grammar.hpp File Reference

Grammar-Constrained Sampling. More...

#include "common.hpp"
#include "tokenizer.hpp"
#include <llama/llama.h>
#include <json-schema-to-grammar.h>
#include <nlohmann/json.hpp>
#include <stdexcept>
#include <string>
#include <vector>

Go to the source code of this file.

Namespaces
namespace	lloyal
	Boundary Tracker Stub for OSS liblloyal.
namespace	lloyal::grammar

Functions
std::string	lloyal::grammar::from_json_schema (const std::string &schema_json)
	Convert JSON schema to GBNF (Grammar BNF) format.
llama_sampler *	lloyal::grammar::init_sampler (const llama_model *model, const std::string &grammar_str, const std::string &root_rule="root")
	Initialize a grammar sampler from GBNF grammar string.
llama_sampler *	lloyal::grammar::init_lazy_sampler (const llama_model *model, const std::string &grammar_str, const std::vector< std::string > &trigger_patterns, const std::vector< llama_token > &trigger_tokens, const std::string &root_rule="root")
	Initialize a lazy grammar sampler from GBNF grammar string.
llama_sampler *	lloyal::grammar::clone_sampler (llama_sampler *smpl)
	Clone a grammar sampler (for fork/branching).
void	lloyal::grammar::free_sampler (llama_sampler *smpl)
	Free a grammar sampler.
void	lloyal::grammar::apply (llama_sampler *smpl, llama_token_data_array *cur_p)
	Apply grammar constraint to candidates.
void	lloyal::grammar::accept (llama_sampler *smpl, llama_token token)
	Accept a token into grammar state.

Detailed Description

Grammar-Constrained Sampling.

Provides JSON schema to GBNF grammar conversion for structured output generation. Uses json_schema_to_grammar() from llama.cpp's common library.

Architecture:

• Calls json_schema_to_grammar() from common/json-schema-to-grammar.h
• Adds error handling, logging, and consistent API
• Manages grammar sampler lifecycle

Definition in file grammar.hpp.