tech-toolchain archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: make -V default behavior change



On Mon, Jun 19, 2017 at 01:42:54PM -0400, Christos Zoulas wrote:
> On Jun 19,  7:13pm, joerg%bec.de@localhost (Joerg Sonnenberger) wrote:
> -- Subject: Re: make -V default behavior change
> 
> | > The code is identical to what it was before with the only thing I changed
> | > is this:
> | > 
> | > -                expandVars = getBoolean(".MAKE.EXPAND_VARIABLES", FALSE);
> | > +                expandVars = getBoolean(".MAKE.EXPAND_VARIABLES", TRUE);
> | 
> | So what happens if .MAKE.EXPAND_VARIABLES is set? Can we really please
> | just go back to the original behavior and add a new option that *always*
> | expands the variable?
> 
> It is a boolean as it was; it's default value has changed. If it
> is not set, then it will default to TRUE now (which is the change
> from the original behavior).
> 
> If it is set, it will act the same as before (it will be set to
> the value the user set). This is sjg's original design. I believe
> he indended the default to be TRUE also, but I will let him speak
> to that. The man page does not specify if the variable will be
> expanded or not in the default case.
> 
> As I mentioned before, the unexpanded output is only useful for
> debugging. Nobody has mentioned a use case that does not involve
> debugging.

Again, you still don't get deterministic behavior of expanded variables.
Franky, this whole thing is a completely mess. So to go forward:
(1) Drop -dV. Useless.
(2) Introduce -v VARIABLE: print the full expansion of ${VARIABLE}.
Ignore .MAKE.EXPAND_VARIABLES, do not expand ${} etc., just plain
variable expansion.
(3) Revert -V behavior.

Joerg


Home | Main Index | Thread Index | Old Index