Jim Idle

ANTLR4 wasn't designed to process grammars written aas a straight copy from the normative spec though. Yeah, it will parse given anything, but that does not mean we should promote...

As part of MySQL Workbench, @mike-lischke has/had a v3 grammar that he wrote to use with the C target. https://dev.mysql.com/blog-archive/parsing-in-mysql-workbench/ Mike's stuff is very good. @mike-lischke (I think I have...

Actually, it looks like MySqlWorkbench now uses an official Oracle v4 grammar. I am not sure where that comes from originally, but we have this: https://github.com/mysql/mysql-workbench/tree/8.0/library/parsers/grammars

> Yes, it's the right Mike, Jim :-D > > And yes, I wrote the original MySQL ANTLR grammar (for use in MySQL Workbench), at that time for ANTLR3 and...

OK - cool. I started on a Go version (but literally just created the project and downloaded your grammars. I was just about to ask if there was any need...

I’ll do Java if I get a chance. On Tue, Mar 5, 2024 at 14:02 Mike Lischke ***@***.***> wrote: > You can ignore MRS (short for MySQL REST Service) completely....

> Test with the same mysql grammar, the java sdk is mush more faster than golang version, maybe there is some implementation difference in golang sdk? You are using this...

> > > Test with the same mysql grammar, the java sdk is mush more faster than golang version, maybe there is some implementation difference in golang sdk? > >...

> > Java SDK is much faster than golang SDK > > I'm curious how much faster. Could you perform a benchmark? It's not relevant really. I improved from essentially...

@parrt we can close this now. The issue is with the grammar, not ANTLR not the runtimes. Also, I have published: https://github.com/jimidle/tsqlparser Which @parrt analysed in a paper when it...