For example, if you have a config that defines OpenGFX as baseset
but for some reason you have no basesets anymore. In that case
bootstrap downloads OpenGFX for you, but it will still show the
error that "OpenGFX was not found" after the bootstrap. This was
an error generated before the bootstrapped kicked in.
Simply muting all errors during bootstrap solves this; as we cannot
show them anyway, this is fine. Any errors that remain after
bootstrap will be generated again anyway.
There are various of ways bootstrap can fail:
- Failing network connection
- Incomplete download
- No write permissions
- Disk full
- (others I forgot)
They all result in a screen with no windows. To ensure we at least
always show something when anything bad happens, if the bootstrap
is not successful, show a screen what the next step for the human
should be.
english (us): 7 changes by 2TallTyler
estonian: 17 changes by siimsoni
hungarian: 100 changes by pnpBrumi
ukrainian: 8 changes by StepanIvasyn
dutch: 24 changes by Afoklala
spanish: 338 changes by MontyMontana
french: 29 changes by MalaGaM
portuguese (brazilian): 1 change by Greavez
This means if you execute a script from a script from a script, ..
for more than 10 times, it bails out now. This should be sufficient
for even the most complex scripts.
MaskWireBits always returns its input unchanged if the input
has only 0 or 1 track bits set.
Having only 0 or 1 track bits sets (i.e. non junction tiles)
is by far the most common case.
Examining the state of neighbouring tiles and the subsequent
masking logic is relatively expensive and can be omitted in this case.
It didn't sit well to me, how I wrote the commit initially. First
casting a variable into another, only to write it back into the
originally feels wrong.
This flow makes a bit more sense to me.
Otherwise that might cause calls to the video-driver, which are
already shut down by now. This causes, depending on the video-driver
crashes or weird effects.
Basically, modal windows had their own thread-locking for what
drawing was possible. This is a bit nonsense now we have a
game-thread. And it makes much more sense to do things like
NewGRFScan and GenerateWorld in the game-thread, and not in a
thread next to the game-thread.
This commit changes that: it removes the threads for NewGRFScan
and GenerateWorld, and just runs the code in the game-thread.
On regular intervals it allows the draw-thread to do a tick,
which gives a much smoother look and feel.
It does slow down NewGRFScan and GenerateWorld ever so slightly
as it spends more time on drawing. But the slowdown is not
measureable on my machines (with 700+ NewGRFs / 4kx4k map and
a Debug build).
Running without a game-thread means NewGRFScan and GenerateWorld
are now blocking.
gui_zoom was never clamp'd between zoom_min/zoom_max.
zoom_min controls how zoomed-in we load sprites. For a value of 1,
no quad-sizes sprites are loaded. If gui_zoom would be 0, meaning
it wants quad-sized sprites to display, it was printing random
stuff to the screen, which could or could not result in crashes.
Otherwise both the draw-thread and game-thread can do it both
at the same time, which gives rather unwanted side-effects.
Calling it from the draw-thread alone is sufficient, as we just
want to create some unpredictable randomness for the player. The
draw-thread is a lot more active (normally) than the game-thread,
so it is the best place of the two to do this.
Additionally, InteractiveRandom() mostly has to do with visuals
that are client-side-only, so more related to drawing than to
game.
v->tile for aircrafts is always zero when in the air. Only when
it starts its landing (or take-off) patterns it becomes a sane
value.
So instead, base the news on the last x/y coordinates of the plane.
english (us): 18 changes by 2TallTyler
korean: 17 changes by telk5093
german: 13 changes by danidoedel, 4 changes by Wuzzy2
finnish: 17 changes by hpiirai
catalan: 17 changes by J0anJosep
lithuanian: 33 changes by devbotas
spanish: 17 changes by MontyMontana
portuguese (brazilian): 20 changes by Greavez
polish: 9 changes by yazalo
This because video-drivers might need to make changes to their
context, which for most video-drivers has to be done in the same
thread as the window was created; main thread in our case.
This allows drawing to happen while the GameLoop is doing an
iteration too.
Sadly, not much drawing currently can be done while the GameLoop
is running, as for example PollEvent() or UpdateWindows() can
influence the game-state. As such, they first need to acquire a
lock on the game-state before they can be called.
Currently, the main advantage is the time spend in Paint(), which
for non-OpenGL drivers can be a few milliseconds. For OpenGL this
is more like 0.05 milliseconds; in these instances this change
doesn't add any benefits for now.
This is an alternative to the former "draw-thread", which moved
the drawing in a thread for some OSes. It has similar performance
gain as this does, although this implementation allows for more
finer control over what suffers when the GameLoop takes too
long: drawing or the next GameLoop. For now they both suffer
equally.
Drawing in a thread is a bit odd, and often leads to surprising
issues. For example, OpenGL would only allow it if you move the
full context to the thread. Which is not always easily done on
all OSes.
In general, the advise is to handle system events and drawing
from the main thread, and do everything else in other threads.
So, let's be more like other games.
Additionally, putting the drawing routine in a thread was only
done for a few targets.
Upcoming commit will move the GameLoop in a thread, which will
work for all targets.
The video drivers using the OpenGL backend are currently our only
accelerated drivers. The options defaults to off for macOS builds and
to on everywhere else.
Co-authored-by: Michael Lutz <michi@icosahedron.de>
chinese (traditional): 5 changes by benny30111
estonian: 1 change by siimsoni
italian: 1 change by AlphaJack
ukrainian: 4 changes by StepanIvasyn
tamil: 37 changes by Aswn
portuguese (brazilian): 19 changes by Greavez
Reworked how the screenshot command works while keeping it backwards
compatible. It can now more freely understand arguments, and has
the ability to make SC_DEFAULTZOOM screenshots.
In other words, it should only (!) return true if A comes for B.
This promise was broken for the situation where two values are
identical. It would return true in these cases too. This is of
course not possible: if two values are identical, neither come
before the other. As such, the sorter was not imposing strict
weak ordering relations.
libstdc++ handled this scenario just fine, but libc++ crashes
badly on this, as it allowed comparing of [begin, end] instead
of [begin, end).
libc++ considered this not a bug (and by specs, they are correct;
just this way of crashing is of course a bit harsh):
https://bugs.llvm.org/show_bug.cgi?id=47903
english (us): 4 changes by 2TallTyler
italian: 4 changes by troccoli
serbian: 251 changes by nkrs
german: 6 changes by ebla71, 2 changes by Wuzzy2
romanian: 3 changes by ALEX11BR
russian: 11 changes by Ln-Wolf
ukrainian: 2 changes by StepanIvasyn
lithuanian: 15 changes by devbotas
spanish: 2 changes by perezdidac
The bootstrap has the _switch_mode to SM_MENU, and never leaves
this mode. Neither is it considered a modal window (while in some
sense it really is). So .. we need to add another "draw anyway"
exception, to make sure bootstrap is being drawn.
When you are downloading a map, all the commands are queued up
for you. Clients joining/leaving is done by the network protocol,
and as such are processed immediately. This means that by the
time you are processing the commands, a client that triggered
it, might already have left.
So, all commands that do something with ClientID, shouldn't
error on an invalid ClientID when DC_EXEC is set, but
gracefully handle the command anyway, to make sure the
game-state is kept in sync with all the clients that did
execute the DoCommand while the now-gone client was still
there.
Additionally, in the small chance a client disconnects between
the server validating a DoCommand and the command being
executed, also just process the command as if the client was
still there. Otherwise, lag or latency can cause clients that
did not receive the disconnect yet to desync.
Strictly seen, there are "N" people -waiting- in front of you
in the queue, but it is nicer to show "N + 1" for the person that
is currently downloading the map. Avoids it showing:
"0 clients in front of you". That just feels a bit off.
swedish: 60 changes by kustridaren
norwegian (bokmal): 12 changes by buzzCraft
czech: 82 changes by PatrikSamuelTauchim, 1 change by tomas-vl
italian: 86 changes by AlphaJack, 9 changes by federico1564S
german: 16 changes by ebla71
romanian: 10 changes by ALEX11BR
ukrainian: 3 changes by StepanIvasyn
spanish: 1 change by MontyMontana
SendError() notifies all clients of the disconnect. This calls
CloseConnection() at the end, which also notified the clients
of the disconnect. Really no need to do it twice.
The status NETWORK_RECV_STATUS_SERVER_ERROR is only set by
SendError(), so in case that is the status, don't let
ClientConnection() send another notification.
CheckCompanyHasMoney() was also executed when not using DC_EXEC,
resulting in an error about shortage of money instead of the
estimation.
This mostly is a problem for AI players, as they will have no
way to know how much it would have cost.
Basically, follow_track.hpp contains a fix for half-tiles, but
this wasn't duplicated for when trying to find a depot and in
a few other places. This makes sure all places act the same.
Patric Stout
translators
Tyler Trahan
Michael Lutz
Matt Kimber
Didac Perez Parera
stormcone
frosch
Jonathan G Rennison
sean
PeterN
dP
Eddi-z
Limyx826
Charles Pigott
Owen Rudge