Add support for C-style alias declarations

[wilzbach]

In Phobos std.typecons and std.traits use this alias syntax:

 alias StorageClasses_opt BasicType FuncDeclarator ;

FuncDeclarator:
    BasicType2opt Identifier FuncDeclaratorSuffix

As AFAICT there's no alternative syntax to this, replacing the few uses sadly doesn't work.
Hence, it would be very appreciated if Dscanner (and thus libdparse) could parse these declarations at least to the degree of not throwing a hard error ;-)
I gave this a shot and it looks like it's very hard to support.

References:

• https://dlang.org/spec/declaration.html#alias
• https://dlang.org/spec/function.html#FuncDeclarator

[wilzbach]

See also: https://issues.dlang.org/show_bug.cgi?id=12299

[codecov-io]

Codecov Report

Merging #150 into master will increase coverage by 0.06%.
The diff coverage is 83.33%.

@@            Coverage Diff             @@
##           master     #150      +/-   ##
==========================================
+ Coverage   81.47%   81.53%   +0.06%     
==========================================
  Files           7        7              
  Lines        4528     4534       +6     
==========================================
+ Hits         3689     3697       +8     
+ Misses        839      837       -2

Impacted Files	Coverage Δ
src/dparse/ast.d	15.14% <ø> (ø)	⬆️
src/dparse/parser.d	87.56% <83.33%> (+0.09%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 5b009b7...06855ee. Read the comment docs.

[ghost]

You forget to visit the parameters

[wilzbach]

My first PR to the parser. Thanks!! (added)

[ghost]

You forget to visit the memberFunctionAttributes

[Hackerpilot]

Looks like this is fixed.

[ghost]

As AFAICT there's no alternative syntax to this

Indeed. I've opend an issue about this https://issues.dlang.org/show_bug.cgi?id=16020 but i think that the only resolution possible is a personal initiative or a DIP. In the absence of the required language change supporting this in libdparse sounds like the right choice.

[wilzbach]

• Misses 4204 4210 +6

That stroke me as quite odd -> #151

[wilzbach]

That stroke me as quite odd -> #151

Thanks a lot for the quick approval!
(Rebased to get a proper coverage report.)

[wilzbach]

Okay @bbasile - how do we proceed here? Does this need approval from @Hackerpilot or can we move ahead?

[ghost]

Does this need approval from @Hackerpilot

I've explained why i agree but initially not to parse them was a clearly a choice, so yes.

[Hackerpilot]

I initially left support for this out as I thought

1. That the syntax would be deprecated / removed
2. That the new syntax was sufficient

[wilzbach]

I initially left support for this out as I thought
That the syntax would be deprecated / removed
That the new syntax was sufficient

Hehe, I think we all were/are still confused by this oversight...

So this means this PR is good to go? :)
I would love to get rid of the exclusion of std.traits and std.typecons.d in the Phobos build .

[Hackerpilot]

Not sure if you noticed, but I clicked the "approve" button in Github.

[wilzbach]

Not sure if you noticed, but I clicked the "approve" button in Github.

I did, but I was just wondering whether there was a reason why you didn't hit merge afterwards ;-)
Thanks a lot!