Introduce __return_one function

[bananastalktome]

Following #20357 and #18845 , I feel it would be useful to have a return function for 1 for use with filters.

[scribu]

Let's not get carried away with these.

I would argue that even __return_zero() isn't common enough.

[bananastalktome]

I've had to create my own __return_zero() in order to set defaults (which cannot be overwritten by users) with filters pre_option_default_pingback_flag, pre_option_enable_xmlrpc, pre_option_use_trackback, etc., but I could just as easily see someone wanting to force open certain options as well with a __return_one(), which is why I decided to add this patch. Many of the options listed at ​http://codex.wordpress.org/Option_Reference use 0 or 1 as their value, so quick returns for their filter would be at least useful to some.

[nacin]

We use 1 and 0 because it is easy to store in a DB field. In reality you can simply use __return_true and __return_false.

[nacin]

It should be noted that will work for most situations. blog_public and a few other options could hypothetically have meaningful values > 1.

[nacin]

Replying to scribu:

I would argue that even __return_zero() isn't common enough.

I just had a pretty good use case for __return_zero() — I wanted to hook into pre_option_{$option} and turn an option off. Normally I would use __return_false but that is the value passed in that needs to be changed for the bypass to work (rather than null). So 0 it is.

[pcfreak30]

I am aware this is old but I have lost track of the number of times I had to create a dummy function for this purpose when the calling code does a typecheck (true !== 1).

So I vote for this to be added as its simple and will get a lot of use I feel. as for __return_zero, I have used this plenty of times as well.

[johnbillion]

See the comment thread on #39240 for alternative approaches. Personally I don't think there's much demand for this as a dedicated function.

[jorbin]

Going back to maybelater as no demand has been demonstrated. Discussion can continue while the ticket is closed if there is a use case that would make a reconsideration of this decision beneficial.

[pcfreak30]

I may have moved on from this long ago, but this issue and request feel so simple, that I don't think you really can give a justification/use case for it outside the number of times you need to create this yourself in the field.

I am also surprised this is still floating around due to how simple it is.

[desrosj]

Closing back out. I agree that this is a maybelater, but later is not 2022.

As noted, there are a few alternatives shared on #39240 that can be used. Just because a change is simple does not mean it's one worth making.