id-utils-3.2d on CygWin

[Marius Gedminas]

GNU id-utils 3.2d compile out of the box on CygWin (1.1.8).  That's the
good news.  The bad news is they do not work.

$ ./configure && make && make check
[... snip ...]
Making check in testsuite
make[1]: Entering directory `/home/gedas/src/id-utils-3.2d/testsuite'
make  check-TESTS
make[2]: Entering directory `/home/gedas/src/id-utils-3.2d/testsuite'
make[3]: Entering directory `/home/gedas/src/id-utils-3.2d/testsuite'
/bin/sh ./consistency .. ../src
../src/mkid -m ../libidu/id-lang.map ..
      0 [main] fnid 1712 handle_exceptions: Exception: STATUS_ACCESS_VIOLATION
   1597 [main] fnid 1712 stackdump: Dumping stack trace to fnid.exe.stackdump
Signal 11
../src/lid: 
Good.  xtokid and lid agree.
All checks successful.
make[3]: Leaving directory `/home/gedas/src/id-utils-3.2d/testsuite'
PASS: consistency
==================
All 1 tests passed
==================
make[2]: Leaving directory `/home/gedas/src/id-utils-3.2d/testsuite'
make[1]: Leaving directory `/home/gedas/src/id-utils-3.2d/testsuite'
make[1]: Entering directory `/home/gedas/src/id-utils-3.2d'
make[1]: Leaving directory `/home/gedas/src/id-utils-3.2d'


It looks like mkid generates corrupt databases.  If I generate ID on a
Linux machine and try lid on CygWin, it works fine.  If I generate ID on
CygWin and try lid on Linux, it segfaults.

I would be glad to assist debugging this problem.  I've tried stepping
through the code with gdb, but couldn't spot the problem.

Marius Gedminas
-- 
MCSE == Must Consult Someone Experienced