thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

common/parser: fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers (#21230)

Browse source 
* Fix call ID detection (Mistral parser mostly) + atomicity for tag-json parsers

* Rename

* Update common/chat-auto-parser-generator.cpp

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

---------

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
This commit is contained in:
Piotr Wilkin (ilintar) 2026-04-03 17:51:52 +02:00 • committed by GitHub
parent af5c13841f
commit f1f793ad06
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 242 additions and 153 deletions
Download patch file Download diff file Expand all files Collapse all files

110
tools/parser/debug-template-parser.cpp
View file

@ -5,15 +5,15 @@
#include "gguf.h"
#include "jinja/runtime.h"
#include "log.h"
#include "nlohmann/json.hpp"
#include "peg-parser.h"
#include <fstream>
#include <numeric>
#include <optional>
#include <sstream>
#include <string>
#include "nlohmann/json.hpp"
#include "peg-parser.h"
using json = nlohmann::ordered_json;
enum class output_mode {
@ -34,14 +34,14 @@ enum class input_message_type {
};
struct debug_options {
std::string template_path;
bool with_tools = true;
bool generation_prompt = true;
bool enable_reasoning = true;
bool debug_jinja = false;
bool force_tool_call = false;
output_mode mode = output_mode::BOTH;
input_message_type input_message = input_message_type::NONE;
std::string template_path;
bool with_tools = true;
bool generation_prompt = true;
bool enable_reasoning = true;
bool debug_jinja = false;
bool force_tool_call = false;
output_mode mode = output_mode::BOTH;
input_message_type input_message = input_message_type::NONE;
};
static std::string read_file(const std::string & path) {
@ -274,7 +274,7 @@ static void render_scenario(const common_chat_template & tmpl,
json final_messages = messages;
if (add_generation_prompt && !messages.empty() && messages.back().value("role", "") == "assistant") {
final_messages.push_back(json{
{ "role", "user" },
{ "role", "user" },
{ "content", "Now please continue with another response." }
});
}
@ -305,7 +305,7 @@ static void render_all_scenarios(const common_chat_template & tmpl,
const json & tools,
bool add_generation_prompt,
bool enable_thinking,
input_message_type message_type) {
input_message_type message_type) {
json user_msg = build_user_message();
auto render_if = [&](input_message_type type, const std::string & name, const json & assistant_msg) {
@ -335,6 +335,24 @@ static void render_all_scenarios(const common_chat_template & tmpl,
}
}
static autoparser::generation_params prepare_params(const debug_options & opts, const json & tools) {
autoparser::generation_params params;
params.messages = json::array({ build_user_message() });
params.reasoning_format = opts.enable_reasoning ? COMMON_REASONING_FORMAT_DEEPSEEK : COMMON_REASONING_FORMAT_NONE;
params.enable_thinking = opts.enable_reasoning;
params.add_generation_prompt = opts.generation_prompt;
if (opts.with_tools) {
params.tools = tools;
params.tool_choice = opts.force_tool_call ? COMMON_CHAT_TOOL_CHOICE_REQUIRED : COMMON_CHAT_TOOL_CHOICE_AUTO;
} else {
params.tools = json();
params.tool_choice = COMMON_CHAT_TOOL_CHOICE_NONE;
}
params.parallel_tool_calls = false;
return params;
}
int main(int argc, char ** argv) {
// Set log level to most verbose to capture all debug output
common_log_set_verbosity_thold(99);
@ -369,49 +387,41 @@ int main(int argc, char ** argv) {
try {
common_chat_template chat_template(template_source, "", "");
// Build tools definition
json tools = opts.with_tools ? build_tools_definition() : json();
// Render template scenarios if requested
if (opts.input_message != input_message_type::NONE &&
(opts.mode == output_mode::TEMPLATE || opts.mode == output_mode::BOTH)) {
autoparser::generation_params params = prepare_params(opts, tools);
common_chat_params parser_data;
if (std::optional<common_chat_params> spec_tmpl =
common_chat_try_specialized_template(chat_template, template_source, params)) {
LOG_ERR("\n");
LOG_ERR("================================================================================\n");
LOG_ERR(" TEMPLATE RENDERING OUTPUT\n");
LOG_ERR("================================================================================\n");
LOG_ERR("This template uses a specialized parser, analysis results will not be available.");
parser_data = *spec_tmpl;
} else {
// Render template scenarios if requested
if (opts.input_message != input_message_type::NONE &&
(opts.mode == output_mode::TEMPLATE || opts.mode == output_mode::BOTH)) {
LOG_ERR("\n");
LOG_ERR("================================================================================\n");
LOG_ERR(" TEMPLATE RENDERING OUTPUT\n");
LOG_ERR("================================================================================\n");
render_all_scenarios(chat_template, tools, opts.generation_prompt, opts.enable_reasoning,
opts.input_message);
}
// Output analysis if requested
if (opts.mode == output_mode::ANALYSIS || opts.mode == output_mode::BOTH) {
LOG_ERR("\n");
LOG_ERR("================================================================================\n");
LOG_ERR(" TEMPLATE ANALYSIS\n");
LOG_ERR("================================================================================\n");
autoparser::autoparser analysis;
analysis.analyze_template(chat_template);
// Generate Parser
autoparser::generation_params params;
params.messages = json::array({ build_user_message() });
params.reasoning_format =
opts.enable_reasoning ? COMMON_REASONING_FORMAT_DEEPSEEK : COMMON_REASONING_FORMAT_NONE;
params.enable_thinking = opts.enable_reasoning;
params.add_generation_prompt = opts.generation_prompt;
if (opts.with_tools) {
params.tools = tools;
params.tool_choice = opts.force_tool_call ? COMMON_CHAT_TOOL_CHOICE_REQUIRED : COMMON_CHAT_TOOL_CHOICE_AUTO;
} else {
params.tools = json();
params.tool_choice = COMMON_CHAT_TOOL_CHOICE_NONE;
render_all_scenarios(chat_template, tools, opts.generation_prompt, opts.enable_reasoning,
opts.input_message);
}
params.parallel_tool_calls = false;
auto parser_data = autoparser::peg_generator::generate_parser(chat_template, params, analysis);
// Output analysis if requested
if (opts.mode == output_mode::ANALYSIS || opts.mode == output_mode::BOTH) {
LOG_ERR("\n");
LOG_ERR("================================================================================\n");
LOG_ERR(" TEMPLATE ANALYSIS\n");
LOG_ERR("================================================================================\n");
autoparser::autoparser analysis;
analysis.analyze_template(chat_template);
// Generate Parser
parser_data = autoparser::peg_generator::generate_parser(chat_template, params, analysis);
}
LOG_ERR("\n=== Generated Parser ===\n");
common_peg_arena arena;