cfml

Need a sanity check. What would you expect the output of `writeOutput( val("") ? "yes" : "no" )` to be? I'm getting "yes" in CF2025 and "no" in all past versions.

without looking at the docs, i would expect it to be "no" because i thought "val()" always returned a number, so i would think that "" would be a 0, aka false.

val returns 0 for empty strings, and 0 is false

I'm assuming ACF returns something like [] or NaN.

I'd run a quick test, but 2025 isn't up on trycf yet

Can you test just `val("")` by itself and `0 ? "yes" : "no"` by itself?

<http://cffiddle.org|cffiddle.org> has a 2025 engine where I confirmed this as well as on a dev copy

val('") results in 0, and 0 has always been falsy up until now

`0 ? "yes" : "no"` results in "no" as expected

check what the type of variable is coming out of val

It's a `java.lang.Double` on ACF 2023 and Lucee 6

that's some crazy nonsense right there is passing an explicit `0` works but passing `val( "" )` doesn't

I'm going to submit this as a bug, but I wanted to make sure I'm not missing something obvious before I do

perhaps they're creating incorrect bytecode in that case

`0 ? "yes" : "no"` results in no, but `val(0) ? "yes" : "no"` results in yes. lovely.

<@U01EY27APNH> - just making sure you saw the above. this definitely has me worried about upgrading.

<@U0BHP93LL> heads up here, can you hurl an engineer at this to see where we zigged instead of zagging on this?

This returns as falsey if there are any nearby Grues in the shadows/darkness as protection for the programmer. (we're looking to see what's up here)

<@U0BHP93LL> <@U01EY27APNH> I logged a bug (CF-4225874) if that's helpful

This is the code of a teenage girl trying to decide?  all kidding aside, looks like a bug as cf2021 its a no

seems to apply only to the ternary operator:
```writeOutput( val("") ? "yes" : "no" );
if ( val("") ) {
        writeOutput("yup");
} else {
        writeOutput("nope");
} ```
outputs `yesnope`  on CF2025

I see you noted that in your ticket, just looking at it now

I got and tested a hotfix from someone at Adobe this morning, so I'm happy to see there's some acknowledgement and movement on it

The patch with the fix has been attached to <https://tracker.adobe.com/#/view/CF-4225874>

Thanks, <@U029V6Y3S91>.
I am unable to run the Jar because of "no main manifest attribute".

Nevermind. Placing the Jar in /lib/updates/ did the trick.