ANTLR is really only good for the Java ecosystem. The quality of the generated C/C++ code is low and it has a large number of dependencies. C++ development also lagged pretty far behind Java when ANTLR 4 was released.
I don't know of any "real" language written in C or C++ that uses ANTLR.
The state of the art for production languages in C/C++ is either hand-written parsers or yacc/bison. I've never even seen flex used, and I've looked at over 20 lexer/parser implementations for "real" languages.
Java, C#, Go, Python, JavaScript, Swift and C++ although everything gets started with Java and filled out to the other targets.
You may be thinking of ANTRL 3. I really didn't much care for the ANTLR 3 C target but I use ANTRL 4 and its C target. It is really much much better than ANTRL 3. TParr says ANTLR 4 is the parser generator he always wanted to make.
However, I'm not still pitching ANTRL as a parser for a real compiler. ANTRL rules the world for little languages, domain specific languages but a production language would almost certainly would use a hand crafted recursive decent compiler.
This is a learning project. So I'd strongly recommend anyone learn modern tools and methods and learn those. bison is C++ native with experimental Java support and nothing else. flex is C and C++ only. Not a fan.
I don't know of any "real" language written in C or C++ that uses ANTLR.
The state of the art for production languages in C/C++ is either hand-written parsers or yacc/bison. I've never even seen flex used, and I've looked at over 20 lexer/parser implementations for "real" languages.