Normalize mb_ereg() return value #6331
Conversation
|
If we don't do this or something similar we can just mark the argument as Not sure who to ask about this, maybe @kocsismate @cmb69 ? |
|
Well, the current behavior is terrible, so returning bool makes sense, but I'm slightly concerned about the potential BC break; some code may rely on the returned length. That code would likely have to be adjusted to check for |
|
Right, there certainly is some potential for breakage. That said, I just checked all usages of mb_ereg/mb_eregi in the top 2k composer packages, and none of them would be affected by this change. That doesn't mean it affects nobody, but at least the impact should be fairly limited. With that in mind, I'm going to land this now to make sure it's in RC2, and may back this out again if I'm estimating the impact incorrectly. |
mb_ereg()/mb_eregi() currently have an inconsistent return value based on whether the $matches parameter is passed or not:
Coupling this behavior to the $matches parameter doesn't make sense -- we know the match length either way, there is no technical reason to distinguish them. However, returning the match length is not particularly useful either, especially due to the need to convert 0-length into 1-length to satisfy "truthy" checks. We could always return
1, which would kind of match the behavior of preg_match() -- however, preg_match() actually returns the number of matches, which is0or1for preg_match(), whilefalsesignals an error. However, mb_ereg() returnsfalseboth for no match and for an error. This would result in an odd1|falsereturn value.The proposal here is to canonicalize mb_ereg() to always return a boolean, where true indicates a match and false indicates no match or error. This also matches the behavior of the mb_ereg_match() and mb_ereg_search() functions.