gettext-tools
===================================================================
All 299 tests passed (28 tests were not run)
   SKIP: gettext-3, -4, -5, -6, -7
   SKIP: msgunfmt-csharp-1
   SKIP: msgunfmt-java-1
   SKIP: recode-sr-latin-1
   SKIP: recode-sr-latin-2
   SKIP: lang-* PASS: -elisp, -ycp, -po, -rst

gettext-3: skipped on non-glibc systems
gettext-4: skipped (no 'traditional french locale')
gettext-5: skipped on non-glibc systems
gettext-6: skipped on non-glibc systems
gettext-7: skipped on non-glibc systems
msgunfmt-csharp-1: skipped because csharp not available
msgunfmt-java-1:   skipped because java support not built
recode-sr-latin-1: skipped (no 'french UTF-8 locale')
recode-sr-latin-2: skipped (no 'japanese EUC-JP locale')


Note that ALL of the lang- tests are now skipped (or, if not skipped, 
they should have been skipped).  Cygwin's setlocale() function
checks that the argument is either "C" or "" (or is in a slightly 
larger list if cygwin1.dll was compiled with _MB_CAPABLE), and fails 
otherwise.  The properly-written lang- tests check the return value 
of setlocale, and if it fails (which it does), then the test itself
is skipped.

This is a shame, because the underlying bindtextdomain stuff actually
works and it would be nice to verify that using the testsuite. Oh 
well. Maybe on cygwin-1.7.


gettext-tools/gnulib-tests
===================================================================
All tests fail when --enable-relocation (the tests do not "activate"
the relocation support, but the gnulib library they link to expect
it).

When not --enable-relocation (the normal case):
All 47 tests passed (2 tests were not run)

test-mbsstr2.sh: skipped (no french Unicode locale)
test-mbsstr3.sh: skipped (no chinese GB18030 locale)


gettext-runtime
===================================================================
All 1 tests passed


gettext-rpath
===================================================================
All 30 tests passed

The lib-link test suite has been modified for cygwin, in two ways:
(1) the test scripts extend the PATH to include the installation
    directories of the sub-dependent libraries, and
(2) the Makefiles for the executable tests extend the PATH to
    include the installation directory of the direct dependent
    library.
In this way, the PATH is sure to include all directories which
contain the DLLs used by the test application. However, all of these 
changes are simply workarounds for the primary defect on windows:
there IS no RPATH support. Modifying the $PATH is no different than
modifying $LD_LIBRARY_PATH on other platforms -- and all of these
rpath tests are *supposed* to be checking that RPATH support actually
works *without* manipulating $LD_LIBRARY_PATH.

No matter how you slice it on windows, it doesn't. So ALL of these
changes are "cheating". Earlier versions of gettext (0.15 and before)
*erroneously* passed many of these tests *without* the workaround
above -- because the tests in which the dependent and sub-dependent
libraries were installed into different prefixes did NOT actually
link against those DLLs, but linked statically. So, naturally, there
was no runtime "can't find the DLL" issue. This was "fixed" in
gettext-0.17 -- so now these tests began to fail, as they always should
have done, without the workarounds described above.


The following six tests WOULD fail on cygwin (regardless of the
"cheating" above), except that they have been removed from the
list of tests actually performed: 
  rpath-2aac  rpath-2abc  ### test files no longer present
  rpath-3aac  rpath-3aaf  ### ditto for these two
  rpath-3abc  rpath-3abf  ### ditto for these two
These are known (and expected) failures on cygwin; you can't create
a shared library that depends on a static lib.

Similarly, the following tests fail on ALL platforms, but have
been removed from the actual test sequence:
  rpath-3aae  rpath-3aag  ### test files present but excluded for all platforms
  rpath-3abe  rpath-3abg  ### test files present but excluded for all platforms

