Re: coccinelle: adjustments for array.cocci?

[Junio C Hamano <gitster@pobox.com>]

Markus Elfring <Markus.Elfring@web.de> writes:

>> I too recall that seemingly redundant entries were noticed during
>> the review and at least back then removing the seemingly redundant
>> ones caused failures in rewriting.
>
> I am curious if the redundancy can be reconsidered once more.
>
> Do you refer to open issues around source code reformatting
> and pretty-printing together with the Coccinelle software here?

Sorry, I do not follow.  

If you are asking if I am interested in following bleeding edge
Coccinelle development and use this project as a guinea pig to do
so, then the answer is no.  I'd rather see us instead staying on the
trailing edge ;-) to make sure that we use common denominator
features that are known to be available in all widely deployed and
perhaps a bit dated versions that come with popular distros.

And if that means we have to accept inefficient ways to express our
patterns, we are willing to pay for that cost.

So, "the A.cocci file uses a set of inefficient expressions that can
be written more concisely like this, using the bleeding edge version
of the syntax" is not a useful improvement for the purpose of this
project, while "the A.cocci file uses a set of inefficient
expressions that can be written more concisely like this, and all
versions of cocci that is newer than X would understand the
notation.  Even distro D that tends to ship with fairly stale
versions of packages ship version X+n, so this change should be
safe" is very much appreciated.

Thanks.