Several JSON message builders conditionally omit fields when their
values are empty or null:
- DMRSlot, P25Control, NXDNControl: source_info omitted when the
callsign/ID lookup returns an empty string
- DStarControl: reflector omitted in RF start messages and when
the reflector pointer is null in network start messages
- YSFControl: reflector omitted in RF start messages
Downstream consumers (such as Display-Driver) that access these
fields via nlohmann::json const operator[] crash with an assertion
failure when the key is absent.
Always emit these fields with an empty string default so that the
JSON schema is consistent regardless of lookup results or message
source.
CDStarControl::clock(),
CM17Control::clock(),
CNXDNControl::clock(),
CP25Control::clock(),
CYSFControl::clock():
- Leave ASAP when it's disabled.
CDMRSlot::enable(),
CDStarControl::enable(),
CM17Control::enable(),
CNXDNControl::enable(),
CP25Control::enable(),
CYSFControl::enable():
- Log a message when Controller get disabled while running, "<MODE>, RF user has timed out" or "<MODE>, network user has timed out", depending of the RF/Net state.
CMMDVMHost::run():
- After getting the modem data, don't handle it (read/write<Controler>, etc) if the given mode is not enabled.
MMDVMHost:
- Rename CMMDVMHost::processEnableCommand() to CMMDVMHost::enableModemMode().
- Add CMMDVMHost::enableMode() and CMMDVMHost::disableMode(), called now from CMMDVMHost::remoteControl(), which clarify the code.
- CMMDVMHost::remoteControl():
* Fix indentation,
* Simplify ENABLE_x/DISABLE_x cases code.
- CMMDVMHost::setMode():
* Enable pocsag network only if m_pocsagEnabled is true,
* Enable ax25 network only if m_ax25Enabled is true (not m_fmEnabled).
Adding 2 byte offset to the data pointer passed to CDMRTA::add() in the "RF" section of CDMRSlot.cpp::writeModem(). Fixes the bug that incorrectly decodes talker alias coming in over RF.
There is two cases that shows "received network late entry"
1. dataType == DT_VOICE_PI_HEADER && m_netState != RS_NET_AUDIO
2. dataType == DT_VOICE_SYNC && m_netState == RS_NET_IDLE
case 1 is already treated. this fix is for case 2.
