Can I combine captures in Sublime Text syntax? - sublimetext3

Question

I'm writing the syntax for V which defines methods with a similar syntax to Go, where:
fn (o MyStruct) my_function(a int) {
// ...
}
I might use something like (I will break it down into push, but just for brevity):
variables:
ident: \b[A-Za-z_][A-Za-z_0-9]*\b
contexts:
fn:
- match: (fn)\s*\({{ident}}\s*({{ident}})\)\s*({{ident}})
captures:
1: keyword
2: entity.name.type.v
3: entity.name.function.v
But the problem is MyStruct and my_function are indexed separately, so methods of the same name (str() is a good example) will not be distinct to the indexer. Is there a way I can combine them into a single entity.name.function.v of value MyStruct.my_function?
I know I could treat the whole definition as the entity, but that's too verbose and won't work when split across multiple lines:
captures:
0: entity.name.function.v

Answer 1

If you apply an unbroken/continuous scope to the text, you can target that in a .tmPreferences file (as opposed to the default entity.name scope) and index that, and make use of a "symbol index transformation" to remove the ) and spaces.
- match: (fn)\s*\({{ident}}\s*(({{ident}})\)\s*({{ident}}))
captures:
1: keyword
2: meta.indexed-unit.v
3: entity.name.type.v
4: entity.name.function.v
Now the scope meta.indexed-unit.v would apply to the text MyStruct) my_function.
A symbol index transformation is basically a regex replacement which applies to the indexed symbols:
https://docs.sublimetext.io/reference/symbols.html#settings-subelements
So your .tmPreferences file might look something like this:
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple Computer//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
<key>scope</key>
<string>source.v meta.indexed-unit.v</string>
<key>settings</key>
<dict>
<key>showInIndexedSymbolList</key>
<string>1</string>
<key>symbolIndexTransformation</key>
<string>
s/\)\s*//;
</string>
</dict>
</dict>
</plist>