MRCI/src/cmd_executor.cpp

822 lines
25 KiB
C++
Raw Normal View History

2019-09-06 23:43:07 -04:00
#include "cmd_executor.h"
// This file is part of MRCI.
// MRCI is free software: you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
// MRCI is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
// You should have received a copy of the GNU General Public License
// along with MRCI under the LICENSE.md file. If not, see
// <http://www.gnu.org/licenses/>.
CmdExecutor::CmdExecutor(RWSharedObjs *rwShare, SharedObjs *rdOnlyShare, QSharedMemory *debugInfo, QObject *parent) : QObject(parent)
{
rwShare->commands = &commands;
rwShare->activeLoopCmds = &activeLoopCmds;
rwShare->pausedCmds = &pausedCmds;
rwShare->moreInputCmds = &moreInputCmds;
rwShare->cmdNames = &cmdNames;
rdOnlyShare->activeLoopCmds = &activeLoopCmds;
rdOnlyShare->pausedCmds = &pausedCmds;
rdOnlyShare->moreInputCmds = &moreInputCmds;
rdOnlyShare->cmdNames = &cmdNames;
loopIndex = 0;
exeDebugInfo = debugInfo;
rdSharedObjs = rdOnlyShare;
rwSharedObjs = rwShare;
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
internalCmds = nullptr;
2019-09-06 23:43:07 -04:00
connect(this, &CmdExecutor::loop, this, &CmdExecutor::exeCmd);
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
void CmdExecutor::buildCmdLoaders()
{
internalCmds = new InternalCommandLoader(rwSharedObjs, this);
cmdLoaders.insert(INTERN_MOD_NAME, internalCmds);
Query db(this);
db.setType(Query::PULL, TABLE_MODULES);
db.addColumn(COLUMN_MOD_NAME);
db.addCondition(COLUMN_LOCKED, false);
db.exec();
for (int i = 0; i < db.rows(); ++i)
{
loadModFile(db.getData(COLUMN_MOD_NAME, i).toString());
}
}
2019-09-06 23:43:07 -04:00
void CmdExecutor::wrCrashDebugInfo(const QString &msg)
{
if (exeDebugInfo->isAttached())
{
exeDebugInfo->lock();
QByteArray data = toTEXT(msg).leftJustified(EXE_DEBUG_INFO_SIZE, static_cast<char>(0), true);
memcpy(exeDebugInfo->data(), data.data(), EXE_DEBUG_INFO_SIZE);
exeDebugInfo->unlock();
}
}
void CmdExecutor::connectExternCmd(ExternCommand *cmd, quint16 cmdId, const QString &cmdName)
{
wrCrashDebugInfo(" exe func: connectExternCmd()\n cmd id: " + QString::number(cmdId) + "\n cmd name: " + cmdName);
auto *cmdOutput = new CommandOutput(cmd);
connect(cmdOutput, &CommandOutput::dataOut, this, &CmdExecutor::externDataToIPC);
connectCommon(cmd, cmdOutput, cmdId, cmdName);
}
void CmdExecutor::connectInternCmd(InternCommand *cmd, quint16 cmdId, const QString &cmdName)
{
wrCrashDebugInfo(" exe func: connectInternCmd()\n cmd id: " + QString::number(cmdId) + "\n cmd name: " + cmdName);
auto *cmdOutput = new CommandOutput(cmd);
connect(cmdOutput, &CommandOutput::dataOut, this, &CmdExecutor::internDataToIPC);
connect(cmd, &InternCommand::backendDataOut, this, &CmdExecutor::backendFromCmd);
connect(cmd, &InternCommand::authOk, this, &CmdExecutor::authOk);
connect(cmd, &InternCommand::termAllCommands, this, &CmdExecutor::termAllCommands);
connect(cmd, &InternCommand::termCommandId, this, &CmdExecutor::termCommandId);
connect(cmd, &InternCommand::reloadCommands, this, &CmdExecutor::buildCommands);
connectCommon(cmd, cmdOutput, cmdId, cmdName);
}
void CmdExecutor::connectCommon(ExternCommand *cmd, CommandOutput *cmdOutput, quint16 cmdId, const QString &cmdName)
{
wrCrashDebugInfo(" exe func: connectCommon()\n cmd id: " + QString::number(cmdId) + "\n cmd name: " + cmdName);
connect(cmdOutput, &CommandOutput::closeChById, this, &CmdExecutor::closeChById);
connect(cmdOutput, &CommandOutput::openChById, this, &CmdExecutor::openChById);
connect(cmdOutput, &CommandOutput::closeChByName, this, &CmdExecutor::closeChByName);
connect(cmdOutput, &CommandOutput::openChByName, this, &CmdExecutor::openChByName);
connect(cmdOutput, &CommandOutput::cmdFinished, this, &CmdExecutor::commandFinished);
connect(cmdOutput, &CommandOutput::enableLoop, this, &CmdExecutor::enableLoop);
connect(cmdOutput, &CommandOutput::enableMoreInput, this, &CmdExecutor::enableMoreInput);
connect(cmd, &ExternCommand::closeChById, cmdOutput, &CommandOutput::closeChIdFromCmdObj);
connect(cmd, &ExternCommand::openChById, cmdOutput, &CommandOutput::openChIdFromCmdObj);
connect(cmd, &ExternCommand::closeChByName, cmdOutput, &CommandOutput::closeChNameFromCmdObj);
connect(cmd, &ExternCommand::openChByName, cmdOutput, &CommandOutput::openChNameFromCmdObj);
connect(cmd, &ExternCommand::cmdFinished, cmdOutput, &CommandOutput::finished);
connect(cmd, &ExternCommand::enableLoop, cmdOutput, &CommandOutput::enableLoopFromCmdObj);
connect(cmd, &ExternCommand::enableMoreInput, cmdOutput, &CommandOutput::enableMoreInputFromCmdObj);
connect(cmd, &ExternCommand::dataToClient, cmdOutput, &CommandOutput::dataFromCmdObj);
connect(cmd, &ExternCommand::castToPeers, this, &CmdExecutor::castToPeers);
connect(cmd, &ExternCommand::toPeer, this, &CmdExecutor::toPeer);
connect(cmd, &ExternCommand::closeSession, this, &CmdExecutor::endSession);
connect(cmd, &ExternCommand::logout, this, &CmdExecutor::logout);
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
cmd->cmdId = cmdId;
cmd->inMoreInputMode = false;
cmd->inLoopMode = false;
cmd->errSent = false;
2019-09-06 23:43:07 -04:00
cmd->setObjectName(cmdName);
cmdOutput->setCmdId(cmdId);
}
void CmdExecutor::enableLoop(quint16 cmdId, bool state)
{
if (state)
{
uniqueAdd(cmdId, activeLoopCmds);
}
else
{
activeLoopCmds.removeAll(cmdId);
}
}
void CmdExecutor::enableMoreInput(quint16 cmdId, bool state)
{
if (state)
{
uniqueAdd(cmdId, moreInputCmds);
}
else
{
moreInputCmds.removeAll(cmdId);
}
}
void CmdExecutor::preExe(ExternCommand *cmdObj, quint16 cmdId)
{
cmdObj->cmdId = cmdId;
cmdObj->errSent = false;
}
void CmdExecutor::preExe(const QList<ExternCommand *> &cmdObjs, quint16 cmdId)
{
for (auto cmdObj : cmdObjs)
{
preExe(cmdObj, cmdId);
}
}
void CmdExecutor::exeCmd(quint16 cmdId, const QByteArray &data, uchar typeId)
{
wrCrashDebugInfo(" exe func: exeCmd()\n cmd id: " + QString::number(cmdId) + "\n type id: " + QString::number(typeId));
if (!commands.contains(cmdId))
{
emit dataToSession(cmdId, toTEXT("err: The requested command id: '" + QString::number(cmdId) + "' does not exists.\n"), ERR);
emit dataToSession(cmdId, QByteArray(), IDLE);
}
else if (!pausedCmds.contains(cmdId))
{
ExternCommand *cmdObj = commands[cmdId];
preExe(cmdObj, cmdId);
preExe(cmdObj->internCommands.values(), cmdId);
cmdObj->procBin(rdSharedObjs, data, typeId);
if (!activeLoopCmds.contains(cmdId) && !moreInputCmds.contains(cmdId))
{
emit cmdObj->cmdFinished();
}
nextLoopCmd();
}
}
void CmdExecutor::nextLoopCmd()
{
if (!activeLoopCmds.isEmpty())
{
if (loopIndex == activeLoopCmds.size())
{
loopIndex = 0;
}
wrCrashDebugInfo(" exe func: nextLoopCmd()\n loop index: " + QString::number(loopIndex));
emit loop(activeLoopCmds[loopIndex++], QByteArray(), TEXT);
}
}
void CmdExecutor::termCommandObj(quint16 cmdId, ExternCommand *cmd, bool del)
{
wrCrashDebugInfo(" exe func: termCommandObj()\n cmd id: " + QString::number(cmdId));
if (moreInputCmds.contains(cmdId) || activeLoopCmds.contains(cmdId))
{
cmd->term();
cmd->inLoopMode = false;
cmd->inMoreInputMode = false;
for (auto internObj : cmd->internCommands.values())
{
if (internObj->inLoopMode || internObj->inMoreInputMode)
{
internObj->term();
internObj->inLoopMode = false;
internObj->inMoreInputMode = false;
}
}
emit dataToSession(cmdId, QByteArray(), IDLE);
}
moreInputCmds.removeAll(cmdId);
activeLoopCmds.removeAll(cmdId);
pausedCmds.removeAll(cmdId);
if (del)
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
wrCrashDebugInfo(" exe func: termCommandObj()\n cmd id: " + QString::number(cmdId) + "\n note: deleting intern commands");
2019-09-06 23:43:07 -04:00
for (auto internObj : cmd->internCommands.values())
{
internObj->aboutToDelete();
internObj->deleteLater();
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
wrCrashDebugInfo(" exe func: termCommandObj()\n cmd id: " + QString::number(cmdId) + "\n note: calling the command object's aboutToDelete()");
2019-09-06 23:43:07 -04:00
cmd->aboutToDelete();
cmd->deleteLater();
commands.remove(cmdId);
cmdNames.remove(cmdId);
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
for (auto&& list : cmdIdsByModName.values())
{
if (list.contains(cmdId))
{
list.removeAll(cmdId);
break;
}
}
2019-09-06 23:43:07 -04:00
emit dataToSession(ASYNC_RM_CMD, wrInt(cmdId, 16), CMD_ID);
}
}
void CmdExecutor::termCommandId(quint16 cmdId)
{
if (commands.contains(cmdId))
{
termCommandObj(cmdId, commands[cmdId]);
}
}
void CmdExecutor::termCommandsInList(const QList<quint16> &cmds, bool del)
{
for (auto&& cmdId : cmds)
{
termCommandObj(cmdId, commands[cmdId], del);
}
}
void CmdExecutor::termAllCommands()
{
termCommandsInList(moreInputCmds, false);
termCommandsInList(activeLoopCmds, false);
termCommandsInList(pausedCmds, false);
}
void CmdExecutor::close()
{
termCommandsInList(commands.keys(), true);
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
for (auto cmdLoader : cmdLoaders.values())
2019-09-06 23:43:07 -04:00
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
cmdLoader->aboutToDelete();
cmdLoader->deleteLater();
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
}
2019-09-06 23:43:07 -04:00
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
for (auto plugin : plugins.values())
{
plugin->unload();
plugin->deleteLater();
2019-09-06 23:43:07 -04:00
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
cleanupDbConnection();
emit okToDelete();
2019-09-06 23:43:07 -04:00
}
void CmdExecutor::commandFinished(quint16 cmdId)
{
emit dataToSession(cmdId, QByteArray(), IDLE);
moreInputCmds.removeAll(cmdId);
activeLoopCmds.removeAll(cmdId);
pausedCmds.removeAll(cmdId);
}
QString CmdExecutor::makeCmdUnique(const QString &name)
{
QString strNum;
QStringList names = cmdNames.values();
for (int j = 1; names.contains(QString(name + strNum).toLower()); ++j)
{
strNum = "_" + QString::number(j);
}
return QString(name + strNum).toLower();
}
void CmdExecutor::procInternRequest(ExternCommand *cmd, quint16 cmdId, const QString &cmdName)
{
wrCrashDebugInfo(" exe func: procInternRequest()\n cmd id: " + QString::number(cmdId));
QStringList internCmdNames = internalCmds->cmdList();
for (auto&& reqCmdName : cmd->internRequest())
{
if (internCmdNames.contains(reqCmdName, Qt::CaseInsensitive))
{
InternCommand *cmdObj = internalCmds->cmdObj(reqCmdName);
if (cmdObj != nullptr)
{
connectInternCmd(cmdObj, cmdId, cmdName);
cmdObj->setParent(cmd);
cmd->internCommands.insert(reqCmdName, cmdObj);
}
}
}
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
quint16 CmdExecutor::getModIdOffs(const QString &name)
{
if (name == INTERN_MOD_NAME)
{
return MAX_CMDS_PER_MOD;
}
else
{
Query db(this);
db.setType(Query::PULL, TABLE_MODULES);
db.addColumn(COLUMN_CMD_ID_OFFS);
db.addCondition(COLUMN_MOD_NAME, name);
db.exec();
return static_cast<quint16>(db.getData(COLUMN_CMD_ID_OFFS).toUInt());
}
}
QString CmdExecutor::getModFile(const QString &modName)
2019-09-06 23:43:07 -04:00
{
Query db(this);
db.setType(Query::PULL, TABLE_MODULES);
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
db.addColumn(COLUMN_MOD_MAIN);
db.addCondition(COLUMN_MOD_NAME, modName);
2019-09-06 23:43:07 -04:00
db.exec();
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
return db.getData(COLUMN_MOD_MAIN).toString();
2019-09-06 23:43:07 -04:00
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
void CmdExecutor::loadModFile(const QString &modName)
{
bool modOk = false;
QStringList ver = QCoreApplication::applicationVersion().split('.');
QString mainFilePath = getModFile(modName);
QString path = QFileInfo(mainFilePath).path();
auto *lib = new QLibrary(mainFilePath, this);
2019-09-06 23:43:07 -04:00
lib->load();
ModImportFunc importFunc = reinterpret_cast<ModImportFunc>(lib->resolve(MOD_IMPORT_FUNC));
if (!importFunc)
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
{
qDebug() << "CmdExecutor::loadModFile() err: failed to load mod lib file: " << mainFilePath << " reason: " << lib->errorString();
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
}
else
{
wrCrashDebugInfo(" exe func: loadModFile()\n path: " + mainFilePath + " \nmod name: " + modName + " \nnote: calling the module's import function.");
CommandLoader *cmdLoader = importFunc();
2019-09-06 23:43:07 -04:00
wrCrashDebugInfo(" exe func: loadModFile()\n path: " + mainFilePath + " \nmod name: " + modName + " \nnote: running import rev negotiations.");
2019-09-06 23:43:07 -04:00
if (!cmdLoader)
{
qDebug() << "CmdExecutor::loadModFile() err: failed to load mod lib file: " << mainFilePath << " reason: the CommandLoader object returned by the import function is null.";
}
else if (cmdLoader->rev() < IMPORT_REV)
{
qDebug() << "CmdExecutor::loadModFile() err: failed to load mod lib file: " << mainFilePath << " reason: module import rev " << cmdLoader->rev() << " not compatible with host rev " << IMPORT_REV << ".";
}
else if (!cmdLoader->hostRevOk(IMPORT_REV, ver[0].toUShort(), ver[1].toUShort(), ver[2].toUShort()))
{
qDebug() << "CmdExecutor::loadModFile() err: failed to load mod lib file: " << mainFilePath << " the module rejected the host rev/version. reason: " << cmdLoader->lastError() << ".";
}
else
{
modOk = true;
wrCrashDebugInfo(" exe func: loadModFile()\n path: " + mainFilePath + " \nmod name: " + modName + " \nnote: calling the module's modPath() function.");
cmdLoader->modPath(path);
2019-09-06 23:43:07 -04:00
cmdLoaders.insert(modName, cmdLoader);
plugins.insert(modName, lib);
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
}
2019-09-06 23:43:07 -04:00
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
if (!modOk)
{
lib->unload();
lib->deleteLater();
2019-09-06 23:43:07 -04:00
}
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
void CmdExecutor::unloadModFile(const QString &modName)
2019-09-06 23:43:07 -04:00
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
if (cmdIdsByModName.contains(modName))
2019-09-06 23:43:07 -04:00
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
termCommandsInList(cmdIdsByModName[modName], true);
2019-09-06 23:43:07 -04:00
wrCrashDebugInfo(" exe func: unloadModFile()\n mod name: " + modName + "\n note: calling the modules's aboutToDelete().");
2019-09-06 23:43:07 -04:00
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
cmdLoaders[modName]->aboutToDelete();
cmdLoaders[modName]->deleteLater();
wrCrashDebugInfo(" exe func: unloadModFile()\n mod name: " + modName + "\n note: calling the modules's library unload function.");
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
plugins[modName]->unload();
plugins[modName]->deleteLater();
2019-09-06 23:43:07 -04:00
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
cmdIdsByModName.remove(modName);
cmdLoaders.remove(modName);
plugins.remove(modName);
2019-09-06 23:43:07 -04:00
}
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
void CmdExecutor::addCommandToList(quint16 cmdId, const QString &cmdName, const QString &modName, ExternCommand *cmdObj)
2019-09-06 23:43:07 -04:00
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
procInternRequest(cmdObj, cmdId, cmdName);
2019-09-06 23:43:07 -04:00
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
commands.insert(cmdId, cmdObj);
cmdNames.insert(cmdId, cmdName);
2019-09-06 23:43:07 -04:00
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
if (cmdIdsByModName.contains(modName))
2019-09-06 23:43:07 -04:00
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
cmdIdsByModName[modName].append(cmdId);
2019-09-06 23:43:07 -04:00
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
else
2019-09-06 23:43:07 -04:00
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
QList<quint16> list;
list.append(cmdId);
cmdIdsByModName.insert(modName, list);
2019-09-06 23:43:07 -04:00
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
emit dataToSession(ASYNC_ADD_CMD, toNEW_CMD(cmdId, cmdName, cmdObj), NEW_CMD);
2019-09-06 23:43:07 -04:00
}
bool CmdExecutor::allowCmdLoad(const QString &cmdName, const QString &modName, const QStringList &exemptList)
{
bool ret = false;
if (exemptList.contains(cmdName, Qt::CaseInsensitive))
{
ret = true;
}
else
{
Query db(this);
db.setType(Query::PULL, TABLE_CMD_RANKS);
db.addColumn(COLUMN_HOST_RANK);
db.addCondition(COLUMN_COMMAND, cmdName);
db.addCondition(COLUMN_MOD_NAME, modName);
db.exec();
if (db.rows())
{
uint cmdRank = db.getData(COLUMN_HOST_RANK).toUInt();
if (cmdRank >= *rdSharedObjs->hostRank)
{
ret = true;
}
}
else if (*rdSharedObjs->hostRank == 1)
{
ret = true;
}
}
return ret;
}
void CmdExecutor::loadInternCmd(CommandLoader *loader, const QString &cmdName, const QString &uniqueName, quint16 id)
{
auto *cmdObj = reinterpret_cast<InternCommand*>(loader->cmdObj(cmdName));
if (cmdObj != nullptr)
{
connectInternCmd(cmdObj, id, uniqueName);
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
addCommandToList(id, uniqueName, INTERN_MOD_NAME, cmdObj);
2019-09-06 23:43:07 -04:00
}
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
void CmdExecutor::loadExternCmd(CommandLoader *loader, const QString &modName, const QString &cmdName, const QString &uniqueName, quint16 id)
2019-09-06 23:43:07 -04:00
{
ExternCommand *cmdObj = loader->cmdObj(cmdName);
if (cmdObj != nullptr)
{
connectExternCmd(cmdObj, id, uniqueName);
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
addCommandToList(id, uniqueName, modName, cmdObj);
2019-09-06 23:43:07 -04:00
}
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
void CmdExecutor::loadCmd(CommandLoader *loader, quint16 cmdId, const QString &modName, const QString &cmdName, const QString &uniqueCmdName)
2019-09-06 23:43:07 -04:00
{
wrCrashDebugInfo(" exe func: loadCmd()\n cmd id: " + QString::number(cmdId) + "\n cmd name: " + cmdName);
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
if (modName == INTERN_MOD_NAME)
2019-09-06 23:43:07 -04:00
{
loadInternCmd(loader, cmdName, uniqueCmdName, cmdId);
}
else
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
loadExternCmd(loader, modName, cmdName, uniqueCmdName, cmdId);
2019-09-06 23:43:07 -04:00
}
}
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
void CmdExecutor::loadCmds(CommandLoader *loader, quint16 idOffs, const QString &modName)
2019-09-06 23:43:07 -04:00
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
wrCrashDebugInfo(" exe func: loadCmds()\n mod name: " + modName + "\n id offs: " + QString::number(idOffs));
2019-09-06 23:43:07 -04:00
QStringList list = loader->cmdList();
QStringList pub = loader->pubCmdList();
QStringList exempt = loader->rankExemptList();
list.sort(Qt::CaseInsensitive);
for (quint16 id = 0; (id < list.size()) && (id < MAX_CMDS_PER_MOD); ++id)
{
QString unique = makeCmdUnique(list[id]);
quint16 cmdId = idOffs + id;
if (validCommandName(unique))
{
if (commands.contains(cmdId))
{
if (!allowCmdLoad(list[id], modName, exempt))
{
termCommandObj(cmdId, commands[cmdId], true);
}
}
else
{
if (rdSharedObjs->userName->isEmpty())
{
if (pub.contains(list[id], Qt::CaseInsensitive))
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
loadCmd(loader, cmdId, modName, list[id], unique);
2019-09-06 23:43:07 -04:00
}
}
else if (allowCmdLoad(list[id], modName, exempt))
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
loadCmd(loader, cmdId, modName, list[id], unique);
2019-09-06 23:43:07 -04:00
}
}
}
else
{
qDebug() << "CmdExecutor::getCmdNames() err: command object name '" << unique << "' is not valid.";
}
}
}
void CmdExecutor::buildCommands()
{
Major bug fixes and a few minor changes Fixed a bug that caused the host to not respond to any commands when a module is present. CmdExecutor::loadMods() had a malformed 'for' loop that caused the command executor to infinite loop. Also re-formed the way the external command loaders (modules) are constructed. Instead, the modules are also intialized at the same time as the internal command loader. The CmdExecutor::loadModFile() and CmdExecutor::unloadModFile() functions now load and unload modules via the module name instead of the path to the module main file. The documentation for ASYNC_ENABLE_MOD and ASYNC_DISABLE_MOD updated to reflect this change. Also part of the module contruction re-form, it fixed a bug that caused the command loaders to operate in a thread different from the command executor. CmdExecutor::loadMods() was refactored into cmdExecutor::buildCmdLoaders(), added the internal command loader to it and have it so this function gets called only when the command executor successfully moves to it's own thread. Fixed a bug that caused the session to crash only when a module is present. It turns out the qobject_cast() function used in CmdExecutor::loadModFile() only pulled the surface class ModCommandLoader so any attempt to access the base class functions from CommandLoader caused the session to crash so to fix this, ModCommandLoader was merged into CommandLoader. The Command loader documentation and test module were updated to reflect this change. Fixed a bug that auto removed any modules that were installed when the host is restarted. The issue was in TCPServer::syncModPath() that used the modules's main file without a suffix when the file did indeed have a suffix, causing the function to think the module no longer exists and remove all of it's files from the host file system. The original intent for this function was to keep the module database table in sync with what was actually in the host file system. It was decided that this is not necessary so the function was removed altogether. Also changed up the way the module files are deleted. The TCPSever class will now use the module name instead of the path to the module's main library file and will use a single use ModDeleteTimer for each module that was requested to be deleted instead of a single continuous timer constantly checking for modules to delete. The ls_cmds command will now list the commands in alphabetical order of the command names and fixed a bug with the -find option that would display nothing if an exact match to any text in the table could not be found. Fixed a bug found in the deployed application that would segmentation fault on startup due to certain library files being included in the deployed installation directory. Updated the linux_build.sh script to include only libsqlite, libQT5, libssl and libicu files. It will no longer blindly copy all library files from the dev machine and then remove files the setup script thinks the application might or might not need. Any module built on import rev1 will simply not work and no attempt will be made to make it work so the host new minimum import rev is now rev2. this update is mustly just bug fixes; nothing that would require any client behaviour changes so just the host patch is bumped. 1.0.0 --> 1.0.1
2019-09-12 01:29:46 -04:00
for (auto&& loaderName : cmdLoaders.keys())
{
loadCmds(cmdLoaders[loaderName], getModIdOffs(loaderName), loaderName);
}
2019-09-06 23:43:07 -04:00
}
bool CmdExecutor::externBlockedTypeId(uchar typeId)
{
// the internal host objects will handle sending the following TypeIDs to the clients
// and peers. any attempt to do so via ExternCommand object will be blocked since these
// data types can cause some behaviour issues if sent at an unexpected time.
return (typeId == PRIV_IPC) || (typeId == PUB_IPC) || (typeId == PING_PEERS) ||
(typeId == PEER_STAT) || (typeId == MY_INFO) || (typeId == PEER_INFO) ||
(typeId == HOST_CERT) || (typeId == IDLE) || (typeId == NEW_CMD);
}
bool CmdExecutor::p2pBlockedTypeId(uchar typeId)
{
// this is used to block P2P specific typeIDs. only toPeers() or internDataToIPC()
// should be allowed to send these frame types, all others use this to block them.
return (typeId == P2P_REQUEST) || (typeId == P2P_OPEN) || (typeId == P2P_CLOSE);
}
void CmdExecutor::externDataToIPC(quint16 cmdId, const QByteArray &data, uchar typeId)
{
if (!externBlockedTypeId(typeId) && !p2pBlockedTypeId(typeId))
{
emit dataToSession(cmdId, data, typeId);
}
}
void CmdExecutor::internDataToIPC(quint16 cmdId, const QByteArray &data, uchar typeId)
{
if ((typeId != IDLE) && (typeId != NEW_CMD))
{
emit dataToSession(cmdId, data, typeId);
}
}
void CmdExecutor::castToPeers(const QByteArray &data, uchar typeId)
{
if (!externBlockedTypeId(typeId) && !p2pBlockedTypeId(typeId))
{
QByteArray castHeader = *rdSharedObjs->wrAbleChIds + wrInt(typeId, 8);
emit dataToSession(ASYNC_CAST, castHeader + data, PUB_IPC);
}
}
void CmdExecutor::toPeer(const QByteArray &dst, const QByteArray &data, uchar typeId)
{
if (!externBlockedTypeId(typeId) && (dst.size() == 28))
{
QByteArray p2pHeader = dst + *rdSharedObjs->sessionId + wrInt(typeId, 8);
if (typeId == P2P_REQUEST)
{
if (!rdSharedObjs->p2pPending->contains(dst))
{
rwSharedObjs->p2pPending->append(dst);
emit dataToSession(ASYNC_P2P, p2pHeader + toPEER_INFO(rdSharedObjs), PUB_IPC);
}
}
else
{
if ((typeId == P2P_CLOSE) || (typeId == P2P_OPEN))
{
if (rdSharedObjs->p2pPending->contains(dst) || rdSharedObjs->p2pAccepted->contains(dst))
{
if (typeId == P2P_CLOSE)
{
rwSharedObjs->p2pPending->removeAll(dst);
rwSharedObjs->p2pAccepted->removeAll(dst);
emit dataToSession(ASYNC_P2P, p2pHeader + dst, PUB_IPC);
}
else if (!rdSharedObjs->p2pAccepted->contains(dst))
{
rwSharedObjs->p2pPending->removeAll(dst);
rwSharedObjs->p2pAccepted->append(dst);
emit dataToSession(ASYNC_P2P, p2pHeader + dst, PUB_IPC);
}
}
}
else
{
emit dataToSession(ASYNC_P2P, p2pHeader + data, PUB_IPC);
}
}
}
}
void CmdExecutor::backendFromCmd(quint16 cmdId, const QByteArray &data, uchar typeId)
{
if ((typeId == PUB_IPC) || (typeId == PRIV_IPC) || (typeId == PUB_IPC_WITH_FEEDBACK))
{
emit dataToSession(cmdId, data, typeId);
}
}
void CmdExecutor::openOrCloseChByName(quint16 cmdId, const QString &ch, const QString &sub, bool open)
{
if (!validChName(ch))
{
emit dataToSession(cmdId, toTEXT("err: '" + ch + "' is not a valid channel name.\n"), ERR);
}
else if (!validChName(sub))
{
emit dataToSession(cmdId, toTEXT("err: '" + sub + "' is not a valid sub channel name.\n"), ERR);
}
else if (!channelSubExists(ch, sub))
{
emit dataToSession(cmdId, toTEXT("err: Sub-channel: '" + sub + "' does not exists.\n"), ERR);
}
else
{
Query db(this);
db.setType(Query::PULL, TABLE_SUB_CHANNELS);
db.addColumn(COLUMN_CHANNEL_ID);
db.addColumn(COLUMN_SUB_CH_ID);
db.addCondition(COLUMN_SUB_CH_NAME, sub);
db.addCondition(COLUMN_CHANNEL_NAME, ch);
db.exec();
if (open)
{
openChById(cmdId, db.getData(COLUMN_CHANNEL_ID).toULongLong(), static_cast<uchar>(db.getData(COLUMN_SUB_CH_ID).toUInt()));
}
else
{
closeChById(cmdId, db.getData(COLUMN_CHANNEL_ID).toULongLong(), static_cast<uchar>(db.getData(COLUMN_SUB_CH_ID).toUInt()));
}
}
}
void CmdExecutor::openChByName(quint16 cmdId, const QString &ch, const QString &sub)
{
openOrCloseChByName(cmdId, ch, sub, true);
}
void CmdExecutor::closeChByName(quint16 cmdId, const QString &ch, const QString &sub)
{
openOrCloseChByName(cmdId, ch, sub, false);
}
void CmdExecutor::openChById(quint16 cmdId, quint64 chId, uchar subId)
{
QByteArray id = wrInt(chId, 64) + wrInt(subId, 8);
if (chId == 0)
{
emit dataToSession(cmdId, toTEXT("err: '0' is not a valid channel id. it must an unsigned integer between 1-18446744073709551615.\n"), ERR);
}
else if (countChs(*rdSharedObjs->chIds) == 6)
{
emit dataToSession(cmdId, toTEXT("err: The maximum amount of open sub-channels reached (6).\n"), ERR);
}
else if (containsChId(id, *rdSharedObjs->chIds))
{
emit dataToSession(cmdId, toTEXT("err: The requested sub-channel is already open.\n"), ERR);
}
else if (!channelSubExists(chId, subId))
{
emit dataToSession(cmdId, toTEXT("err: The requested sub-channel does not exists.\n"), ERR);
}
else if (channelAccessLevel(rdSharedObjs, chId) > lowestAcessLevel(chId, subId))
{
emit dataToSession(cmdId, toTEXT("err: Access denied.\n"), ERR);
}
else
{
wrOpenCh(rwSharedObjs, id);
}
}
void CmdExecutor::closeChById(quint16 cmdId, quint64 chId, uchar subId)
{
QByteArray id = wrInt(chId, 64) + wrInt(subId, 8);
if (chId == 0)
{
emit dataToSession(cmdId, toTEXT("err: '0' is not a valid channel id. it must an integer between 1-18446744073709551615.\n"), ERR);
}
else if (!containsChId(id, *rdSharedObjs->chIds))
{
emit dataToSession(cmdId, toTEXT("err: The requested sub-channel is not open.\n"), ERR);
}
else
{
QByteArray peerStat;
wrCloseCh(rwSharedObjs, id, peerStat);
if (!peerStat.isEmpty())
{
emit dataToSession(ASYNC_LIMITED_CAST, peerStat, PUB_IPC);
}
}
}