Upgrade to version 0.3.0

new QML Chat
Qwen chat model
This commit is contained in:
Petr Mironychev
2024-10-14 00:09:08 +02:00
parent 5c98de7440
commit 0c045e65df
71 changed files with 1920 additions and 1070 deletions

View File

@ -23,13 +23,13 @@
#include <QNetworkAccessManager>
#include <QNetworkReply>
#include <llmcore/RequestConfig.hpp>
#include <texteditor/textdocument.h>
#include "DocumentContextReader.hpp"
#include "LLMProvidersManager.hpp"
#include "PromptTemplateManager.hpp"
#include "QodeAssistUtils.hpp"
#include "core/LLMRequestConfig.hpp"
#include "logger/Logger.hpp"
#include "llmcore/PromptTemplateManager.hpp"
#include "llmcore/ProvidersManager.hpp"
#include "settings/ContextSettings.hpp"
#include "settings/GeneralSettings.hpp"
@ -39,7 +39,7 @@ LLMClientInterface::LLMClientInterface()
: m_requestHandler(this)
{
connect(&m_requestHandler,
&LLMRequestHandler::completionReceived,
&LLMCore::RequestHandler::completionReceived,
this,
&LLMClientInterface::sendCompletionToClient);
}
@ -80,7 +80,7 @@ void LLMClientInterface::sendData(const QByteArray &data)
} else if (method == "exit") {
// TODO make exit handler
} else {
logMessage(QString("Unknown method: %1").arg(method));
LOG_MESSAGE(QString("Unknown method: %1").arg(method));
}
}
@ -88,9 +88,9 @@ void LLMClientInterface::handleCancelRequest(const QJsonObject &request)
{
QString id = request["params"].toObject()["id"].toString();
if (m_requestHandler.cancelRequest(id)) {
logMessage(QString("Request %1 cancelled successfully").arg(id));
LOG_MESSAGE(QString("Request %1 cancelled successfully").arg(id));
} else {
logMessage(QString("Request %1 not found").arg(id));
LOG_MESSAGE(QString("Request %1 not found").arg(id));
}
}
@ -148,10 +148,10 @@ void LLMClientInterface::handleCompletion(const QJsonObject &request)
{
auto updatedContext = prepareContext(request);
LLMConfig config;
config.requestType = RequestType::Fim;
config.provider = LLMProvidersManager::instance().getCurrentFimProvider();
config.promptTemplate = PromptTemplateManager::instance().getCurrentFimTemplate();
LLMCore::LLMConfig config;
config.requestType = LLMCore::RequestType::Fim;
config.provider = LLMCore::ProvidersManager::instance().getCurrentFimProvider();
config.promptTemplate = LLMCore::PromptTemplateManager::instance().getCurrentFimTemplate();
config.url = QUrl(QString("%1%2").arg(Settings::generalSettings().url(),
Settings::generalSettings().endPoint()));
@ -159,18 +159,19 @@ void LLMClientInterface::handleCompletion(const QJsonObject &request)
{"stream", true},
{"stop",
QJsonArray::fromStringList(config.promptTemplate->stopWords())}};
config.multiLineCompletion = Settings::generalSettings().multiLineCompletion();
if (Settings::contextSettings().useSpecificInstructions())
config.providerRequest["system"] = Settings::contextSettings().specificInstractions();
if (Settings::contextSettings().useSystemPrompt())
config.providerRequest["system"] = Settings::contextSettings().systemPrompt();
config.promptTemplate->prepareRequest(config.providerRequest, updatedContext);
config.provider->prepareRequest(config.providerRequest);
config.provider->prepareRequest(config.providerRequest, LLMCore::RequestType::Fim);
m_requestHandler.sendLLMRequest(config, request);
}
ContextData LLMClientInterface::prepareContext(const QJsonObject &request,
const QStringView &accumulatedCompletion)
LLMCore::ContextData LLMClientInterface::prepareContext(const QJsonObject &request,
const QStringView &accumulatedCompletion)
{
QJsonObject params = request["params"].toObject();
QJsonObject doc = params["doc"].toObject();
@ -182,8 +183,8 @@ ContextData LLMClientInterface::prepareContext(const QJsonObject &request,
filePath);
if (!textDocument) {
logMessage("Error: Document is not available for" + filePath.toString());
return ContextData{};
LOG_MESSAGE("Error: Document is not available for" + filePath.toString());
return LLMCore::ContextData{};
}
int cursorPosition = position["character"].toInt();
@ -218,11 +219,11 @@ void LLMClientInterface::sendCompletionToClient(const QString &completion,
result[LanguageServerProtocol::isIncompleteKey] = !isComplete;
response[LanguageServerProtocol::resultKey] = result;
logMessage(
LOG_MESSAGE(
QString("Completions: \n%1")
.arg(QString::fromUtf8(QJsonDocument(completions).toJson(QJsonDocument::Indented))));
logMessage(QString("Full response: \n%1")
LOG_MESSAGE(QString("Full response: \n%1")
.arg(QString::fromUtf8(QJsonDocument(response).toJson(QJsonDocument::Indented))));
QString requestId = request["id"].toString();
@ -250,7 +251,7 @@ void LLMClientInterface::logPerformance(const QString &requestId,
const QString &operation,
qint64 elapsedMs)
{
logMessage(QString("Performance: %1 %2 took %3 ms").arg(requestId, operation).arg(elapsedMs));
LOG_MESSAGE(QString("Performance: %1 %2 took %3 ms").arg(requestId, operation).arg(elapsedMs));
}
void LLMClientInterface::parseCurrentMessage() {}