Function llama_grammar_accept_token

llama.cpp:8656–8680

}

void llama_grammar_accept_token(struct llama_context * ctx, struct llama_grammar * grammar, llama_token token) {
    const int64_t t_start_sample_us = ggml_time_us();

    if (token == llama_token_eos(&ctx->model)) {
        for (const auto & stack : grammar->stacks) {
            if (stack.empty()) {
                return;
            }
        }
        GGML_ASSERT(false);
    }

    const std::string piece = llama_token_to_piece(ctx, token);

    // Note terminating 0 in decoded string
    const auto   decoded     = decode_utf8(piece.c_str(), grammar->partial_utf8);
    const auto & code_points = decoded.first;
    for (auto it = code_points.begin(), end = code_points.end() - 1; it != end; ++it) {
        grammar->stacks = llama_grammar_accept(grammar->rules, grammar->stacks, *it);
    }
    grammar->partial_utf8 = decoded.second;
    GGML_ASSERT(!grammar->stacks.empty());

    ctx->t_sample_us += ggml_time_us() - t_start_sample_us;
}

//
// Beam search