Re: [docs] [yocto-docs][PATCH v2] Update the documentation for SRCPV

["Antonin Godard" <antonin.godard@bootlin.com>]

Hi Quentin,

On Wed Nov 27, 2024 at 3:08 PM CET, Quentin Schulz via lists.yoctoproject.org wrote:
[...]
>>>>> I think we could simplify the example. The point is not necessarily to
>>>>> use % with a git-named recipe but rather that a git-named recipe is not
>>>>> going to match "git" but the "PV" defined in the recipe. What do you think?
>>>>
>>>> Ok, something like:
>>>>
>>>>     If your PV is equal to "a.b.c+git", you can match it with "a.b.c%" in
>>>>     PREFERRED_VERSION?
>>>>
>>>> This makes more sense with the current state of things, and simplifies the
>>>> example a bit.
>>>>
>>>
>>> """
>>> If you want to select a recipe named foo_git.bb which has :term:`PV` set
>>> to `1.2.3+git`, you can do so by setting :term:`PREFERRED_VERSION_foo`
>>> to `1.2.3%` (i.e. using simply `git` will not work as the name of the
>>> recipe isn't used, but rather its :term:`PV`).
>>> """
>>>
>>> or something like that?
>>>
>>> Is PV really extended with the git hash when there's a + in there or is
>>> it only PKGV that gets it? Basically, do you even need % character when
>>> we have a + in PV now?
>>>
>>> Not sure it's clear, lemme know if I need to rephrase my question.
>> 
>> So yes you got it right: PV will never contain the hash, only PKGV will (and
>> only during do_install iirc, so bitbake-getvar PKGV will not show the hash).
>> 
>> Which makes me wonder if giving the example above is any useful, since you could
>> just set PREFERRED_VERSION_foo to "1.2.3+git".
>> 
>
> It is useful! The point is to highlight the "version" in the filename of 
> the recipe is not the one you should use in PREFERRED_VERSION, but 
> rather the value of PV. Incidentally, PV = version in the filename by 
> default I believe (maybe for the git ones too?) but it is 
> recommended/required to set PV for git recipes. So in order to pick the 
> git recipe, you need to use the PV set inside the git recipe and not use 
> "git" along. I believe this is what the original author wanted this 
> example to transpire?

I didn't think of it that way, but yes, maybe the author wanted to transpire
this + the fact that in the hash wasn't predictable hence the "%". Or only one
or the other.

In any case, I get what you mean, and I'll update the description in a v3.

Thanks,
Antonin

-- 
Antonin Godard, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com