Mail Archives: djgpp/1996/08/13/20:30:09
In article <199608131727 DOT VAA07673 AT video DOT yars DOT free DOT net>, "Alexander V. Lukyanov" <lav AT video DOT yars DOT free DOT net> writes:
>> From: Joshua Cannon Butcher <lchandar AT mindspring DOT com>
>
>> 7) And for everyone in General, why do you have to stray away from
>> industry standards? Calling object files .O files instead of .OBJ,
>> calling C++ files .CC instead of .CPP, and .a instead of .LIB for
>> library. HELLO! Its not copyright infringement to use the same
>> extensions, and it would make the transition for existing C and C++
>> users to use. This is quite frankly scaring me, and almost makes me
>> want to pay the $500 for Borland C++ 5.0 so I can have the "standard" of
>> the computer programming industry.
>
> Why do you call those silly dos extensions "industry standard"?
> GCC is primarily Unix compiler and it was ported to dos. All Unices
> have .o, .cc, .a etc as convensional suffixes. I think they are older than
> dos. (Probably .obj was used in other os'es, like CP/M, but I don't know)
>
> Well, you can call your files as you want. See how:
> ar q mylib.lib files...
> gcc myprog.cpp mylib.lib
>
> You can even do 'ren ar.exe lib.exe'
>
This argument over "standard" DOS names vs. Unix is silly in my opinion, so I
won't join in. Except to say that if .o, .a and .cc scare you, you probably
should spend $500 and feel secure. Nuff said.
I remember someone saying that if you say "gcc myprog.o mylib.a" the linker
doesn't exclude routines in 'mylib' that aren't used. To do this one must use
the "-l" option. (eg. "gcc myprog.o -lmy" to link libmy.a).
Adios
Mark Wodrich.
- Raw text -