When building Python with:

    ./configure --disable-unicode

the test suite has many spurious failures because
modules and tests assume that unicode support is available.

Installing this Python (using "make install") fails at
the end of the first run of compileall.py, since that
script detects and reports failures; the output
contains many lines with the message

    Sorry: ValueError: ('Unicode escapes not legal when
Unicode disabled',)

This should be fixed, or --disable-unicode should be
removed.

Logged In: YES 
user_id=3066

The trunk and release24-maint branch will no longer compile
with --diable-unicode.  I get the following error message
from the trunk:

c++ -pthread  -Xlinker -export-dynamic -o python \
                Modules/python.o \
                libpython2.5.a -lpthread -ldl  -lutil   -lm
libpython2.5.a(posixmodule.o)(.text+0x39a0): In function
`posix_tmpnam':
../Modules/posixmodule.c:6161: warning: the use of
`tmpnam_r' is dangerous, better use `mkstemp'
libpython2.5.a(posixmodule.o)(.text+0x3906): In function
`posix_tempnam':
../Modules/posixmodule.c:6116: warning: the use of `tempnam'
is dangerous, better use `mkstemp'
libpython2.5.a(_codecsmodule.o)(.text+0xf9): In function
`codec_encode':
../Modules/_codecsmodule.c:108: undefined reference to
`PyUnicode_GetDefaultEncoding'
libpython2.5.a(_codecsmodule.o)(.text+0x165): In function
`codec_decode':
../Modules/_codecsmodule.c:141: undefined reference to
`PyUnicode_GetDefaultEncoding'
collect2: ld returned 1 exit status

The release24-maint branch gives similar results.

Logged In: YES 
user_id=1188172

Okay, both branches compile again. The test failures remain,
though.

The --disable-unicode option and related code should be removed from 3.0
and probably from 2.6, too.

I renamed --enable-unicode to --with-wide-unicode in r59829; this should
handle the Py3k case.

I added suitable Py_USING_UNICODE directives in r59830 so that the trunk
compiles again without unicode.