Re: CVS commit: src/usr.bin/make

[christos%zoulas.com@localhost (Christos Zoulas)]

On Jun 19,  6:58am, martin%duskware.de@localhost (Martin Husemann) wrote:
-- Subject: Re: CVS commit: src/usr.bin/make

| There was a discussion way back, exactly about what the default
| behaviour should be:
| 
| http://mail-index.netbsd.org/tech-toolchain/2012/08/06/msg001900.html
| 
| If we now want to change the default (as you did), we should at least
| have a follow-up discussion on tech-toolchain.

I just read all 4 messages and there was no strong consensus either way:

Iain:
    I don't like the thought of changing the result of options that are
    already present, since it is not beyond conceivable that doing so would
    break somebodys setup somewhere..

    but I agree that -V can be a bit irritating, and providing a modifier to
    cause it to produce the expanded value directly does sound useful...

joerg:

    My usage distribution is far less screwed into either direction, so I am
    not too sure about changing it.

simon:

    Being able to get the raw value of a variable is definitely useful,
    but I'm wondering if it makes a good default.
    99% of the time I end up doing:

    make -V VAR
    make -V '${VAR}


I ask again: Do you want to have 'make -V VAR' do something useful by
default, or do you want it to produce debugging output because this
is what it did historically?

I am convinced that this behavior is an accident; whoever wrote
the code thought that Var_Value() returns the value of a variable,
and did not realize that it returns the unexpanded value (because
it worked for the cases she/he checked). It is my fault because I
I introduced -V in 1.31 in 1996 without realizing that.

christos