Re: [commits] Horde branch master updated. e82250e4a2d28e0464ad26f3d6affb5ae51d80b2
Jan Schneider <[email protected]>
| Newsgroups | gmane.comp.horde.devel |
|---|---|
| Message-ID | <[email protected]> |
Zitat von Michael J Rubinsky <[email protected]>: > Quoting Michael J Rubinsky <[email protected]>: > >> Quoting Michael J Rubinsky <[email protected]>: >> >>> Quoting Michael J Rubinsky <[email protected]>: >>> >>>> Quoting Jan Schneider <[email protected]>: >>>> >>>>>> commit 6697c620ead20833d8ca077f64e4cc26ed1ad8d8 >>>>>> Author: Michael J Rubinsky <[email protected]> >>>>>> Date: Wed Jun 24 12:31:18 2015 -0400 >>>>>> >>>>>> Allow a string value of '0'. >>>>>> >>>>>> Part of fix for Bug: 14025 >>>>>> >>>>>> framework/Perms/lib/Horde/Perms/Permission.php | 5 +++-- >>>>>> 1 files changed, 3 insertions(+), 2 deletions(-) >>>>>> >>>>>> http://github.com/horde/horde/commit/6697c620ead20833d8ca077f64e4cc26ed1ad8d8 >>>>> >>>>> How about checking for strlen() instead? >>>> >>>> I tried that, even when using strval() first, and it was still >>>> taking it as an "empty" value. >>> >>> Though a quick test script shows that this should not be the case. >>> I'll see if there was something else going on. >> >> Yeah, I think I tested that before changing the key values to be >> strings in Horde :/ Updated in Git. > > This also takes an integer value of 0 and returns a length of 1, so > this technically breaks previous behavior of allowing an integer > value of 0 "clear" the permission. Not sure if that falls under the > "previous behavior is broken, so it's ok" school of breaking BC or > not.... Ah, I thought that integer 0 should always have been a valid value. If not, then using strlen() would indeed be a BC break. -- Jan Schneider The Horde Project http://www.horde.org/ https://www.facebook.com/hordeproject -- dev mailing list Frequently Asked Questions: http://wiki.horde.org/FAQ To unsubscribe, mail: [email protected]