from test import test_support

from test.test_support import bigmemtest, _1G, _2G, _4G, precisionbigmemtest

import unittest

import operator

import string

import sys

# Bigmem testing houserules:

#

#  - Try not to allocate too many large objects. It's okay to rely on

#    refcounting semantics, but don't forget that 's = create_largestring()'

#    doesn't release the old 's' (if it exists) until well after its new

#    value has been created. Use 'del s' before the create_largestring call.

#

#  - Do *not* compare large objects using assertEquals or similar. It's a

#    lengty operation and the errormessage will be utterly useless due to

#    its size. To make sure whether a result has the right contents, better

#    to use the strip or count methods, or compare meaningful slices.

#

#  - Don't forget to test for large indices, offsets and results and such,

#    in addition to large sizes.

#

#  - When repeating an object (say, a substring, or a small list) to create

#    a large object, make the subobject of a length that is not a power of

#    2. That way, int-wrapping problems are more easily detected.

#

#  - While the bigmemtest decorator speaks of 'minsize', all tests will

#    actually be called with a much smaller number too, in the normal

#    test run (5Kb currently.) This is so the tests themselves get frequent

#    testing. Consequently, always make all large allocations based on the

#    passed-in 'size', and don't rely on the size being very large. Also,

#    memuse-per-size should remain sane (less than a few thousand); if your

#    test uses more, adjust 'size' upward, instead.

class StrTest(unittest.TestCase):

    @bigmemtest(minsize=_2G, memuse=2)

    def test_capitalize(self, size):

        SUBSTR = ' abc def ghi'

        s = '-' * size + SUBSTR

        caps = s.capitalize()

        self.assertEquals(caps[-len(SUBSTR):],

                         SUBSTR.capitalize())

        self.assertEquals(caps.lstrip('-'), SUBSTR)

    @bigmemtest(minsize=_2G + 10, memuse=1)

    def test_center(self, size):

        SUBSTR = ' abc def ghi'

        s = SUBSTR.center(size)

        self.assertEquals(len(s), size)

        lpadsize = rpadsize = (len(s) - len(SUBSTR)) // 2

        if len(s) % 2:

            lpadsize += 1

        self.assertEquals(s[lpadsize:-rpadsize], SUBSTR)

        self.assertEquals(s.strip(), SUBSTR.strip())

    @precisionbigmemtest(size=_2G - 1, memuse=1)

    def test_center_unicode(self, size):

        SUBSTR = u' abc def ghi'

        try:

            s = SUBSTR.center(size)

        except OverflowError:

            pass # acceptable on 32-bit

        else:

            self.assertEquals(len(s), size)

            lpadsize = rpadsize = (len(s) - len(SUBSTR)) // 2

            if len(s) % 2:

                lpadsize += 1

            self.assertEquals(s[lpadsize:-rpadsize], SUBSTR)

            self.assertEquals(s.strip(), SUBSTR.strip())

            del s

    @bigmemtest(minsize=_2G, memuse=2)

    def test_count(self, size):

        SUBSTR = ' abc def ghi'

        s = '.' * size + SUBSTR

        self.assertEquals(s.count('.'), size)

        s += '.'

        self.assertEquals(s.count('.'), size + 1)

        self.assertEquals(s.count(' '), 3)

        self.assertEquals(s.count('i'), 1)

        self.assertEquals(s.count('j'), 0)

    @bigmemtest(minsize=_2G + 2, memuse=3)

    def test_decode(self, size):

        s = '.' * size

        self.assertEquals(len(s.decode('utf-8')), size)

    def basic_encode_test(self, size, enc, c=u'.', expectedsize=None):

        if expectedsize is None:

            expectedsize = size

        s = c * size

        self.assertEquals(len(s.encode(enc)), expectedsize)

    @bigmemtest(minsize=_2G + 2, memuse=3)

    def test_encode(self, size):

        return self.basic_encode_test(size, 'utf-8')

    @precisionbigmemtest(size=_4G / 6 + 2, memuse=2)

    def test_encode_raw_unicode_escape(self, size):

        try:

            return self.basic_encode_test(size, 'raw_unicode_escape')

        except MemoryError:

            pass # acceptable on 32-bit

    @precisionbigmemtest(size=_4G / 5 + 70, memuse=3)

    def test_encode_utf7(self, size):

        try:

            return self.basic_encode_test(size, 'utf7')

        except MemoryError:

            pass # acceptable on 32-bit

    @precisionbigmemtest(size=_4G / 4 + 5, memuse=6)

    def test_encode_utf32(self, size):

        try:

            return self.basic_encode_test(size, 'utf32', expectedsize=4*size+4)

        except MemoryError:

            pass # acceptable on 32-bit

    @precisionbigmemtest(size=_2G-1, memuse=2)

    def test_decodeascii(self, size):

        return self.basic_encode_test(size, 'ascii', c='A')

    @precisionbigmemtest(size=_4G / 5, memuse=6+2)

    def test_unicode_repr_oflw(self, size):

        try:

            s = u"\uAAAA"*size

            r = repr(s)

        except MemoryError:

            pass # acceptable on 32-bit

        else:

            self.failUnless(s == eval(r))

    @bigmemtest(minsize=_2G, memuse=2)

    def test_endswith(self, size):

        SUBSTR = ' abc def ghi'

        s = '-' * size + SUBSTR

        self.failUnless(s.endswith(SUBSTR))

        self.failUnless(s.endswith(s))

        s2 = '...' + s

        self.failUnless(s2.endswith(s))

        self.failIf(s.endswith('a' + SUBSTR))

        self.failIf(SUBSTR.endswith(s))

    @bigmemtest(minsize=_2G + 10, memuse=2)

    def test_expandtabs(self, size):

        s = '-' * size

        tabsize = 8

        self.assertEquals(s.expandtabs(), s)

        del s

        slen, remainder = divmod(size, tabsize)

        s = '       \t' * slen

        s = s.expandtabs(tabsize)

        self.assertEquals(len(s), size - remainder)

        self.assertEquals(len(s.strip(' ')), 0)

    @bigmemtest(minsize=_2G, memuse=2)

    def test_find(self, size):

        SUBSTR = ' abc def ghi'

        sublen = len(SUBSTR)

        s = ''.join([SUBSTR, '-' * size, SUBSTR])

        self.assertEquals(s.find(' '), 0)

        self.assertEquals(s.find(SUBSTR), 0)

        self.assertEquals(s.find(' ', sublen), sublen + size)

        self.assertEquals(s.find(SUBSTR, len(SUBSTR)), sublen + size)

        self.assertEquals(s.find('i'), SUBSTR.find('i'))

        self.assertEquals(s.find('i', sublen),

                         sublen + size + SUBSTR.find('i'))

        self.assertEquals(s.find('i', size),

                         sublen + size + SUBSTR.find('i'))

        self.assertEquals(s.find('j'), -1)

    @bigmemtest(minsize=_2G, memuse=2)

    def test_index(self, size):

        SUBSTR = ' abc def ghi'

        sublen = len(SUBSTR)

        s = ''.join([SUBSTR, '-' * size, SUBSTR])

        self.assertEquals(s.index(' '), 0)

        self.assertEquals(s.index(SUBSTR), 0)

        self.assertEquals(s.index(' ', sublen), sublen + size)

        self.assertEquals(s.index(SUBSTR, sublen), sublen + size)

        self.assertEquals(s.index('i'), SUBSTR.index('i'))

        self.assertEquals(s.index('i', sublen),

                         sublen + size + SUBSTR.index('i'))

        self.assertEquals(s.index('i', size),

                         sublen + size + SUBSTR.index('i'))

        self.assertRaises(ValueError, s.index, 'j')

    @bigmemtest(minsize=_2G, memuse=2)

    def test_isalnum(self, size):

        SUBSTR = '123456'

        s = 'a' * size + SUBSTR

        self.failUnless(s.isalnum())

        s += '.'

        self.failIf(s.isalnum())

    @bigmemtest(minsize=_2G, memuse=2)

    def test_isalpha(self, size):

        SUBSTR = 'zzzzzzz'

        s = 'a' * size + SUBSTR

        self.failUnless(s.isalpha())

        s += '.'

        self.failIf(s.isalpha())

    @bigmemtest(minsize=_2G, memuse=2)

    def test_isdigit(self, size):

        SUBSTR = '123456'

        s = '9' * size + SUBSTR

        self.failUnless(s.isdigit())

        s += 'z'

        self.failIf(s.isdigit())

    @bigmemtest(minsize=_2G, memuse=2)

    def test_islower(self, size):

        chars = ''.join([ chr(c) for c in range(255) if not chr(c).isupper() ])

        repeats = size // len(chars) + 2

        s = chars * repeats

        self.failUnless(s.islower())

        s += 'A'

        self.failIf(s.islower())

    @bigmemtest(minsize=_2G, memuse=2)

    def test_isspace(self, size):

        whitespace = ' \f\n\r\t\v'

        repeats = size // len(whitespace) + 2

        s = whitespace * repeats

        self.failUnless(s.isspace())

        s += 'j'

        self.failIf(s.isspace())

    @bigmemtest(minsize=_2G, memuse=2)

    def test_istitle(self, size):

        SUBSTR = '123456'

        s = ''.join(['A', 'a' * size, SUBSTR])

        self.failUnless(s.istitle())

        s += 'A'

        self.failUnless(s.istitle())

        s += 'aA'

        self.failIf(s.istitle())

    @bigmemtest(minsize=_2G, memuse=2)

    def test_isupper(self, size):

        chars = ''.join([ chr(c) for c in range(255) if not chr(c).islower() ])

        repeats = size // len(chars) + 2

        s = chars * repeats

        self.failUnless(s.isupper())

        s += 'a'

        self.failIf(s.isupper())

    @bigmemtest(minsize=_2G, memuse=2)

    def test_join(self, size):

        s = 'A' * size

        x = s.join(['aaaaa', 'bbbbb'])

        self.assertEquals(x.count('a'), 5)

        self.assertEquals(x.count('b'), 5)

        self.failUnless(x.startswith('aaaaaA'))

        self.failUnless(x.endswith('Abbbbb'))

    @bigmemtest(minsize=_2G + 10, memuse=1)

    def test_ljust(self, size):

        SUBSTR = ' abc def ghi'

        s = SUBSTR.ljust(size)

        self.failUnless(s.startswith(SUBSTR + '  '))

        self.assertEquals(len(s), size)

        self.assertEquals(s.strip(), SUBSTR.strip())

    @bigmemtest(minsize=_2G + 10, memuse=2)

    def test_lower(self, size):

        s = 'A' * size

        s = s.lower()

        self.assertEquals(len(s), size)

        self.assertEquals(s.count('a'), size)

    @bigmemtest(minsize=_2G + 10, memuse=1)

    def test_lstrip(self, size):

        SUBSTR = 'abc def ghi'

        s = SUBSTR.rjust(size)

        self.assertEquals(len(s), size)

        self.assertEquals(s.lstrip(), SUBSTR.lstrip())

        del s

        s = SUBSTR.ljust(size)

        self.assertEquals(len(s), size)

        stripped = s.lstrip()

        self.failUnless(stripped is s)

    @bigmemtest(minsize=_2G + 10, memuse=2)

    def test_replace(self, size):

        replacement = 'a'

        s = ' ' * size

        s = s.replace(' ', replacement)

        self.assertEquals(len(s), size)

        self.assertEquals(s.count(replacement), size)

        s = s.replace(replacement, ' ', size - 4)

        self.assertEquals(len(s), size)

        self.assertEquals(s.count(replacement), 4)

        self.assertEquals(s[-10:], '      aaaa')

    @bigmemtest(minsize=_2G, memuse=2)

    def test_rfind(self, size):

        SUBSTR = ' abc def ghi'

        sublen = len(SUBSTR)

        s = ''.join([SUBSTR, '-' * size, SUBSTR])

        self.assertEquals(s.rfind(' '), sublen + size + SUBSTR.rfind(' '))

        self.assertEquals(s.rfind(SUBSTR), sublen + size)

        self.assertEquals(s.rfind(' ', 0, size), SUBSTR.rfind(' '))

        self.assertEquals(s.rfind(SUBSTR, 0, sublen + size), 0)

        self.assertEquals(s.rfind('i'), sublen + size + SUBSTR.rfind('i'))

        self.assertEquals(s.rfind('i', 0, sublen), SUBSTR.rfind('i'))

        self.assertEquals(s.rfind('i', 0, sublen + size),

                          SUBSTR.rfind('i'))

        self.assertEquals(s.rfind('j'), -1)

    @bigmemtest(minsize=_2G, memuse=2)

    def test_rindex(self, size):

        SUBSTR = ' abc def ghi'

        sublen = len(SUBSTR)

        s = ''.join([SUBSTR, '-' * size, SUBSTR])

        self.assertEquals(s.rindex(' '),

                          sublen + size + SUBSTR.rindex(' '))

        self.assertEquals(s.rindex(SUBSTR), sublen + size)

        self.assertEquals(s.rindex(' ', 0, sublen + size - 1),

                          SUBSTR.rindex(' '))

        self.assertEquals(s.rindex(SUBSTR, 0, sublen + size), 0)

        self.assertEquals(s.rindex('i'),

                          sublen + size + SUBSTR.rindex('i'))

        self.assertEquals(s.rindex('i', 0, sublen), SUBSTR.rindex('i'))

        self.assertEquals(s.rindex('i', 0, sublen + size),

                          SUBSTR.rindex('i'))

        self.assertRaises(ValueError, s.rindex, 'j')

    @bigmemtest(minsize=_2G + 10, memuse=1)

    def test_rjust(self, size):

        SUBSTR = ' abc def ghi'

        s = SUBSTR.ljust(size)

        self.failUnless(s.startswith(SUBSTR + '  '))

        self.assertEquals(len(s), size)

        self.assertEquals(s.strip(), SUBSTR.strip())

    @bigmemtest(minsize=_2G + 10, memuse=1)

    def test_rstrip(self, size):

        SUBSTR = ' abc def ghi'

        s = SUBSTR.ljust(size)

        self.assertEquals(len(s), size)

        self.assertEquals(s.rstrip(), SUBSTR.rstrip())

        del s

        s = SUBSTR.rjust(size)

        self.assertEquals(len(s), size)

        stripped = s.rstrip()

        self.failUnless(stripped is s)

    # The test takes about size bytes to build a string, and then about

    # sqrt(size) substrings of sqrt(size) in size and a list to

    # hold sqrt(size) items. It's close but just over 2x size.

    @bigmemtest(minsize=_2G, memuse=2.1)

    def test_split_small(self, size):

        # Crudely calculate an estimate so that the result of s.split won't

        # take up an inordinate amount of memory

        chunksize = int(size ** 0.5 + 2)

        SUBSTR = 'a' + ' ' * chunksize

        s = SUBSTR * chunksize

        l = s.split()

        self.assertEquals(len(l), chunksize)

        self.assertEquals(set(l), set(['a']))

        del l

        l = s.split('a')

        self.assertEquals(len(l), chunksize + 1)

        self.assertEquals(set(l), set(['', ' ' * chunksize]))

    # Allocates a string of twice size (and briefly two) and a list of

    # size.  Because of internal affairs, the s.split() call produces a

    # list of size times the same one-character string, so we only

    # suffer for the list size. (Otherwise, it'd cost another 48 times

    # size in bytes!) Nevertheless, a list of size takes

    # 8*size bytes.

    @bigmemtest(minsize=_2G + 5, memuse=10)

    def test_split_large(self, size):

        s = ' a' * size + ' '

        l = s.split()

        self.assertEquals(len(l), size)

        self.assertEquals(set(l), set(['a']))

        del l

        l = s.split('a')

        self.assertEquals(len(l), size + 1)

        self.assertEquals(set(l), set([' ']))

    @bigmemtest(minsize=_2G, memuse=2.1)

    def test_splitlines(self, size):

        # Crudely calculate an estimate so that the result of s.split won't

        # take up an inordinate amount of memory

        chunksize = int(size ** 0.5 + 2) // 2

        SUBSTR = ' ' * chunksize + '\n' + ' ' * chunksize + '\r\n'

        s = SUBSTR * chunksize

        l = s.splitlines()

        self.assertEquals(len(l), chunksize * 2)

        self.assertEquals(set(l), set([' ' * chunksize]))

    @bigmemtest(minsize=_2G, memuse=2)

    def test_startswith(self, size):

        SUBSTR = ' abc def ghi'

        s = '-' * size + SUBSTR

        self.failUnless(s.startswith(s))

        self.failUnless(s.startswith('-' * size))

        self.failIf(s.startswith(SUBSTR))

    @bigmemtest(minsize=_2G, memuse=1)

    def test_strip(self, size):

        SUBSTR = '   abc def ghi   '

        s = SUBSTR.rjust(size)

        self.assertEquals(len(s), size)

        self.assertEquals(s.strip(), SUBSTR.strip())

        del s

        s = SUBSTR.ljust(size)

        self.assertEquals(len(s), size)

        self.assertEquals(s.strip(), SUBSTR.strip())

    @bigmemtest(minsize=_2G, memuse=2)

    def test_swapcase(self, size):

        SUBSTR = "aBcDeFG12.'\xa9\x00"

        sublen = len(SUBSTR)

        repeats = size // sublen + 2

        s = SUBSTR * repeats

        s = s.swapcase()

        self.assertEquals(len(s), sublen * repeats)

        self.assertEquals(s[:sublen * 3], SUBSTR.swapcase() * 3)

        self.assertEquals(s[-sublen * 3:], SUBSTR.swapcase() * 3)

    @bigmemtest(minsize=_2G, memuse=2)

    def test_title(self, size):

        SUBSTR = 'SpaaHAaaAaham'

        s = SUBSTR * (size // len(SUBSTR) + 2)

        s = s.title()

        self.failUnless(s.startswith((SUBSTR * 3).title()))

        self.failUnless(s.endswith(SUBSTR.lower() * 3))

    @bigmemtest(minsize=_2G, memuse=2)

    def test_translate(self, size):

        trans = string.maketrans('.aZ', '-!$')

        SUBSTR = 'aZz.z.Aaz.'

        sublen = len(SUBSTR)

        repeats = size // sublen + 2

        s = SUBSTR * repeats

        s = s.translate(trans)

        self.assertEquals(len(s), repeats * sublen)

        self.assertEquals(s[:sublen], SUBSTR.translate(trans))

        self.assertEquals(s[-sublen:], SUBSTR.translate(trans))

        self.assertEquals(s.count('.'), 0)

        self.assertEquals(s.count('!'), repeats * 2)

        self.assertEquals(s.count('z'), repeats * 3)

    @bigmemtest(minsize=_2G + 5, memuse=2)

    def test_upper(self, size):

        s = 'a' * size

        s = s.upper()

        self.assertEquals(len(s), size)

        self.assertEquals(s.count('A'), size)

    @bigmemtest(minsize=_2G + 20, memuse=1)

    def test_zfill(self, size):

        SUBSTR = '-568324723598234'

        s = SUBSTR.zfill(size)

        self.failUnless(s.endswith('0' + SUBSTR[1:]))

        self.failUnless(s.startswith('-0'))

        self.assertEquals(len(s), size)

        self.assertEquals(s.count('0'), size - len(SUBSTR))

    @bigmemtest(minsize=_2G + 10, memuse=2)

    def test_format(self, size):

        s = '-' * size

        sf = '%s' % (s,)

        self.failUnless(s == sf)

        del sf

        sf = '..%s..' % (s,)

        self.assertEquals(len(sf), len(s) + 4)

        self.failUnless(sf.startswith('..-'))

        self.failUnless(sf.endswith('-..'))

        del s, sf

        size //= 2

        edge = '-' * size

        s = ''.join([edge, '%s', edge])

        del edge

        s = s % '...'

        self.assertEquals(len(s), size * 2 + 3)

        self.assertEquals(s.count('.'), 3)

        self.assertEquals(s.count('-'), size * 2)

    @bigmemtest(minsize=_2G + 10, memuse=2)

    def test_repr_small(self, size):

        s = '-' * size

        s = repr(s)

        self.assertEquals(len(s), size + 2)

        self.assertEquals(s[0], "'")

        self.assertEquals(s[-1], "'")

        self.assertEquals(s.count('-'), size)

        del s

        # repr() will create a string four times as large as this 'binary

        # string', but we don't want to allocate much more than twice

        # size in total.  (We do extra testing in test_repr_large())

        size = size // 5 * 2

        s = '\x00' * size

        s = repr(s)

        self.assertEquals(len(s), size * 4 + 2)

        self.assertEquals(s[0], "'")

        self.assertEquals(s[-1], "'")

        self.assertEquals(s.count('\\'), size)

        self.assertEquals(s.count('0'), size * 2)

    @bigmemtest(minsize=_2G + 10, memuse=5)

    def test_repr_large(self, size):

        s = '\x00' * size

        s = repr(s)

        self.assertEquals(len(s), size * 4 + 2)

        self.assertEquals(s[0], "'")

        self.assertEquals(s[-1], "'")

        self.assertEquals(s.count('\\'), size)

        self.assertEquals(s.count('0'), size * 2)

    @bigmemtest(minsize=2**32 / 5, memuse=6+2)

    def test_unicode_repr(self, size):

        s = u"\uAAAA" * size

        self.failUnless(len(repr(s)) > size)

    # This test is meaningful even with size < 2G, as long as the

    # doubled string is > 2G (but it tests more if both are > 2G :)

    @bigmemtest(minsize=_1G + 2, memuse=3)

    def test_concat(self, size):

        s = '.' * size