equal(cpu[index], gpu[index], float_check, path + [index])
        elif (t is tuple):
            assert len(cpu) == len(gpu), "CPU and GPU list have different lengths at {} CPU: {} GPU: {}".format(path, len(cpu), len(gpu))
            for index in range(len(cpu)):
                _assert_equal(cpu[index], gpu[index], float_check, path + [index])
        elif (t is pytypes.GeneratorType):
            index = 0
            # generator has no zip :( so we have to do this the hard way
            done = False
            while not done:
                sub_cpu = None
                sub_gpu = None
                try:
                    sub_cpu = next(cpu)
                except StopIteration:
                    done = True

                try:
                    sub_gpu = next(gpu)
                except StopIteration:
                    done = True

                if done:
                    assert sub_cpu == sub_gpu and sub_cpu == None, "CPU and GPU generators have different lengths at {}".format(path)
                else:
                    _assert_equal(sub_cpu, sub_gpu, float_check, path + [index])

                index = index + 1
        elif (t is dict):
            # The order of key/values is not guaranteed in python dicts, nor are they guaranteed by Spark
            # so sort the items to do our best with ignoring the order of dicts
            cpu_items = list(cpu.items()).sort(key=_RowCmp)
            gpu_items = list(gpu.items()).sort(key=_RowCmp)
            _assert_equal(cpu_items, gpu_items, float_check, path + ["map"])
        elif (t is int):
            assert cpu == gpu, "GPU and CPU int values are different at {}".format(path)
        elif (t is float):
            if (math.isnan(cpu)):
                assert math.isnan(gpu), "GPU and CPU float values are different at {}".format(path)
            else:
                assert float_check(cpu, gpu), "GPU and CPU float values are different {}".format(path)
        elif isinstance(cpu, str):
>           assert cpu == gpu, "GPU and CPU string values are different at {}".format(path)
E           AssertionError: GPU and CPU string values are different at [1261, 'regexp_replace(a, \\p{Punct}, x, 1)']

../../src/main/python/asserts.py:85: AssertionError
-------------------------------------------------------------------- Captured stdout call ---------------------------------------------------------------------
### CPU RUN ###
### GPU RUN ###
### COLLECT: GPU TOOK 0.5829105377197266 CPU TOOK 3.3358943462371826 ###
--- CPU OUTPUT
+++ GPU OUTPUT
@@ -1259,7 +1259,7 @@
 Row(regexp_replace(a, \p{Lower}, x, 1)='xR<`40', regexp_replace(a, \p{Upper}, x, 1)='tx<`40', regexp_replace(a, \p{ASCII}, x, 1)='xxxxxx', regexp_replace(a, \p{Alpha}, x, 1)='xx<`40', regexp_replace(a, \p{Digit}, x, 1)='tR<`xx', regexp_replace(a, \p{Alnum}, x, 1)='xx<`xx', regexp_replace(a, \p{Punct}, x, 1)='tRxx40', regexp_replace(a, \p{Graph}, x, 1)='xxxxxx', regexp_replace(a, \p{Print}, x, 1)='xxxxxx', regexp_replace(a, \p{Blank}, x, 1)='tR<`40', regexp_replace(a, \p{Cntrl}, x, 1)='tR<`40', regexp_replace(a, \p{XDigit}, x, 1)='tR<`xx', regexp_replace(a, \p{Space}, x, 1)='tR<`40', regexp_replace(a, \P{Lower}, x, 1)='txxxxx', regexp_replace(a, \P{Upper}, x, 1)='xRxxxx', regexp_replace(a, \P{ASCII}, x, 1)='tR<`40', regexp_replace(a, \P{Alpha}, x, 1)='tRxxxx', regexp_replace(a, \P{Digit}, x, 1)='xxxx40', regexp_replace(a, \P{Alnum}, x, 1)='tRxx40', regexp_replace(a, \P{Punct}, x, 1)='xx<`xx', regexp_replace(a, \P{Graph}, x, 1)='tR<`40', regexp_replace(a, \P{Print}, x, 1)='tR<`40', regexp_replace(a, \P{Blank}, x, 1)='xxxxxx', regexp_replace(a, \P{Cntrl}, x, 1)='xxxxxx', regexp_replace(a, \P{XDigit}, x, 1)='xxxx40', regexp_replace(a, \P{Space}, x, 1)='xxxxxx')
 Row(regexp_replace(a, \p{Lower}, x, 1)='MW?{04', regexp_replace(a, \p{Upper}, x, 1)='xx?{04', regexp_replace(a, \p{ASCII}, x, 1)='xxxxxx', regexp_replace(a, \p{Alpha}, x, 1)='xx?{04', regexp_replace(a, \p{Digit}, x, 1)='MW?{xx', regexp_replace(a, \p{Alnum}, x, 1)='xx?{xx', regexp_replace(a, \p{Punct}, x, 1)='MWxx04', regexp_replace(a, \p{Graph}, x, 1)='xxxxxx', regexp_replace(a, \p{Print}, x, 1)='xxxxxx', regexp_replace(a, \p{Blank}, x, 1)='MW?{04', regexp_replace(a, \p{Cntrl}, x, 1)='MW?{04', regexp_replace(a, \p{XDigit}, x, 1)='MW?{xx', regexp_replace(a, \p{Space}, x, 1)='MW?{04', regexp_replace(a, \P{Lower}, x, 1)='xxxxxx', regexp_replace(a, \P{Upper}, x, 1)='MWxxxx', regexp_replace(a, \P{ASCII}, x, 1)='MW?{04', regexp_replace(a, \P{Alpha}, x, 1)='MWxxxx', regexp_replace(a, \P{Digit}, x, 1)='xxxx04', regexp_replace(a, \P{Alnum}, x, 1)='MWxx04', regexp_replace(a, \P{Punct}, x, 1)='xx?{xx', regexp_replace(a, \P{Graph}, x, 1)='MW?{04', regexp_replace(a, \P{Print}, x, 1)='MW?{04', regexp_replace(a, \P{Blank}, x, 1)='xxxxxx', regexp_replace(a, \P{Cntrl}, x, 1)='xxxxxx', regexp_replace(a, \P{XDigit}, x, 1)='xxxx04', regexp_replace(a, \P{Space}, x, 1)='xxxxxx')
 Row(regexp_replace(a, \p{Lower}, x, 1)='xx!{50', regexp_replace(a, \p{Upper}, x, 1)='ma!{50', regexp_replace(a, \p{ASCII}, x, 1)='xxxxxx', regexp_replace(a, \p{Alpha}, x, 1)='xx!{50', regexp_replace(a, \p{Digit}, x, 1)='ma!{xx', regexp_replace(a, \p{Alnum}, x, 1)='xx!{xx', regexp_replace(a, \p{Punct}, x, 1)='maxx50', regexp_replace(a, \p{Graph}, x, 1)='xxxxxx', regexp_replace(a, \p{Print}, x, 1)='xxxxxx', regexp_replace(a, \p{Blank}, x, 1)='ma!{50', regexp_replace(a, \p{Cntrl}, x, 1)='ma!{50', regexp_replace(a, \p{XDigit}, x, 1)='mx!{xx', regexp_replace(a, \p{Space}, x, 1)='ma!{50', regexp_replace(a, \P{Lower}, x, 1)='maxxxx', regexp_replace(a, \P{Upper}, x, 1)='xxxxxx', regexp_replace(a, \P{ASCII}, x, 1)='ma!{50', regexp_replace(a, \P{Alpha}, x, 1)='maxxxx', regexp_replace(a, \P{Digit}, x, 1)='xxxx50', regexp_replace(a, \P{Alnum}, x, 1)='maxx50', regexp_replace(a, \P{Punct}, x, 1)='xx!{xx', regexp_replace(a, \P{Graph}, x, 1)='ma!{50', regexp_replace(a, \P{Print}, x, 1)='ma!{50', regexp_replace(a, \P{Blank}, x, 1)='xxxxxx', regexp_replace(a, \P{Cntrl}, x, 1)='xxxxxx', regexp_replace(a, \P{XDigit}, x, 1)='xaxx50', regexp_replace(a, \P{Space}, x, 1)='xxxxxx')
-Row(regexp_replace(a, \p{Lower}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Upper}, x, 1)='Ýx½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{ASCII}, x, 1)='Ýx½¨x\x97Ðxxö', regexp_replace(a, \p{Alpha}, x, 1)='Ýx½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Digit}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Alnum}, x, 1)='Ýx½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Punct}, x, 1)='ÝM½¨x\x97Ð\x0fxö', regexp_replace(a, \p{Graph}, x, 1)='Ýx½¨x\x97Ð\x0fxö', regexp_replace(a, \p{Print}, x, 1)='Ýx½¨x\x97Ð\x0fxö', regexp_replace(a, \p{Blank}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Cntrl}, x, 1)='ÝM½¨\\\x97Ðx]ö', regexp_replace(a, \p{XDigit}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Space}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \P{Lower}, x, 1)='xxxxxxxxxx', regexp_replace(a, \P{Upper}, x, 1)='xMxxxxxxxx', regexp_replace(a, \P{ASCII}, x, 1)='xMxx\\xx\x0f]x', regexp_replace(a, \P{Alpha}, x, 1)='xMxxxxxxxx', regexp_replace(a, \P{Digit}, x, 1)='xxxxxxxxxx', regexp_replace(a, \P{Alnum}, x, 1)='xMxxxxxxxx', regexp_replace(a, \P{Punct}, x, 1)='xxxx\\xxx]x', regexp_replace(a, \P{Graph}, x, 1)='xMxx\\xxx]x', regexp_replace(a, \P{Print}, x, 1)='xMxx\\xxx]x', regexp_replace(a, \P{Blank}, x, 1)='xxxxxxxxxx', regexp_replace(a, \P{Cntrl}, x, 1)='xxxxxxx\x0fxx', regexp_replace(a, \P{XDigit}, x, 1)='xxxxxxxxxx', regexp_replace(a, \P{Space}, x, 1)='xxxxxxxxxx')
+Row(regexp_replace(a, \p{Lower}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Upper}, x, 1)='Ýx½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{ASCII}, x, 1)='Ýx½¨x\x97Ðxxö', regexp_replace(a, \p{Alpha}, x, 1)='Ýx½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Digit}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Alnum}, x, 1)='Ýx½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Punct}, x, 1)='ÝM½¨\\\x97Ð\x0fxö', regexp_replace(a, \p{Graph}, x, 1)='Ýx½¨\\\x97Ð\x0fxö', regexp_replace(a, \p{Print}, x, 1)='Ýx½¨\\\x97Ð\x0fxö', regexp_replace(a, \p{Blank}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Cntrl}, x, 1)='ÝM½¨\\\x97Ðx]ö', regexp_replace(a, \p{XDigit}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \p{Space}, x, 1)='ÝM½¨\\\x97Ð\x0f]ö', regexp_replace(a, \P{Lower}, x, 1)='xxxxxxxxxx', regexp_replace(a, \P{Upper}, x, 1)='xMxxxxxxxx', regexp_replace(a, \P{ASCII}, x, 1)='xMxx\\xx\x0f]x', regexp_replace(a, \P{Alpha}, x, 1)='xMxxxxxxxx', regexp_replace(a, \P{Digit}, x, 1)='xxxxxxxxxx', regexp_replace(a, \P{Alnum}, x, 1)='xMxxxxxxxx', regexp_replace(a, \P{Punct}, x, 1)='xxxxxxxx]x', regexp_replace(a, \P{Graph}, x, 1)='xMxxxxxx]x', regexp_replace(a, \P{Print}, x, 1)='xMxxxxxx]x', regexp_replace(a, \P{Blank}, x, 1)='xxxxxxxxxx', regexp_replace(a, \P{Cntrl}, x, 1)='xxxxxxx\x0fxx', regexp_replace(a, \P{XDigit}, x, 1)='xxxxxxxxxx', regexp_replace(a, \P{Space}, x, 1)='xxxxxxxxxx')
 Row(regexp_replace(a, \p{Lower}, x, 1)='YE.[64', regexp_replace(a, \p{Upper}, x, 1)='xx.[64', regexp_replace(a, \p{ASCII}, x, 1)='xxxxxx', regexp_replace(a, \p{Alpha}, x, 1)='xx.[64', regexp_replace(a, \p{Digit}, x, 1)='YE.[xx', regexp_replace(a, \p{Alnum}, x, 1)='xx.[xx', regexp_replace(a, \p{Punct}, x, 1)='YExx64', regexp_replace(a, \p{Graph}, x, 1)='xxxxxx', regexp_replace(a, \p{Print}, x, 1)='xxxxxx', regexp_replace(a, \p{Blank}, x, 1)='YE.[64', regexp_replace(a, \p{Cntrl}, x, 1)='YE.[64', regexp_replace(a, \p{XDigit}, x, 1)='Yx.[xx', regexp_replace(a, \p{Space}, x, 1)='YE.[64', regexp_replace(a, \P{Lower}, x, 1)='xxxxxx', regexp_replace(a, \P{Upper}, x, 1)='YExxxx', regexp_replace(a, \P{ASCII}, x, 1)='YE.[64', regexp_replace(a, \P{Alpha}, x, 1)='YExxxx', regexp_replace(a, \P{Digit}, x, 1)='xxxx64', regexp_replace(a, \P{Alnum}, x, 1)='YExx64', regexp_replace(a, \P{Punct}, x, 1)='xx.[xx', regexp_replace(a, \P{Graph}, x, 1)='YE.[64', regexp_replace(a, \P{Print}, x, 1)='YE.[64', regexp_replace(a, \P{Blank}, x, 1)='xxxxxx', regexp_replace(a, \P{Cntrl}, x, 1)='xxxxxx', regexp_replace(a, \P{XDigit}, x, 1)='xExx64', regexp_replace(a, \P{Space}, x, 1)='xxxxxx')
 Row(regexp_replace(a, \p{Lower}, x, 1)='xW@`88', regexp_replace(a, \p{Upper}, x, 1)='ex@`88', regexp_replace(a, \p{ASCII}, x, 1)='xxxxxx', regexp_replace(a, \p{Alpha}, x, 1)='xx@`88', regexp_replace(a, \p{Digit}, x, 1)='eW@`xx', regexp_replace(a, \p{Alnum}, x, 1)='xx@`xx', regexp_replace(a, \p{Punct}, x, 1)='eWxx88', regexp_replace(a, \p{Graph}, x, 1)='xxxxxx', regexp_replace(a, \p{Print}, x, 1)='xxxxxx', regexp_replace(a, \p{Blank}, x, 1)='eW@`88', regexp_replace(a, \p{Cntrl}, x, 1)='eW@`88', regexp_replace(a, \p{XDigit}, x, 1)='xW@`xx', regexp_replace(a, \p{Space}, x, 1)='eW@`88', regexp_replace(a, \P{Lower}, x, 1)='exxxxx', regexp_replace(a, \P{Upper}, x, 1)='xWxxxx', regexp_replace(a, \P{ASCII}, x, 1)='eW@`88', regexp_replace(a, \P{Alpha}, x, 1)='eWxxxx', regexp_replace(a, \P{Digit}, x, 1)='xxxx88', regexp_replace(a, \P{Alnum}, x, 1)='eWxx88', regexp_replace(a, \P{Punct}, x, 1)='xx@`xx', regexp_replace(a, \P{Graph}, x, 1)='eW@`88', regexp_replace(a, \P{Print}, x, 1)='eW@`88', regexp_replace(a, \P{Blank}, x, 1)='xxxxxx', regexp_replace(a, \P{Cntrl}, x, 1)='xxxxxx', regexp_replace(a, \P{XDigit}, x, 1)='exxx88', regexp_replace(a, \P{Space}, x, 1)='xxxxxx')
 Row(regexp_replace(a, \p{Lower}, x, 1)='Bx{^57', regexp_replace(a, \p{Upper}, x, 1)='xg{^57', regexp_replace(a, \p{ASCII}, x, 1)='xxxxxx', regexp_replace(a, \p{Alpha}, x, 1)='xx{^57', regexp_replace(a, \p{Digit}, x, 1)='Bg{^xx', regexp_replace(a, \p{Alnum}, x, 1)='xx{^xx', regexp_replace(a, \p{Punct}, x, 1)='Bgxx57', regexp_replace(a, \p{Graph}, x, 1)='xxxxxx', regexp_replace(a, \p{Print}, x, 1)='xxxxxx', regexp_replace(a, \p{Blank}, x, 1)='Bg{^57', regexp_replace(a, \p{Cntrl}, x, 1)='Bg{^57', regexp_replace(a, \p{XDigit}, x, 1)='xg{^xx', regexp_replace(a, \p{Space}, x, 1)='Bg{^57', regexp_replace(a, \P{Lower}, x, 1)='xgxxxx', regexp_replace(a, \P{Upper}, x, 1)='Bxxxxx', regexp_replace(a, \P{ASCII}, x, 1)='Bg{^57', regexp_replace(a, \P{Alpha}, x, 1)='Bgxxxx', regexp_replace(a, \P{Digit}, x, 1)='xxxx57', regexp_replace(a, \P{Alnum}, x, 1)='Bgxx57', regexp_replace(a, \P{Punct}, x, 1)='xx{^xx', regexp_replace(a, \P{Graph}, x, 1)='Bg{^57', regexp_replace(a, \P{Print}, x, 1)='Bg{^57', regexp_replace(a, \P{Blank}, x, 1)='xxxxxx', regexp_replace(a, \P{Cntrl}, x, 1)='xxxxxx', regexp_replace(a, \P{XDigit}, x, 1)='Bxxx57', regexp_replace(a, \P{Space}, x, 1)='xxxxxx')

[revans2]

This one looks like a real problem too and I need to dig into it a bit more.

[revans2]

It looks like we tried to include \ in the list of characters for Punct, but failed to escape it when we generated the character set, so it ended up just escaping another character instead.