Re: Multiple rules for same target: what is the order of prerequisites in $^ and what does $< expand to?

On 9/27/23 21:39, Philip Guenther wrote:
> On Wed, Sep 27, 2023 at 7:52 AM Renaud Pacalet <[email protected] <mailto:[email protected]>> wrote:
> 
>     On 9/27/23 16:32, Bahman Movaqar wrote:
>      > On Wed, 2023-09-27 at 12:16 +0200, Renaud Pacalet wrote:
>      >
>      >> 3. Ignore all other implicit rules (which seems to contradict 4.11).
>      >
>      > Make only ignores the implicit rules for which there are
>      > explicit/static equivalents.
> 
> 
> Per 10.1 "Using Implicit Rules", make will only use an implicit rule for a target if there isn't an explicit rule with a recipe.  If a target has an explicit rule with a recipe, then make won't bother looking for implicit rules and so won't use them to add more prerequisites.
> 
> Note that there are additional requirements on what implicit rules will be used even when they are considered: if they have a prerequisite that doesn't exist and make doesn't see how to build, then it'll ignore the rule.
> 
>     Thanks for your answer, Bahman. Not sure what "equivalent" means but with the following Makefile:
> 
>     %.a: %.b
> 
>     a.a: a.c
>           echo $^
> 
> 
> a.a has an explicit rule with a recipe so the %.a:%.b rule has no effect on the a.a target.

Thanks a lot Philip. This is in line with my own conclusions. I guess my algorithm is thus probably correct. Note that the findings of Bahman (target-specific variables are taken into accounts even if declared in a pattern rule while an explicit rule with recipe exists) complicate things. I will continue my experiments and, if I find a way to explain all this clearly enough, I'll submit a proposal of improvement for 4.11 which I find misleading.

Best regards,

Renaud.