F/A-18E and F/A-18F are added to their factions, Growler is currently in
its respective factions but was released in 2009, this will be changed
if needed.
If we don't do this, the uvicorn server may log its shutdown after the
Qt application has closed, and the signal this attempts to emit may not
be valid. Disconnect the log signals when the application exits to
prevent that.
There's actually another solution that I thought would be better, but I
couldn't get it to work:
https://www.pyinstaller.org/en/stable/feature-notes.html#automatic-hiding-and-minimization-of-console-window-under-windows
describes a way to have pyinstaller hide or minimize the console rather
than disabling it entirely. I was never really fond of getting rid of
the console window in the first place, but it did bother some users. If
we could get the hide or minimize option working, that'd probably avoid
bothering users, but also make the logs much easier to find, get us out
of the trouble of maintaining our own log viewer, and fix the problem
mentioned in the comment I add here (the log window only works if
there's only one in memory log handler).
Another option would be ditching our log window and instead just having
that menu item open the log file or directory in whatever program the OS
defaults to (probably notepad). It would still have the quirk of maybe
needing to open more than one location, since logging is use
configurable.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/3278.
There are a few different code paths for deleting packages depending on
the state of the package, and two of them were deleting items from a
list they were iterating over without first making a copy, causing each
iteration of the loop to skip over a flight, making it still used since
the flight was never deleted, but unreachable since the package that
owned it was deleted.
This only happened when the package was canceled in its draft state
(the user clicked the X without ever saving the package), or if the user
canceled a package mid fast forward (the controls for which aren't even
visible to users) while only a portion of the package was active. In
both cases, only even numbered flights were lost.
There's a better fix lurking in here somewhere but the interaction with
the Qt model complicates it. Fortunately that mess would be cleaned up
automatically by moving this dialog into React, which we'll do some day.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/3257.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/3258.
fix a number of regressions in the flight waypoint list by changing the
indexing and finding a work-around to blocking of signals
This PR addresses some (but not all) recently reported issues with the
waypoints screen reported in Issue #3188 . This PR was tested by:
- Changing the waypoint altitude and confirming it shows up correctly
when reloading the waypoint list window and on the map
- Adding a waypoint and confirming that it shows up immediately and
persists on reload
- Deleting a waypoint (except the first waypoint) and confirming that it
is removed immediately and persists on reload,
Known issues: first waypoint (typically hold) cannot be deleted -- still
looking into this one.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/3188.
Moves the stretch to the bottom of the page to avoid awkward whitespace
in the middle. Presumably the totals used to be at the bottom (since
that's a normal place for a total), but it was moved to the top,
probably since that was the most interesting data and we didn't want to
scroll though all the details to find that one point.
This also removes the unused code path where the total would be shown at
the bottom.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/1288.
This property affects safe compat because the ID is what gets preserved
in the save, but it's unfortunately also used as the display name, which
means changing the display name breaks save compat. It also prevents us
from changing display names without breaking faction definitions.
This is the first step in fixing that. The next is adding a separate
display_name property that can be updated without breaking either of
those.
The weapon laser codes can be set more easily from the weapon laser code
combo box. Setting the properties explicitly here will just cause
conflicts and annoying UI bugs. Hide those properties from the UI.
This makes it possible to have the right laser code set for hot start
aircraft that (typically) do not allow changing laser codes when the
engine is on.
This PR addresses #3066 by restricting the list of control points in the
new unit transfer dialog to control points reachable from the origin.
This change centralizes the logic for reachable nodes to the
TransitNetworkBuilder class.
This PR was tested by:
1. Loading save from #3066
2. Using cheat menu to destroy runway at Wadi al Jandali
3. Purchasing units at any of the other control points
4. Pass the turn to allow the purchase to complete
5. Initiating a unit transfer from the other control point and
confirming that Wadi al Jandali does not show up in the list
Steps 2-4 are needed as no ground units show up at Melez when loading
the save directly from the latest dev build.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/3066.
Use the new data from pydcs to improve the properties UI:
* Use human readable names
* Use appropriate control types
* Limit min and max values as appropriate for each property
* Show labels
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/3090.
The flight plan used to not include a waypoint for departure, so a few
places would create one for the sake of the UI, or were built to assume
there was a missing waypoint that was okay to ignore. At some point we
added them to the flight plan, but never updated the UI, so the waypoint
list in the flight dialog started counting from 1 instead of 0, and the
openapi endpoint wrongly reported two departure waypoints to the front-
end.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/3037.
If we're going to show a game over dialog, we need to do so before
moving to the next turn, but we will still want to show the debriefing
window. Move those steps to happen after the debrief window.
https://github.com/dcs-liberation/dcs_liberation/issues/978
Any real end game dialog needs a "new game" button. If only the main
window can usefully call the NGW we'd have to plumb that object through
and call into it from that dialog, which is gross. Just make it easier
to call the wizard.
https://github.com/dcs-liberation/dcs_liberation/issues/978
The aircraft banners are only used for the unit info window, and that's
not a normal part of gameplay. We spend a bit over 1.5 seconds
preloading this data at startup (about 25% of the non-game load startup
time). This data is only expensive to load in aggregate, and we never
need all of it. The unit info window is not noticeably slowed by this,
but startup is noticeably faster without it.
It's not feasible to actually check the parking limits because we can't
identify parking limits for carriers until the theater is populated.
Doing so is expensive (and depends on other NGW inputs). Instead,
compare against the version of the campaign and guess.
A new (minor) campaign version has been introduced which makes this
required to improve the UI hint. Campaigns that are compatible with the
new rules should update their version to advertise support.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/2909.
This also changes the window close button of the air wing configuration
dialog to cancel rather than revert and continue, because otherwise
there's no way for the user to back out of the dialog without fixing all
the overfull bases first.
Fixes https://github.com/dcs-liberation/dcs_liberation/issues/2910.
This does show the theoretical parking use of full squadrons even when
the new rules are not enabled. Since limits can be enabled manually
later in the game, it's still useful information, even if it's a bit
misleading.
https://github.com/dcs-liberation/dcs_liberation/issues/2910
