from __future__ import with_statement

from test import test_support

import unittest

import codecs

import sys, StringIO, _testcapi

class Queue(object):

    """

    queue: write bytes at one end, read bytes from the other end

    """

    def __init__(self):

        self._buffer = ""

    def write(self, chars):

        self._buffer += chars

    def read(self, size=-1):

        if size<0:

            s = self._buffer

            self._buffer = ""

            return s

        else:

            s = self._buffer[:size]

            self._buffer = self._buffer[size:]

            return s

class ReadTest(unittest.TestCase):

    def check_partial(self, input, partialresults):

        # get a StreamReader for the encoding and feed the bytestring version

        # of input to the reader byte by byte. Read everything available from

        # the StreamReader and check that the results equal the appropriate

        # entries from partialresults.

        q = Queue()

        r = codecs.getreader(self.encoding)(q)

        result = u""

        for (c, partialresult) in zip(input.encode(self.encoding), partialresults):

            q.write(c)

            result += r.read()

            self.assertEqual(result, partialresult)

        # check that there's nothing left in the buffers

        self.assertEqual(r.read(), u"")

        self.assertEqual(r.bytebuffer, "")

        self.assertEqual(r.charbuffer, u"")

        # do the check again, this time using a incremental decoder

        d = codecs.getincrementaldecoder(self.encoding)()

        result = u""

        for (c, partialresult) in zip(input.encode(self.encoding), partialresults):

            result += d.decode(c)

            self.assertEqual(result, partialresult)

        # check that there's nothing left in the buffers

        self.assertEqual(d.decode("", True), u"")

        self.assertEqual(d.buffer, "")

        # Check whether the rest method works properly

        d.reset()

        result = u""

        for (c, partialresult) in zip(input.encode(self.encoding), partialresults):

            result += d.decode(c)

            self.assertEqual(result, partialresult)

        # check that there's nothing left in the buffers

        self.assertEqual(d.decode("", True), u"")

        self.assertEqual(d.buffer, "")

        # check iterdecode()

        encoded = input.encode(self.encoding)

        self.assertEqual(

            input,

            u"".join(codecs.iterdecode(encoded, self.encoding))

        )

    def test_readline(self):

        def getreader(input):

            stream = StringIO.StringIO(input.encode(self.encoding))

            return codecs.getreader(self.encoding)(stream)

        def readalllines(input, keepends=True, size=None):

            reader = getreader(input)

            lines = []

            while True:

                line = reader.readline(size=size, keepends=keepends)

                if not line:

                    break

                lines.append(line)

            return "|".join(lines)

        s = u"foo\nbar\r\nbaz\rspam\u2028eggs"

        sexpected = u"foo\n|bar\r\n|baz\r|spam\u2028|eggs"

        sexpectednoends = u"foo|bar|baz|spam|eggs"

        self.assertEqual(readalllines(s, True), sexpected)

        self.assertEqual(readalllines(s, False), sexpectednoends)

        self.assertEqual(readalllines(s, True, 10), sexpected)

        self.assertEqual(readalllines(s, False, 10), sexpectednoends)

        # Test long lines (multiple calls to read() in readline())

        vw = []

        vwo = []

        for (i, lineend) in enumerate(u"\n \r\n \r \u2028".split()):

            vw.append((i*200)*u"\3042" + lineend)

            vwo.append((i*200)*u"\3042")

        self.assertEqual(readalllines("".join(vw), True), "".join(vw))

        self.assertEqual(readalllines("".join(vw), False),"".join(vwo))

        # Test lines where the first read might end with \r, so the

        # reader has to look ahead whether this is a lone \r or a \r\n

        for size in xrange(80):

            for lineend in u"\n \r\n \r \u2028".split():

                s = 10*(size*u"a" + lineend + u"xxx\n")

                reader = getreader(s)

                for i in xrange(10):

                    self.assertEqual(

                        reader.readline(keepends=True),

                        size*u"a" + lineend,

                    )

                reader = getreader(s)

                for i in xrange(10):

                    self.assertEqual(

                        reader.readline(keepends=False),

                        size*u"a",

                    )

    def test_bug1175396(self):

        s = [

            '<%!--===================================================\r\n',

            '    BLOG index page: show recent articles,\r\n',

            '    today\'s articles, or articles of a specific date.\r\n',

            '========================================================--%>\r\n',

            '<%@inputencoding="ISO-8859-1"%>\r\n',

            '<%@pagetemplate=TEMPLATE.y%>\r\n',

            '<%@import=import frog.util, frog%>\r\n',

            '<%@import=import frog.objects%>\r\n',

            '<%@import=from frog.storageerrors import StorageError%>\r\n',

            '<%\r\n',

            '\r\n',

            'import logging\r\n',

            'log=logging.getLogger("Snakelets.logger")\r\n',

            '\r\n',

            '\r\n',

            'user=self.SessionCtx.user\r\n',

            'storageEngine=self.SessionCtx.storageEngine\r\n',

            '\r\n',

            '\r\n',

            'def readArticlesFromDate(date, count=None):\r\n',

            '    entryids=storageEngine.listBlogEntries(date)\r\n',

            '    entryids.reverse() # descending\r\n',

            '    if count:\r\n',

            '        entryids=entryids[:count]\r\n',

            '    try:\r\n',

            '        return [ frog.objects.BlogEntry.load(storageEngine, date, Id) for Id in entryids ]\r\n',

            '    except StorageError,x:\r\n',

            '        log.error("Error loading articles: "+str(x))\r\n',

            '        self.abort("cannot load articles")\r\n',

            '\r\n',

            'showdate=None\r\n',

            '\r\n',

            'arg=self.Request.getArg()\r\n',

            'if arg=="today":\r\n',

            '    #-------------------- TODAY\'S ARTICLES\r\n',

            '    self.write("<h2>Today\'s articles</h2>")\r\n',

            '    showdate = frog.util.isodatestr() \r\n',

            '    entries = readArticlesFromDate(showdate)\r\n',

            'elif arg=="active":\r\n',

            '    #-------------------- ACTIVE ARTICLES redirect\r\n',

            '    self.Yredirect("active.y")\r\n',

            'elif arg=="login":\r\n',

            '    #-------------------- LOGIN PAGE redirect\r\n',

            '    self.Yredirect("login.y")\r\n',

            'elif arg=="date":\r\n',

            '    #-------------------- ARTICLES OF A SPECIFIC DATE\r\n',

            '    showdate = self.Request.getParameter("date")\r\n',

            '    self.write("<h2>Articles written on %s</h2>"% frog.util.mediumdatestr(showdate))\r\n',

            '    entries = readArticlesFromDate(showdate)\r\n',

            'else:\r\n',

            '    #-------------------- RECENT ARTICLES\r\n',

            '    self.write("<h2>Recent articles</h2>")\r\n',

            '    dates=storageEngine.listBlogEntryDates()\r\n',

            '    if dates:\r\n',

            '        entries=[]\r\n',

            '        SHOWAMOUNT=10\r\n',

            '        for showdate in dates:\r\n',

            '            entries.extend( readArticlesFromDate(showdate, SHOWAMOUNT-len(entries)) )\r\n',

            '            if len(entries)>=SHOWAMOUNT:\r\n',

            '                break\r\n',

            '                \r\n',

        ]

        stream = StringIO.StringIO("".join(s).encode(self.encoding))

        reader = codecs.getreader(self.encoding)(stream)

        for (i, line) in enumerate(reader):

            self.assertEqual(line, s[i])

    def test_readlinequeue(self):

        q = Queue()

        writer = codecs.getwriter(self.encoding)(q)

        reader = codecs.getreader(self.encoding)(q)

        # No lineends

        writer.write(u"foo\r")

        self.assertEqual(reader.readline(keepends=False), u"foo")

        writer.write(u"\nbar\r")

        self.assertEqual(reader.readline(keepends=False), u"")

        self.assertEqual(reader.readline(keepends=False), u"bar")

        writer.write(u"baz")

        self.assertEqual(reader.readline(keepends=False), u"baz")

        self.assertEqual(reader.readline(keepends=False), u"")

        # Lineends

        writer.write(u"foo\r")

        self.assertEqual(reader.readline(keepends=True), u"foo\r")

        writer.write(u"\nbar\r")

        self.assertEqual(reader.readline(keepends=True), u"\n")

        self.assertEqual(reader.readline(keepends=True), u"bar\r")

        writer.write(u"baz")

        self.assertEqual(reader.readline(keepends=True), u"baz")

        self.assertEqual(reader.readline(keepends=True), u"")

        writer.write(u"foo\r\n")

        self.assertEqual(reader.readline(keepends=True), u"foo\r\n")

    def test_bug1098990_a(self):

        s1 = u"xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy\r\n"

        s2 = u"offending line: ladfj askldfj klasdj fskla dfzaskdj fasklfj laskd fjasklfzzzzaa%whereisthis!!!\r\n"

        s3 = u"next line.\r\n"

        s = (s1+s2+s3).encode(self.encoding)

        stream = StringIO.StringIO(s)

        reader = codecs.getreader(self.encoding)(stream)

        self.assertEqual(reader.readline(), s1)

        self.assertEqual(reader.readline(), s2)

        self.assertEqual(reader.readline(), s3)

        self.assertEqual(reader.readline(), u"")

    def test_bug1098990_b(self):

        s1 = u"aaaaaaaaaaaaaaaaaaaaaaaa\r\n"

        s2 = u"bbbbbbbbbbbbbbbbbbbbbbbb\r\n"

        s3 = u"stillokay:bbbbxx\r\n"

        s4 = u"broken!!!!badbad\r\n"

        s5 = u"againokay.\r\n"

        s = (s1+s2+s3+s4+s5).encode(self.encoding)

        stream = StringIO.StringIO(s)

        reader = codecs.getreader(self.encoding)(stream)

        self.assertEqual(reader.readline(), s1)

        self.assertEqual(reader.readline(), s2)

        self.assertEqual(reader.readline(), s3)

        self.assertEqual(reader.readline(), s4)

        self.assertEqual(reader.readline(), s5)

        self.assertEqual(reader.readline(), u"")

class UTF16Test(ReadTest):

    encoding = "utf-16"

    spamle = '\xff\xfes\x00p\x00a\x00m\x00s\x00p\x00a\x00m\x00'

    spambe = '\xfe\xff\x00s\x00p\x00a\x00m\x00s\x00p\x00a\x00m'

    def test_only_one_bom(self):

        _,_,reader,writer = codecs.lookup(self.encoding)

        # encode some stream

        s = StringIO.StringIO()

        f = writer(s)

        f.write(u"spam")

        f.write(u"spam")

        d = s.getvalue()

        # check whether there is exactly one BOM in it

        self.assert_(d == self.spamle or d == self.spambe)

        # try to read it back

        s = StringIO.StringIO(d)

        f = reader(s)

        self.assertEquals(f.read(), u"spamspam")

    def test_badbom(self):

        s = StringIO.StringIO("\xff\xff")

        f = codecs.getreader(self.encoding)(s)

        self.assertRaises(UnicodeError, f.read)

        s = StringIO.StringIO("\xff\xff\xff\xff")

        f = codecs.getreader(self.encoding)(s)

        self.assertRaises(UnicodeError, f.read)

    def test_partial(self):

        self.check_partial(

            u"\x00\xff\u0100\uffff",

            [

                u"", # first byte of BOM read

                u"", # second byte of BOM read => byteorder known

                u"",

                u"\x00",

                u"\x00",

                u"\x00\xff",

                u"\x00\xff",

                u"\x00\xff\u0100",

                u"\x00\xff\u0100",

                u"\x00\xff\u0100\uffff",

            ]

        )

    def test_errors(self):

        self.assertRaises(UnicodeDecodeError, codecs.utf_16_decode, "\xff", "strict", True)

class UTF16LETest(ReadTest):

    encoding = "utf-16-le"

    def test_partial(self):

        self.check_partial(

            u"\x00\xff\u0100\uffff",

            [

                u"",

                u"\x00",

                u"\x00",

                u"\x00\xff",

                u"\x00\xff",

                u"\x00\xff\u0100",

                u"\x00\xff\u0100",

                u"\x00\xff\u0100\uffff",

            ]

        )

    def test_errors(self):

        self.assertRaises(UnicodeDecodeError, codecs.utf_16_le_decode, "\xff", "strict", True)

class UTF16BETest(ReadTest):

    encoding = "utf-16-be"

    def test_partial(self):

        self.check_partial(

            u"\x00\xff\u0100\uffff",

            [

                u"",

                u"\x00",

                u"\x00",

                u"\x00\xff",

                u"\x00\xff",

                u"\x00\xff\u0100",

                u"\x00\xff\u0100",

                u"\x00\xff\u0100\uffff",

            ]

        )

    def test_errors(self):

        self.assertRaises(UnicodeDecodeError, codecs.utf_16_be_decode, "\xff", "strict", True)

class UTF8Test(ReadTest):

    encoding = "utf-8"

    def test_partial(self):

        self.check_partial(

            u"\x00\xff\u07ff\u0800\uffff",

            [

                u"\x00",

                u"\x00",

                u"\x00\xff",

                u"\x00\xff",

                u"\x00\xff\u07ff",

                u"\x00\xff\u07ff",

                u"\x00\xff\u07ff",

                u"\x00\xff\u07ff\u0800",

                u"\x00\xff\u07ff\u0800",

                u"\x00\xff\u07ff\u0800",

                u"\x00\xff\u07ff\u0800\uffff",

            ]

        )

class UTF7Test(ReadTest):

    encoding = "utf-7"

    # No test_partial() yet, because UTF-7 doesn't support it.

class UTF16ExTest(unittest.TestCase):

    def test_errors(self):

        self.assertRaises(UnicodeDecodeError, codecs.utf_16_ex_decode, "\xff", "strict", 0, True)

    def test_bad_args(self):

        self.assertRaises(TypeError, codecs.utf_16_ex_decode)

class ReadBufferTest(unittest.TestCase):

    def test_array(self):

        import array

        self.assertEqual(

            codecs.readbuffer_encode(array.array("c", "spam")),

            ("spam", 4)

        )

    def test_empty(self):

        self.assertEqual(codecs.readbuffer_encode(""), ("", 0))

    def test_bad_args(self):

        self.assertRaises(TypeError, codecs.readbuffer_encode)

        self.assertRaises(TypeError, codecs.readbuffer_encode, 42)

class CharBufferTest(unittest.TestCase):

    def test_string(self):

        self.assertEqual(codecs.charbuffer_encode("spam"), ("spam", 4))

    def test_empty(self):

        self.assertEqual(codecs.charbuffer_encode(""), ("", 0))

    def test_bad_args(self):

        self.assertRaises(TypeError, codecs.charbuffer_encode)

        self.assertRaises(TypeError, codecs.charbuffer_encode, 42)

class UTF8SigTest(ReadTest):

    encoding = "utf-8-sig"

    def test_partial(self):

        self.check_partial(

            u"\ufeff\x00\xff\u07ff\u0800\uffff",

            [

                u"",

                u"",

                u"", # First BOM has been read and skipped

                u"",

                u"",

                u"\ufeff", # Second BOM has been read and emitted

                u"\ufeff\x00", # "\x00" read and emitted

                u"\ufeff\x00", # First byte of encoded u"\xff" read

                u"\ufeff\x00\xff", # Second byte of encoded u"\xff" read

                u"\ufeff\x00\xff", # First byte of encoded u"\u07ff" read

                u"\ufeff\x00\xff\u07ff", # Second byte of encoded u"\u07ff" read

                u"\ufeff\x00\xff\u07ff",

                u"\ufeff\x00\xff\u07ff",

                u"\ufeff\x00\xff\u07ff\u0800",

                u"\ufeff\x00\xff\u07ff\u0800",

                u"\ufeff\x00\xff\u07ff\u0800",

                u"\ufeff\x00\xff\u07ff\u0800\uffff",

            ]

        )

    def test_bug1601501(self):

        # SF bug #1601501: check that the codec works with a buffer

        unicode("\xef\xbb\xbf", "utf-8-sig")

    def test_bom(self):

        d = codecs.getincrementaldecoder("utf-8-sig")()

        s = u"spam"

        self.assertEqual(d.decode(s.encode("utf-8-sig")), s)

    def test_stream_bom(self):

        unistring = u"ABC\u00A1\u2200XYZ"

        bytestring = codecs.BOM_UTF8 + "ABC\xC2\xA1\xE2\x88\x80XYZ"

        reader = codecs.getreader("utf-8-sig")

        for sizehint in [None] + range(1, 11) + \

                        [64, 128, 256, 512, 1024]:

            istream = reader(StringIO.StringIO(bytestring))

            ostream = StringIO.StringIO()

            while 1:

                if sizehint is not None:

                    data = istream.read(sizehint)

                else:

                    data = istream.read()

                if not data:

                    break

                ostream.write(data)

            got = ostream.getvalue()

            self.assertEqual(got, unistring)

    def test_stream_bare(self):

        unistring = u"ABC\u00A1\u2200XYZ"

        bytestring = "ABC\xC2\xA1\xE2\x88\x80XYZ"

        reader = codecs.getreader("utf-8-sig")

        for sizehint in [None] + range(1, 11) + \

                        [64, 128, 256, 512, 1024]:

            istream = reader(StringIO.StringIO(bytestring))

            ostream = StringIO.StringIO()

            while 1:

                if sizehint is not None:

                    data = istream.read(sizehint)

                else:

                    data = istream.read()

                if not data:

                    break

                ostream.write(data)

            got = ostream.getvalue()

            self.assertEqual(got, unistring)

class EscapeDecodeTest(unittest.TestCase):

    def test_empty(self):

        self.assertEquals(codecs.escape_decode(""), ("", 0))

class RecodingTest(unittest.TestCase):

    def test_recoding(self):

        f = StringIO.StringIO()

        f2 = codecs.EncodedFile(f, "unicode_internal", "utf-8")

        f2.write(u"a")

        f2.close()

        # Python used to crash on this at exit because of a refcount

        # bug in _codecsmodule.c

# From RFC 3492

punycode_testcases = [

    # A Arabic (Egyptian):

    (u"\u0644\u064A\u0647\u0645\u0627\u0628\u062A\u0643\u0644"

     u"\u0645\u0648\u0634\u0639\u0631\u0628\u064A\u061F",

     "egbpdaj6bu4bxfgehfvwxn"),

    # B Chinese (simplified):

    (u"\u4ED6\u4EEC\u4E3A\u4EC0\u4E48\u4E0D\u8BF4\u4E2D\u6587",

     "ihqwcrb4cv8a8dqg056pqjye"),

    # C Chinese (traditional):

    (u"\u4ED6\u5011\u7232\u4EC0\u9EBD\u4E0D\u8AAA\u4E2D\u6587",

     "ihqwctvzc91f659drss3x8bo0yb"),

    # D Czech: Pro<ccaron>prost<ecaron>nemluv<iacute><ccaron>esky

    (u"\u0050\u0072\u006F\u010D\u0070\u0072\u006F\u0073\u0074"

     u"\u011B\u006E\u0065\u006D\u006C\u0075\u0076\u00ED\u010D"

     u"\u0065\u0073\u006B\u0079",

     "Proprostnemluvesky-uyb24dma41a"),

    # E Hebrew:

    (u"\u05DC\u05DE\u05D4\u05D4\u05DD\u05E4\u05E9\u05D5\u05D8"

     u"\u05DC\u05D0\u05DE\u05D3\u05D1\u05E8\u05D9\u05DD\u05E2"

     u"\u05D1\u05E8\u05D9\u05EA",

     "4dbcagdahymbxekheh6e0a7fei0b"),

    # F Hindi (Devanagari):

    (u"\u092F\u0939\u0932\u094B\u0917\u0939\u093F\u0928\u094D"

    u"\u0926\u0940\u0915\u094D\u092F\u094B\u0902\u0928\u0939"

    u"\u0940\u0902\u092C\u094B\u0932\u0938\u0915\u0924\u0947"

    u"\u0939\u0948\u0902",

    "i1baa7eci9glrd9b2ae1bj0hfcgg6iyaf8o0a1dig0cd"),

    #(G) Japanese (kanji and hiragana):

    (u"\u306A\u305C\u307F\u3093\u306A\u65E5\u672C\u8A9E\u3092"

    u"\u8A71\u3057\u3066\u304F\u308C\u306A\u3044\u306E\u304B",

     "n8jok5ay5dzabd5bym9f0cm5685rrjetr6pdxa"),

    # (H) Korean (Hangul syllables):