rhizome VS grammars-v4

There was actually an existing, similar in concept, Ruby JIT, Rhyzome: https://github.com/chrisseaton/rhizome.

It's good to see (IMO) the development of a JIT in a strict sense. I'm personally skeptical about the current approach (invoking a compiler separately).

On the other hand, it's important to know that JITs may take a long development time to be performant, and that they also complicate the performance profile of a virtual machine.

Lots of compiler tutorials are like this - there's very little out there to explain how compilers really work.

This is my effort - trying to show genuine data structures and processes.

https://github.com/chrisseaton/rhizome

Perhaps the one from ANTLR's collection [0] is a good start (there are also others ANTLR VB6 grammars documented elsewhere). It does require knowing ANTLR, but that should be less effort for someone already familiar with language implementation, particularly, the visitor pattern (my favorite reference [1]).

[0] https://github.com/antlr/grammars-v4/tree/master/vb6

[1] https://craftinginterpreters.com/representing-code.html

Where is the SQLite test suite, please? I'd be very interested.

There are already SQL grammars, check https://github.com/antlr/grammars-v4 specifically in here I think https://github.com/antlr/grammars-v4/tree/master/sql I contributed to one of them, and I wrote my own for some personal work. Be warned, it's very involved, very complex and MSSQL is rather ill-defined.

Names bracket identifiers) in SQL are bloody awful. Sometimes square brackets are even compulsory, and why you can usually replace [...] with the SQL standard "..." , not always! Trust me, it gets worse.

I don't find antlr grammars to be brittle, and while they can lose in performance (by how much I don't know, perhaps quite considerably) they are very easy to maintain and I am very fortunate to have antlr to work with.

This grammar "library" was cited as an example of what the format could look like:.

https://github.com/antlr/grammars-v4

There is everything from assembly and C++ to glsl and scripting languages, arithmetic, games, and other weird formats.

> Which brings me to the other approach: steering the LLM's output __as it is generating tokens__

A relevant PR:

https://github.com/ggerganov/llama.cpp/pull/1773

The plan is to support arbitrary grammar files to constrain tokens as they are generated, like the ones here:

https://github.com/antlr/grammars-v4

Have a look at jooq - I know this has been used to rewrite SQL from one dialect to another, so it MUST be capable of collating code activity metrics. Look here. Otherwise, you might want to look into writing your own parser. ANTLR has a T-SQL dialect parser script here.

Find a Perl grammar file for ANTLR, like https://github.com/antlr/grammars-v4/tree/master/perl Save the grammar file as Perl.g4 in your project. Now, you can create the Kotlin program: import org.antlr.v4.runtime.* import org.antlr.v4.runtime.tree.ParseTree import java.io.File

Antlr is another option. You could generate a parser using the JSON antlr grammar.