scott@slp53.sl.home (Scott Lurndal) wrote:
Michael S <already5chosen@yahoo.com> writes:On Fri, 18 Sep 2026 12:55:11 -0000 (UTC)
cross@spitfire.i.gajendra.net (Dan Cross) wrote:
In article <118i2mc$q9l1$1@dont-email.me>, bart <bc@freeuk.com>>
wrote:On 17/09/2026 23:06, Keith Thompson wrote: =20=20bart <bc@freeuk.com> writes: =20>On 17/09/2026 08:24, tTh wrote: =20=20On 9/17/26 02:49, bart wrote: =20>And I've no idea where gcc would look for its headers, other=C2=A0=C2=A0 You just have to read the fscking manual.
than where it keeps its system headers, or how to set it up
to look permanently in certain places. =20
https://gcc.gnu.org/onlinedocs/gcc/Environment-Variables.html
=20
Nobody uses environment variables any more. =20
Obviously untrue. =20
Let's say they're out of fashion. =20
Why would you say such a thing? While they may not be the most
elegant solution to a number of problems, they're very much in
use, and "in fashion", at least on Unix-derived systems.
=20
- Dan C.
=20
Would you design a new program which behavior can be modified by
environment variables? I don't mean standard environment variables,
like locale (although that is also less than great) but environment
variables specific to your program?
Absolutely.
And they would be well documented in the manual page for the program.
LD_DEBUG, for example, is quite useful in certain usage cases and
effectively impossible to handle with a command line option flag.
Why impossible?
Because devs of gnu ld.so were inconsistent or for other reasons?
Les messages affichés proviennent d'usenet.