[OE-core] subtle weirdness when you combine "_append" with "+="?

[Khem Raj]

> On Aug 16, 2016, at 8:56 AM, Ulf Magnusson <ulfalizer at gmail.com> wrote:
> 
> On Tue, Aug 16, 2016 at 5:49 PM, Robert P. J. Day <rpjday at crashcourse.ca> wrote:
>> On Tue, 16 Aug 2016, Ulf Magnusson wrote:
>> 
>>> On Tue, Aug 16, 2016 at 4:02 PM, Robert P. J. Day <rpjday at crashcourse.ca> wrote:
>>>> 
>>>>  was about to submit a small number of patches to clean up redundancy
>>>> when people combine "_append" with "+=" (because it offends my
>>>> delicate sensibilities), and ran across this in oe-core,
>>>> unfs3_0.9.22.r497.bb:
>>>> 
>>>>  DEPENDS_append_class-nativesdk += "flex-nativesdk"
>>>> 
>>>> uh, what?
>>>> 
>>>>  most of the time, i assume the above doesn't hurt, it's just ...
>>>> silly. but normally, with "_append", you *need* to add the leading
>>>> space explicitly, and that's not being done above. so does that mean
>>>> that combining "_append" with "+=" *does* generate a leading space?
>>>> that just makes my head hurt -- the possibility that "_append" is
>>>> being used in a way that normally makes it fail, only to have "+="
>>>> step in and save the day. at which point "_append" saves processing
>>>> that until the end of parsing? yeesh.
>>>> 
>>>>  thoughts?
>>> 
>>> By the point the += is handled, the override won't have been interpreted
>>> yet. My guess is that += fetches the value of the variable
>>> "DEPENDS_append_class-nativesdk", gets back the empty string, and
>>> adds a space followed by "flex-nativesdk" to that.
>>> 
>>> The resulting " flex-nativesdk" is then interpreted as usual when the
>>> overrides are handled.
>>> 
>>> You might like the note I added to
>>> https://www.yoctoproject.org/docs/2.2/bitbake-user-manual/bitbake-user-manual.html#override-style-operation-advantages
>>> by the way. :)
>> 
>>  that note is pretty much what i've been whining about for a long
>> time. :-) in any event, when one sees something like the above:
> 
> The intent of the note was to discourage use of '+=' together with _append,
> because it's redundant and potentially confusing. Do you think it fails
> (even in context)? :/

the _append/_prepend in conjunction with += is a undocumented behavior
however, I do not see it as much a side effect. But in future bitbake may
silently change its behavior, so I agree its always good to stay in safe
waters.

> 
>> 
>>  DEPENDS_append_class-nativesdk += "flex-nativesdk"
>> 
>> what is the *proper* cleanup?
> 
> I'd say the following:
> 
> DEPENDS_append_class-nativesdk = " flex-nativesdk"
> 
> Cheers,
> Ulf
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core at lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 204 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.openembedded.org/pipermail/openembedded-core/attachments/20160816/29ccd3c6/attachment-0002.sig>