On Mon, Jan 08, 2001 at 10:21:12PM +0100, Richard Levitte - VMS Whacker wrote:
>> Why? Actually a number of exported header files *do* include e_os2.h
>> (without including e_os.h, which is not exported).
> The conditionals in e_os2.h depend on macros defined in e_os.h. That
> makes e_os2.h dependent on e_os.h, and the whole thing becomes ridiculous.
That would be MSDOS and VMS, right? So the #ifs in e_os2.h should be
expanded according to the definitions in e_os.h. If MSDOS and VMS are
in general not meant to be defined on those systems, the exported
headers should not define them either.
BTW, even e_os.h tests for MSDOS before it defines it. So either
it is defined anyway and e_os.h does not have to define it, or e_os.h
in itself is ridiculous.
>> However, e_os.h is not installed by "make install".
> Oh yes, it is!
Er, you are right. I moved it back into EXHEADER for simplicity
because the link in the source code include directory is needed.
To separate these things cleanly, there should be another type
of headers beside EXHEADER that is looked at by "make links",
but not by "make install". (We could put those files into
a new directory in the source directory tree.)
--
Bodo Möller <[EMAIL PROTECTED]>
PGP http://www.informatik.tu-darmstadt.de/TI/Mitarbeiter/moeller/0x36d2c658.html
* TU Darmstadt, Theoretische Informatik, Alexanderstr. 10, D-64283 Darmstadt
* Tel. +49-6151-16-6628, Fax +49-6151-16-6036
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
