[Wliu]

Something I've noticed while using the new API docs page is that function arguments no longer display their default value if they have one. This is pretty detrimental since 1) you no longer know if you need to include that argument, and 2) there is no way of finding out what the default value is without clicking File.

Old:


New:

[Potato Uno]

Yes that has been very annoying. I have to check the old docs to figure out the optional parameters. Specifying in the docs, if not on the prototype, that a function has optional parameters (and if so, their default values) would be really nice.

[BAILOPAN]

The include files show default values. The new documentation parser can't slice out the correct text yet.

[Miu]

why do const vars have to hold constant expressions in 1.7? o.o

[BAILOPAN]

@Miu: I would be surprised if that was a regression (i.e., you should see the same behavior in 1.6) - but if it is, feel free to file a bug. Pawn treats a leading "const" as a compiler constant rather than a variable with immutable storage. So it requires that the expression evaluates at compile-time (and Pawn only supports primitive cells as constants).

The experimental v2 parser does change this, but I'm not aware that the behavior was different in earlier releases.

[Miu]

there's no regression if you think new syntax "const int var" corresponds to old syntax "const var"; I was just thinking about "new const var"

by 1.7 I meant transitional syntax btw, I don't think there's a change between the compilers

[BAILOPAN]

Ah - yeah, in Transitional Syntax I think the old behaviour is counter-intuitive. I haven't decided whether it's worth changing in spcomp yet.

[nergal]

not to hijack the current subtopic but will the keyword "any" be changed in the future?

[Potato Uno]

From a technical standpoint, is 1.7 basically the same as 1.6 except spcomp does some voodoo magic that transforms methodmap code or new syntax declarations back to the same bytecode as if the source code was written in 1.6? (If that even makes any sense - I think you get what I mean.)

[asherkin]

Originally Posted by nergal not to hijack the current subtopic but will the keyword "any" be changed in the future?

Yes, its replacement is still undecided I believe.

Originally Posted by Potato Uno From a technical standpoint, is 1.7 basically the same as 1.6 except spcomp does some voodoo magic that transforms methodmap code or new syntax declarations back to the same bytecode as if the source code was written in 1.6? (If that even makes any sense - I think you get what I mean.)

Regarding only SourcePawn, yes. (Obviously there are numerous other changes between 1.6.x and 1.7.x than the introduction of the Transitional Syntax.)