Is there a way to use additional parsing info for the lexer?

[kortenkamp]

Describe the solution you'd like For most languages, calling a function needs a certain, fixed number of parameters. E.g., call( 1, 2) is a different function call than call (5). It would be nice to be able to give a visual clue whether that call is actually defined – that is, allow for the information whether call is a token of type known_function or unknown_function.

Are there any workarounds? Maybe it is possible to extract that information when parsing the line somehow. I would be grateful for hints about any existing parsers that can do similar things.

Additional context A similiar problem would be to identify whether a certain identifier denotes a built-in function or a user-defined function. Again, this depends on parsing of the full file.

[bobbylight]

This is a tricky problem for a couple of reasons:

• Highlighting functions differently depending on factors you mention like built-in vs. not require knowledge of semantics of the current language, whith RSTA doesn't havea OOTB
• Highlighting is done one line at a time, and RSTA doesn't store the highlighting sttae of any other part of the file (this is different than many other code editor components and was done to minimize memory usage to facilitate supporting much larger files)

That said, it is certainly possible to do these things, but it would take legwork in the client app.

• Highlighting known/stdlib/etc. functions differently than others could be done via a custom TokenMaker and overriding addToken with logic like:

if (tokenType == TokenTypes.FUNCTION && stdlibFunctionSet.contains(new String(array, start, end))) {
  tokenType = myCustomTokenType;
}

• As for tracking functions, variables, etc. defined in the current editor, one hook that's baked into RSTA that can do this is addParser(). The passed-in Parser is called after a small delay whenever the editor's contents are modified. Parsers are designed to squiggle-underline warnings and errors, but they can also be used to build say an AST of the editor's contents, which can then be used by your TokenMaker to decide how to colorize tokens.