cgen error for using a default := p or { T{} } construct in a generic function

[spytheman]

V doctor:

V full version: V 0.4.7 f61a4f1.d5c2ebc
OS: linux, Ubuntu 20.04.6 LTS
Processor: 2 cpus, 64bit, little endian, Intel(R) Core(TM) i3-3225 CPU @ 3.30GHz

getwd: /space/v/oo
vexe: /space/v/oo/v
vexe mtime: 2024-09-04 19:52:03

vroot: OK, value: /space/v/oo
VMODULES: OK, value: /home/delian/.vmodules
VTMP: OK, value: /tmp/v_1000

Git version: git version 2.46.0
Git vroot status: weekly.2024.36-8-gd5c2ebce
.git/config present: true

CC version: cc (Ubuntu 10.5.0-1ubuntu1~20.04) 10.5.0
thirdparty/tcc status: thirdparty-linux-amd64 0134e9b9

What did you do?
./v -g -o vdbg cmd/v && ./vdbg a.v

pub fn f[T](defaults ?T) {
	default := defaults or { T{} }
	dump(default)
}	
		 
f(123)

What did you expect to see?

a compiled program

What did you see instead?

================== C compilation error (from tcc): ==============
cc: /tmp/v_1000/a.01J6Z9GC8NRPDKZDBDK9QGBFNR.tmp.c:13299: error: switch expected
=================================================================
(You can pass `-cg`, or `-show-c-output` as well, to print all the C error messages).

Note

You can use the 👍 reaction to increase the issue's priority for developers.

Please note that only the 👍 reaction to the issue itself counts as a vote.
Other reactions and those to comments will not be taken into account.

[spytheman]

Note that a very similar program works fine:

pub fn f[T](defaults ?T) {
	the_default := defaults or { T{} }
	dump(the_default)
}	
		 
f(123)

[spytheman]

Note also that default is a reserved keyword in C, and it is already in vlib/v/gen/c/cgen.v:23:const c_reserved .

Perhaps it is just a matter of escaping it, in yet another place?

[JalonSolov]

C name collision. default is used in switch statements to specify the catch-all branch (like else in V's match).

Yet another case where all V variables should've been named _v_<whatever> to avoid problems.

[JalonSolov]

We were typing at the same time, you were just faster. :-)

[medvednikov]

@JalonSolov my goal has been to generate clean readable code. Like it was written by a human.

Every variable starting with v_ would break that. I'd prefer to just handle all reserved c keywords.

[spytheman]

The reserved C keywords are a limited set, and can be handled.

However, in C programs, there can be global symbols, like function names and variable names, and that set is unlimited. We allow #include, so we get those, and we do not know what they are ahead of time, because we do not parse the C headers ourselves.

Due to that, the V names (that are also user controlled), can still clash with the global names from the C headers, that a V program includes.

The potential for conflicts for variables is minimized, because C allows shadowing of global names with ones from a local scope, and V mostly ensures that it always declares and initializes its own variables.