mirror of
https://github.com/saitohirga/WSJT-X.git
synced 2024-08-01 12:22:30 -04:00
2cfbb15b4f
The concept of a nominal receive and transmit frequency has been
introduced. This is used as a base frequency for Doppler correction,
frequency setting and reporting. The start up frequency is now zero
which is updated by the first rig control status report. This needs
more work to accommodate calling frequency plus working frequency
operation as is used for random MS operation etc..
The main window frequency display now shows the transmit dial
frequency while transmitting.
The mode changing logic sequence has been changed such that the rig is
correctly put into and taken out of split mode as required by the
target mode. This also avoids the "other" VFO having its frequency
changed when entering a mode that does not use split operating like
WSPR.
The main window band combo box edit may now be used to input an kHz
offset from the current MHz dial frequency. This is intended for
setting a sked or working frequency on the VHF and up bands. For
example the working frequency for 23cms might be set to 1296MHz and a
working frequency of 1296.3MHz would be selected by selecting the
23cms band with the combo box drop down list and then entering 300k
into the band combo box edit widget.
When using JT4 modes a CTRL+Click on the waterfall adjusts the nominal
frequency such that the frequency clicked on becomes the Tx and Rx
frequency using the fixed 1000Hz DF that JT4 modes use. This will
probably be extended to all QSO modes when used in VHF & up mode. This
assumes that 1000Hz is an optimal DF for both Tx and Rx and therefore
one can "net" to an off frequency, but visible on the waterfall,
caller with one click.
Improvements to OmniRig rig control including use of the serial port
control lines RTS or DTR, on the CAT serial port used by OmniRig, for
PTT control.
Incrementing transaction sequence numbers added to messages to and from
the rig control thread. This enables round trip status to be tracked
and associated with a request. For example a command that might cause
several asynchronous status updates can now be tracked in the
originating thread such that it is clear which updates are caused by
executing the request. This in turn allows updates to be held until
the request is complete i.e. the state is consistent with the results
of the request.
Messages to the rig control thread are now posted as a new state
(Transceiver::TransceiverState) object. The rig control thread tracks
requests and actions any differences between the prior requests and
the new state.
The rig control thread is now stored on the heap so that it can be
closed down and released as needed. Along with this the rig control
close down semantics are better defined avoiding some potential
deadlock situations.
If the rig is placed into split mode it will be reverted to simplex
mode when the rig connection is closed.
When using direct rig control via Hamlib, rigs that have A/B VFO
arrangements and no method to query the current VFO like many Icoms
and the Yaesu FT-817/857/897(D) series now have smarted frequency
updating requiring no VFO changes when changing the frequency. This
is particularly important when doing Tx Doppler correction to avoid
glitches.
The implementation of emulated split operating mode ("Fake It") is
simplified and improved.
A dummy Hamlib transceiver for PTT control on a separate port is no
long instantiated if CAT or VOX PTT control is selected.
The resolution and any rounding of the rig CAT frequency set and get
commands is determined automatically upon opening the rig connection.
This is needed to determine the rate of frequency updates for Doppler
tracking. It also allows the rig to be more accurately controlled.
Frequency calibration is calculated separately for the receive and
transmit frequencies.
Whether the rig modulation mode should be controlled is now a
constructor argument rather than being passed with individual rig
control requests.
Doppler shift correction is considerably enhanced with simpler
controls and much better rig control. A new mode of tracking called
"receive only" is introduced for those with rigs that cannot be QSY:ed
via CAT when transmitting. Such rigs have a Doppler correction
calculated for the middle of the next transmit period just before
transmission starts. While using Doppler tracking it is now possible
to adjust the sked frequency either using the new kHz offset feature
of the main window band combo box or by directly tuning the rig VFO
knob while holding down the CTRL key.
The astronomical data window that includes Doppler tracking control is
now opened and closed using a checkable menu item to avoid it being
accidentally closed.
Debug configuration rig control diagnostic messages now have a
facility argument for clearer and more standardized trace messages.
git-svn-id: svn+ssh://svn.code.sf.net/p/wsjt/wsjt/branches/wsjtx@6590 ab8295b8-cf94-4d9e-aec4-7959e3be5d79
66 lines
2.2 KiB
C++
66 lines
2.2 KiB
C++
#include "EmulateSplitTransceiver.hpp"
|
|
|
|
EmulateSplitTransceiver::EmulateSplitTransceiver (std::unique_ptr<Transceiver> wrapped, QObject * parent)
|
|
: Transceiver {parent}
|
|
, wrapped_ {std::move (wrapped)}
|
|
, rx_frequency_ {0}
|
|
, tx_frequency_ {0}
|
|
, split_ {false}
|
|
{
|
|
// Connect update signal of wrapped Transceiver object instance to ours.
|
|
connect (wrapped_.get (), &Transceiver::update, this, &EmulateSplitTransceiver::handle_update);
|
|
|
|
// Connect other signals of wrapped Transceiver object to our
|
|
// parent matching signals.
|
|
connect (wrapped_.get (), &Transceiver::resolution, this, &Transceiver::resolution);
|
|
connect (wrapped_.get (), &Transceiver::finished, this, &Transceiver::finished);
|
|
connect (wrapped_.get (), &Transceiver::failure, this, &Transceiver::failure);
|
|
}
|
|
|
|
void EmulateSplitTransceiver::set (TransceiverState const& s, unsigned sequence_number) noexcept
|
|
{
|
|
#if WSJT_TRACE_CAT
|
|
qDebug () << "EmulateSplitTransceiver::set: state:" << s << "#:" << sequence_number;
|
|
#endif
|
|
// save for use in updates
|
|
rx_frequency_ = s.frequency ();
|
|
tx_frequency_ = s.tx_frequency ();
|
|
split_ = s.split ();
|
|
|
|
TransceiverState emulated_state {s};
|
|
if (s.ptt () && split_) emulated_state.frequency (s.tx_frequency ());
|
|
emulated_state.split (false);
|
|
emulated_state.tx_frequency (0);
|
|
wrapped_->set (emulated_state, sequence_number);
|
|
}
|
|
|
|
void EmulateSplitTransceiver::handle_update (TransceiverState const& state,
|
|
unsigned sequence_number)
|
|
{
|
|
#if WSJT_TRACE_CAT
|
|
qDebug () << "EmulateSplitTransceiver::handle_update: from wrapped:" << state;
|
|
#endif
|
|
|
|
if (state.split ())
|
|
{
|
|
Q_EMIT failure (tr ("Emulated split mode requires rig to be in simplex mode"));
|
|
}
|
|
else
|
|
{
|
|
TransceiverState new_state {state};
|
|
// Follow the rig if in RX mode.
|
|
if (state.ptt ()) new_state.frequency (rx_frequency_);
|
|
|
|
// These are always what was requested in prior set state operation
|
|
new_state.tx_frequency (tx_frequency_);
|
|
new_state.split (split_);
|
|
|
|
#if WSJT_TRACE_CAT
|
|
qDebug () << "EmulateSplitTransceiver::handle_update: signalling:" << state;
|
|
#endif
|
|
|
|
// signal emulated state
|
|
Q_EMIT update (new_state, sequence_number);
|
|
}
|
|
}
|