# -*- coding: iso-8859-1 -*-

# Copyright (C) 2001,2002 Python Software Foundation

# csv package unit tests

import sys

import os

import unittest

from StringIO import StringIO

import tempfile

import csv

import gc

from test import test_support

class Test_Csv(unittest.TestCase):

    """

    Test the underlying C csv parser in ways that are not appropriate

    from the high level interface. Further tests of this nature are done

    in TestDialectRegistry.

    """

    def _test_arg_valid(self, ctor, arg):

        self.assertRaises(TypeError, ctor)

        self.assertRaises(TypeError, ctor, None)

        self.assertRaises(TypeError, ctor, arg, bad_attr = 0)

        self.assertRaises(TypeError, ctor, arg, delimiter = 0)

        self.assertRaises(TypeError, ctor, arg, delimiter = 'XX')

        self.assertRaises(csv.Error, ctor, arg, 'foo')

        self.assertRaises(TypeError, ctor, arg, delimiter=None)

        self.assertRaises(TypeError, ctor, arg, delimiter=1)

        self.assertRaises(TypeError, ctor, arg, quotechar=1)

        self.assertRaises(TypeError, ctor, arg, lineterminator=None)

        self.assertRaises(TypeError, ctor, arg, lineterminator=1)

        self.assertRaises(TypeError, ctor, arg, quoting=None)

        self.assertRaises(TypeError, ctor, arg,

                          quoting=csv.QUOTE_ALL, quotechar='')

        self.assertRaises(TypeError, ctor, arg,

                          quoting=csv.QUOTE_ALL, quotechar=None)

    def test_reader_arg_valid(self):

        self._test_arg_valid(csv.reader, [])

    def test_writer_arg_valid(self):

        self._test_arg_valid(csv.writer, StringIO())

    def _test_default_attrs(self, ctor, *args):

        obj = ctor(*args)

        # Check defaults

        self.assertEqual(obj.dialect.delimiter, ',')

        self.assertEqual(obj.dialect.doublequote, True)

        self.assertEqual(obj.dialect.escapechar, None)

        self.assertEqual(obj.dialect.lineterminator, "\r\n")

        self.assertEqual(obj.dialect.quotechar, '"')

        self.assertEqual(obj.dialect.quoting, csv.QUOTE_MINIMAL)

        self.assertEqual(obj.dialect.skipinitialspace, False)

        self.assertEqual(obj.dialect.strict, False)

        # Try deleting or changing attributes (they are read-only)

        self.assertRaises(TypeError, delattr, obj.dialect, 'delimiter')

        self.assertRaises(TypeError, setattr, obj.dialect, 'delimiter', ':')

        self.assertRaises(AttributeError, delattr, obj.dialect, 'quoting')

        self.assertRaises(AttributeError, setattr, obj.dialect,

                          'quoting', None)

    def test_reader_attrs(self):

        self._test_default_attrs(csv.reader, [])

    def test_writer_attrs(self):

        self._test_default_attrs(csv.writer, StringIO())

    def _test_kw_attrs(self, ctor, *args):

        # Now try with alternate options

        kwargs = dict(delimiter=':', doublequote=False, escapechar='\\',

                      lineterminator='\r', quotechar='*',

                      quoting=csv.QUOTE_NONE, skipinitialspace=True,

                      strict=True)

        obj = ctor(*args, **kwargs)

        self.assertEqual(obj.dialect.delimiter, ':')

        self.assertEqual(obj.dialect.doublequote, False)

        self.assertEqual(obj.dialect.escapechar, '\\')

        self.assertEqual(obj.dialect.lineterminator, "\r")

        self.assertEqual(obj.dialect.quotechar, '*')

        self.assertEqual(obj.dialect.quoting, csv.QUOTE_NONE)

        self.assertEqual(obj.dialect.skipinitialspace, True)

        self.assertEqual(obj.dialect.strict, True)

    def test_reader_kw_attrs(self):

        self._test_kw_attrs(csv.reader, [])

    def test_writer_kw_attrs(self):

        self._test_kw_attrs(csv.writer, StringIO())

    def _test_dialect_attrs(self, ctor, *args):

        # Now try with dialect-derived options

        class dialect:

            delimiter='-'

            doublequote=False

            escapechar='^'

            lineterminator='$'

            quotechar='#'

            quoting=csv.QUOTE_ALL

            skipinitialspace=True

            strict=False

        args = args + (dialect,)

        obj = ctor(*args)

        self.assertEqual(obj.dialect.delimiter, '-')

        self.assertEqual(obj.dialect.doublequote, False)

        self.assertEqual(obj.dialect.escapechar, '^')

        self.assertEqual(obj.dialect.lineterminator, "$")

        self.assertEqual(obj.dialect.quotechar, '#')

        self.assertEqual(obj.dialect.quoting, csv.QUOTE_ALL)

        self.assertEqual(obj.dialect.skipinitialspace, True)

        self.assertEqual(obj.dialect.strict, False)

    def test_reader_dialect_attrs(self):

        self._test_dialect_attrs(csv.reader, [])

    def test_writer_dialect_attrs(self):

        self._test_dialect_attrs(csv.writer, StringIO())

    def _write_test(self, fields, expect, **kwargs):

        fd, name = tempfile.mkstemp()

        fileobj = os.fdopen(fd, "w+b")

        try:

            writer = csv.writer(fileobj, **kwargs)

            writer.writerow(fields)

            fileobj.seek(0)

            self.assertEqual(fileobj.read(),

                             expect + writer.dialect.lineterminator)

        finally:

            fileobj.close()

            os.unlink(name)

    def test_write_arg_valid(self):

        self.assertRaises(csv.Error, self._write_test, None, '')

        self._write_test((), '')

        self._write_test([None], '""')

        self.assertRaises(csv.Error, self._write_test,

                          [None], None, quoting = csv.QUOTE_NONE)

        # Check that exceptions are passed up the chain

        class BadList:

            def __len__(self):

                return 10;

            def __getitem__(self, i):

                if i > 2:

                    raise IOError

        self.assertRaises(IOError, self._write_test, BadList(), '')

        class BadItem:

            def __str__(self):

                raise IOError

        self.assertRaises(IOError, self._write_test, [BadItem()], '')

    def test_write_bigfield(self):

        # This exercises the buffer realloc functionality

        bigstring = 'X' * 50000

        self._write_test([bigstring,bigstring], '%s,%s' % \

                         (bigstring, bigstring))

    def test_write_quoting(self):

        self._write_test(['a',1,'p,q'], 'a,1,"p,q"')

        self.assertRaises(csv.Error,

                          self._write_test,

                          ['a',1,'p,q'], 'a,1,p,q',

                          quoting = csv.QUOTE_NONE)

        self._write_test(['a',1,'p,q'], 'a,1,"p,q"',

                         quoting = csv.QUOTE_MINIMAL)

        self._write_test(['a',1,'p,q'], '"a",1,"p,q"',

                         quoting = csv.QUOTE_NONNUMERIC)

        self._write_test(['a',1,'p,q'], '"a","1","p,q"',

                         quoting = csv.QUOTE_ALL)

    def test_write_escape(self):

        self._write_test(['a',1,'p,q'], 'a,1,"p,q"',

                         escapechar='\\')

        self.assertRaises(csv.Error,

                          self._write_test,

                          ['a',1,'p,"q"'], 'a,1,"p,\\"q\\""',

                          escapechar=None, doublequote=False)

        self._write_test(['a',1,'p,"q"'], 'a,1,"p,\\"q\\""',

                         escapechar='\\', doublequote = False)

        self._write_test(['"'], '""""',

                         escapechar='\\', quoting = csv.QUOTE_MINIMAL)

        self._write_test(['"'], '\\"',

                         escapechar='\\', quoting = csv.QUOTE_MINIMAL,

                         doublequote = False)

        self._write_test(['"'], '\\"',

                         escapechar='\\', quoting = csv.QUOTE_NONE)

        self._write_test(['a',1,'p,q'], 'a,1,p\\,q',

                         escapechar='\\', quoting = csv.QUOTE_NONE)

    def test_writerows(self):

        class BrokenFile:

            def write(self, buf):

                raise IOError

        writer = csv.writer(BrokenFile())

        self.assertRaises(IOError, writer.writerows, [['a']])

        fd, name = tempfile.mkstemp()

        fileobj = os.fdopen(fd, "w+b")

        try:

            writer = csv.writer(fileobj)

            self.assertRaises(TypeError, writer.writerows, None)

            writer.writerows([['a','b'],['c','d']])

            fileobj.seek(0)

            self.assertEqual(fileobj.read(), "a,b\r\nc,d\r\n")

        finally:

            fileobj.close()

            os.unlink(name)

    def _read_test(self, input, expect, **kwargs):

        reader = csv.reader(input, **kwargs)

        result = list(reader)

        self.assertEqual(result, expect)

    def test_read_oddinputs(self):

        self._read_test([], [])

        self._read_test([''], [[]])

        self.assertRaises(csv.Error, self._read_test,

                          ['"ab"c'], None, strict = 1)

        # cannot handle null bytes for the moment

        self.assertRaises(csv.Error, self._read_test,

                          ['ab\0c'], None, strict = 1)

        self._read_test(['"ab"c'], [['abc']], doublequote = 0)

    def test_read_eol(self):

        self._read_test(['a,b'], [['a','b']])

        self._read_test(['a,b\n'], [['a','b']])

        self._read_test(['a,b\r\n'], [['a','b']])

        self._read_test(['a,b\r'], [['a','b']])

        self.assertRaises(csv.Error, self._read_test, ['a,b\rc,d'], [])

        self.assertRaises(csv.Error, self._read_test, ['a,b\nc,d'], [])

        self.assertRaises(csv.Error, self._read_test, ['a,b\r\nc,d'], [])

    def test_read_escape(self):

        self._read_test(['a,\\b,c'], [['a', 'b', 'c']], escapechar='\\')

        self._read_test(['a,b\\,c'], [['a', 'b,c']], escapechar='\\')

        self._read_test(['a,"b\\,c"'], [['a', 'b,c']], escapechar='\\')

        self._read_test(['a,"b,\\c"'], [['a', 'b,c']], escapechar='\\')

        self._read_test(['a,"b,c\\""'], [['a', 'b,c"']], escapechar='\\')

        self._read_test(['a,"b,c"\\'], [['a', 'b,c\\']], escapechar='\\')

    def test_read_quoting(self):

        self._read_test(['1,",3,",5'], [['1', ',3,', '5']])

        self._read_test(['1,",3,",5'], [['1', '"', '3', '"', '5']],

                        quotechar=None, escapechar='\\')

        self._read_test(['1,",3,",5'], [['1', '"', '3', '"', '5']],

                        quoting=csv.QUOTE_NONE, escapechar='\\')

        # will this fail where locale uses comma for decimals?

        self._read_test([',3,"5",7.3, 9'], [['', 3, '5', 7.3, 9]],

                        quoting=csv.QUOTE_NONNUMERIC)

        self.assertRaises(ValueError, self._read_test,

                          ['abc,3'], [[]],

                          quoting=csv.QUOTE_NONNUMERIC)

    def test_read_bigfield(self):

        # This exercises the buffer realloc functionality and field size

        # limits.

        limit = csv.field_size_limit()

        try:

            size = 50000

            bigstring = 'X' * size

            bigline = '%s,%s' % (bigstring, bigstring)

            self._read_test([bigline], [[bigstring, bigstring]])

            csv.field_size_limit(size)

            self._read_test([bigline], [[bigstring, bigstring]])

            self.assertEqual(csv.field_size_limit(), size)

            csv.field_size_limit(size-1)

            self.assertRaises(csv.Error, self._read_test, [bigline], [])

            self.assertRaises(TypeError, csv.field_size_limit, None)

            self.assertRaises(TypeError, csv.field_size_limit, 1, None)

        finally:

            csv.field_size_limit(limit)

    def test_read_linenum(self):

        for r in (csv.reader(['line,1', 'line,2', 'line,3']),

                  csv.DictReader(['line,1', 'line,2', 'line,3'],

                                 fieldnames=['a', 'b', 'c'])):

            self.assertEqual(r.line_num, 0)

            r.next()

            self.assertEqual(r.line_num, 1)

            r.next()

            self.assertEqual(r.line_num, 2)

            r.next()

            self.assertEqual(r.line_num, 3)

            self.assertRaises(StopIteration, r.next)

            self.assertEqual(r.line_num, 3)

class TestDialectRegistry(unittest.TestCase):

    def test_registry_badargs(self):

        self.assertRaises(TypeError, csv.list_dialects, None)

        self.assertRaises(TypeError, csv.get_dialect)

        self.assertRaises(csv.Error, csv.get_dialect, None)

        self.assertRaises(csv.Error, csv.get_dialect, "nonesuch")

        self.assertRaises(TypeError, csv.unregister_dialect)

        self.assertRaises(csv.Error, csv.unregister_dialect, None)

        self.assertRaises(csv.Error, csv.unregister_dialect, "nonesuch")

        self.assertRaises(TypeError, csv.register_dialect, None)

        self.assertRaises(TypeError, csv.register_dialect, None, None)

        self.assertRaises(TypeError, csv.register_dialect, "nonesuch", 0, 0)

        self.assertRaises(TypeError, csv.register_dialect, "nonesuch",

                          badargument=None)

        self.assertRaises(TypeError, csv.register_dialect, "nonesuch",

                          quoting=None)

        self.assertRaises(TypeError, csv.register_dialect, [])

    def test_registry(self):

        class myexceltsv(csv.excel):

            delimiter = "\t"

        name = "myexceltsv"

        expected_dialects = csv.list_dialects() + [name]

        expected_dialects.sort()

        csv.register_dialect(name, myexceltsv)

        try:

            self.failUnless(csv.get_dialect(name).delimiter, '\t')

            got_dialects = csv.list_dialects()

            got_dialects.sort()

            self.assertEqual(expected_dialects, got_dialects)

        finally:

            csv.unregister_dialect(name)

    def test_register_kwargs(self):

        name = 'fedcba'

        csv.register_dialect(name, delimiter=';')

        try:

            self.failUnless(csv.get_dialect(name).delimiter, '\t')

            self.failUnless(list(csv.reader('X;Y;Z', name)), ['X', 'Y', 'Z'])

        finally:

            csv.unregister_dialect(name)

    def test_incomplete_dialect(self):

        class myexceltsv(csv.Dialect):

            delimiter = "\t"

        self.assertRaises(csv.Error, myexceltsv)

    def test_space_dialect(self):

        class space(csv.excel):

            delimiter = " "

            quoting = csv.QUOTE_NONE

            escapechar = "\\"

        fd, name = tempfile.mkstemp()

        fileobj = os.fdopen(fd, "w+b")

        try:

            fileobj.write("abc def\nc1ccccc1 benzene\n")

            fileobj.seek(0)

            rdr = csv.reader(fileobj, dialect=space())

            self.assertEqual(rdr.next(), ["abc", "def"])

            self.assertEqual(rdr.next(), ["c1ccccc1", "benzene"])

        finally:

            fileobj.close()

            os.unlink(name)

    def test_dialect_apply(self):

        class testA(csv.excel):

            delimiter = "\t"

        class testB(csv.excel):

            delimiter = ":"

        class testC(csv.excel):

            delimiter = "|"

        csv.register_dialect('testC', testC)

        try:

            fd, name = tempfile.mkstemp()

            fileobj = os.fdopen(fd, "w+b")

            try:

                writer = csv.writer(fileobj)

                writer.writerow([1,2,3])

                fileobj.seek(0)

                self.assertEqual(fileobj.read(), "1,2,3\r\n")

            finally:

                fileobj.close()

                os.unlink(name)

            fd, name = tempfile.mkstemp()

            fileobj = os.fdopen(fd, "w+b")

            try:

                writer = csv.writer(fileobj, testA)

                writer.writerow([1,2,3])

                fileobj.seek(0)

                self.assertEqual(fileobj.read(), "1\t2\t3\r\n")

            finally:

                fileobj.close()

                os.unlink(name)

            fd, name = tempfile.mkstemp()

            fileobj = os.fdopen(fd, "w+b")

            try:

                writer = csv.writer(fileobj, dialect=testB())

                writer.writerow([1,2,3])

                fileobj.seek(0)

                self.assertEqual(fileobj.read(), "1:2:3\r\n")

            finally:

                fileobj.close()

                os.unlink(name)

            fd, name = tempfile.mkstemp()

            fileobj = os.fdopen(fd, "w+b")

            try:

                writer = csv.writer(fileobj, dialect='testC')

                writer.writerow([1,2,3])

                fileobj.seek(0)

                self.assertEqual(fileobj.read(), "1|2|3\r\n")

            finally:

                fileobj.close()

                os.unlink(name)

            fd, name = tempfile.mkstemp()

            fileobj = os.fdopen(fd, "w+b")

            try:

                writer = csv.writer(fileobj, dialect=testA, delimiter=';')

                writer.writerow([1,2,3])

                fileobj.seek(0)

                self.assertEqual(fileobj.read(), "1;2;3\r\n")

            finally:

                fileobj.close()

                os.unlink(name)

        finally:

            csv.unregister_dialect('testC')

    def test_bad_dialect(self):

        # Unknown parameter

        self.assertRaises(TypeError, csv.reader, [], bad_attr = 0)

        # Bad values

        self.assertRaises(TypeError, csv.reader, [], delimiter = None)

        self.assertRaises(TypeError, csv.reader, [], quoting = -1)

        self.assertRaises(TypeError, csv.reader, [], quoting = 100)

class TestCsvBase(unittest.TestCase):

    def readerAssertEqual(self, input, expected_result):

        fd, name = tempfile.mkstemp()

        fileobj = os.fdopen(fd, "w+b")

        try:

            fileobj.write(input)

            fileobj.seek(0)

            reader = csv.reader(fileobj, dialect = self.dialect)

            fields = list(reader)

            self.assertEqual(fields, expected_result)

        finally:

            fileobj.close()

            os.unlink(name)

    def writerAssertEqual(self, input, expected_result):

        fd, name = tempfile.mkstemp()

        fileobj = os.fdopen(fd, "w+b")

        try:

            writer = csv.writer(fileobj, dialect = self.dialect)

            writer.writerows(input)

            fileobj.seek(0)

            self.assertEqual(fileobj.read(), expected_result)

        finally:

            fileobj.close()

            os.unlink(name)

class TestDialectExcel(TestCsvBase):

    dialect = 'excel'

    def test_single(self):

        self.readerAssertEqual('abc', [['abc']])

    def test_simple(self):

        self.readerAssertEqual('1,2,3,4,5', [['1','2','3','4','5']])

    def test_blankline(self):

        self.readerAssertEqual('', [])

    def test_empty_fields(self):

        self.readerAssertEqual(',', [['', '']])

    def test_singlequoted(self):

        self.readerAssertEqual('""', [['']])

    def test_singlequoted_left_empty(self):

        self.readerAssertEqual('"",', [['','']])

    def test_singlequoted_right_empty(self):

        self.readerAssertEqual(',""', [['','']])

    def test_single_quoted_quote(self):

        self.readerAssertEqual('""""', [['"']])

    def test_quoted_quotes(self):

        self.readerAssertEqual('""""""', [['""']])

    def test_inline_quote(self):

        self.readerAssertEqual('a""b', [['a""b']])

    def test_inline_quotes(self):

        self.readerAssertEqual('a"b"c', [['a"b"c']])

    def test_quotes_and_more(self):

        # Excel would never write a field containing '"a"b', but when

        # reading one, it will return 'ab'.

        self.readerAssertEqual('"a"b', [['ab']])

    def test_lone_quote(self):

        self.readerAssertEqual('a"b', [['a"b']])

    def test_quote_and_quote(self):

        # Excel would never write a field containing '"a" "b"', but when

        # reading one, it will return 'a "b"'.

        self.readerAssertEqual('"a" "b"', [['a "b"']])

    def test_space_and_quote(self):

        self.readerAssertEqual(' "a"', [[' "a"']])

    def test_quoted(self):

        self.readerAssertEqual('1,2,3,"I think, therefore I am",5,6',

                               [['1', '2', '3',

                                 'I think, therefore I am',

                                 '5', '6']])

    def test_quoted_quote(self):

        self.readerAssertEqual('1,2,3,"""I see,"" said the blind man","as he picked up his hammer and saw"',

                               [['1', '2', '3',

                                 '"I see," said the blind man',

                                 'as he picked up his hammer and saw']])

    def test_quoted_nl(self):

        input = '''\

1,2,3,"""I see,""

said the blind man","as he picked up his

hammer and saw"

9,8,7,6'''

        self.readerAssertEqual(input,

                               [['1', '2', '3',

                                   '"I see,"\nsaid the blind man',

                                   'as he picked up his\nhammer and saw'],

                                ['9','8','7','6']])

    def test_dubious_quote(self):