Remove invalid pop directives

[willrowe]

pop only accepts true or an integer larger than zero.

The invalid directives break the syntax in pre-release builds of Sublime. The syntax would not even be loaded. This will ensure compatibility going forward.

[jfcherng]

Just for reference: https://discordapp.com/channels/280102180189634562/280157121897562114/677222161106206720

[Medalink]

@willrowe we should target a specific version of sublime (within packages.json) for this change to retain backward compatibility I think. Comments say it was "ignored" before but to be safe I think I rather mark this change as a forward only change. Can you add the specific build numbers with >= in packages?

[jfcherng]

I don't use this package but just wonder what's the point of using/keeping pop: false? It does nothing in ST 3XXX and causes an error/warning in ST 4XXX.

[Medalink]

I don't use this package but just wonder what's the point of using/keeping pop: false? It does nothing in ST 3XXX and causes an error/warning in ST 4XXX.

Are we 100% certain it does nothing in ST3? If we are or can find a document example (or proven test) that this is indeed true I am willing to make the change. However if we are not then it's much more risky to let this flow backwards when it has worked this way for years.

[willrowe]

I believe it would be best to fix this for ST3, even though it technically works by being ignored. false has never been a valid value for pop. In the ST3 docs it explicitly says:

• pop. Pops the current context off the stack. The only accepted value for this key is true.

I defer to you, of course, if you would still like me to add the version constraint.

[Medalink]

I believe it would be best to fix this for ST3, even though it technically works by being ignored. false has never been a valid value for pop. In the ST3 docs it explicitly says:

• pop. Pops the current context off the stack. The only accepted value for this key is true.

I defer to you, of course, if you would still like me to add the version constraint.

Ahh perfect, this is enough to convince me we can safely remove it. Thanks!

[Medalink]

Thanks for the contributions!

[willrowe]

Thanks for taking a look at this @Medalink, really appreciate the package!