Remove custom 'p' length modifier from custom snprintf implementation #5103
Conversation
|
I think it's pretty likely that this one is used by extensions. We originally used Can you check usages in PECL extensions for this? |
|
On Tue, 21 Jan 2020, 13:18 Nikita Popov, ***@***.***> wrote:
I think it's pretty likely that this one is used by extensions. We
originally used %pd to print zend_long, but later switched to
ZEND_LONG_FMT because it plays better with compiler checks for printf.
Can you check usages in PECL extensions for this?
I thought so, is there a tool to easily check PECL extensions? Or does it
need to be done manually?
… —
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#5103?email_source=notifications&email_token=AB4KLAFV2RZODZOH33I52OTQ63Y27A5CNFSM4KJSV4S2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEJPWKTA#issuecomment-576677196>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AB4KLAB6OLGLFO34P4CZUPDQ63Y27ANCNFSM4KJSV4SQ>
.
|
|
I think @petk used to do this, maybe he can share how it's done? |
Girgias
force-pushed
the
drop-custom-p-length-modifier-snprintf
branch
from
11592e7 to
dc526c2
Compare
|
Made a second commit which removes it from the custom |
|
Speaking of, can you please also apply the |
|
To search occurrences in extensions back then, I've used very dummy technique:
|
|
@petk did you have time to look into it? |
|
Small ping @petk did you have time to look into this? |
|
Hello, nope. Don't wait for me here with this check. No time for this script atm :) |
Mind providing the script so that I can run it myself or too complicated? |
|
This has been done in 0d6358f |
From my understanding this gives a
zend_longrepresentation, however it seems to be unused going from this grep search: https://gist.github.com/Girgias/06253c18fde29af25ed50b29c0d8da38Possibly because the long format macros also take care of this modifier? Or maybe it's just mostly used in conjunction with
spprintf(which I'll need to make a separate search for).(Drive by the mysqlnd usage seems to suggest to output a pointer which we don't really support fully from my understanding).