With the removal of `using namespace std;` we inadvertently started
using some of the global math.h versions of these functions, which take
different types and resulting in computation errors.
The fix is to prefix all these functions with std:: so that we go back
to using the std:: versions.
If a timer expires while stopped, it should trigger when TIMER ON is
run. Instead, on QB64 it triggers randomly after the TIMER ON happens.
The basic issue is that `qbevent` needs to be set to trigger the timer,
but TIMER ON doesn't do that. The regular timer logic that does that
already set it when the timer expired while sleeping, so it won't set it
again. The simplest solution is to just alway set qbevent = 1 when TIMER
ON is done. It's slightly less efficent but doesn't hurt to set it even
when there are no timers that expired.
Fixes: #293
Timer's were not firing at the right time if they were started shortly
after the program started, instead they would fire at twice the interval
time (and then work correctly after that).
The issue was a mistaken assumption about `time_now`, with the idea that
if `last_time == 0` then `time_now` will be large enough such that the
interval check will pass. This is wrong because in most cases `time_now`
starts at zero at program start, so when `last_time == 0` it will take
one full interval of the timer before `time_now` is large enough for the
interval check to pass (at which point the timer is initialized and runs
normally).
This simply refactors the timer logic so that `last_time == 0` is
checked first, rather than if the interval has expired. This doesn't
change how the normal logic works, but ensures that the value of
`time_now` does not matter for initializing a timer.
Fixes: #273
Currently main() includes logic that is intended to sync time() with
GetTicks() for the purpose of using GetTicks() to get millisecond
accuracy with time(), which only has second accuracy. Unfortunately, the
'syncing' up of these time sources results in an average of a half
second delay in starting a QB64-PE program.
This logic is easly replaced with std::chrono, which provides a real
time clock which is also millisecond accurate. That removes the need to
use time() and GetTicks() together to get millisecond accuracy, and
means the delay syncing them is no longer necessary.
I also separated most of the "delay" and "time" related functions into
datetime.cpp, and included the new std::chrono code into that file.
Since I needed to call some of the rounding functions in datetime.cpp I
also moved that stuff out into its own .cpp and header files to clean
things up a bit.
Fixes: #282
The current GLUT initialization logic is flawed because it allows the
QB64 code portion of the program to start on a separate thread at the
same time that the GLUT code is starting. This results in a race where
some commands won't work for a brief period at the beginning of the
program (with "won't work" being very inconsistent, some return invalid
values, some have a chance at seg faulting).
The same issue also leads to us adding many `while (!window_exists)`
checks in an attempt to solve this race for some of the commands.
Unfortunately this solution is very inconsistently applied leading to
some deadlock situations, and really it's a silly solution when this
race is entirely our creation anyway.
To fix this, the logic was changed such that we perform all of the GLUT
initialization besides calling `glutMainLoop()` before we ever start the
thread that runs the actual QB64 code. By doing it this way we ensure
that the GLUT initialization has already taken place before the code
runs and thus the race is gone.
Things get a bit more interesting with $SCREENHIDE, because that simply
delays the execution of the GLUT initialization indefinitely until
_ScreenShow is done. This was previously very buggy since some commands
rely on FreeGLUT being up and will simply hang the entire program if
run. Other commands have logic to catch this and simply return zero.
The above issue is solved with the `NEEDS_GLUT()` and `OPTIONAL_GLUT()`
macros. Both of them simply check if the GLUT initialization has taken
place and exit the current function if it has not. The difference
between the two is that `NEEDS_GLUT()` throws an 'illegal function call'
error while `OPTIONAL_GLUT()` simply exits with no error. The choice of
behavior of each function was based upon its previous behavior - if it
checked `screen_hide` and exited with no error previously, then
`OPTIONAL_GLUT()` was used. If it deadlocked or similar then
`NEEDS_GLUT()` was used (so instead of deadlocking, it now produces an
error). In this way, programs can now never get stuck due to the use of
`$SCREENHIDE` and all the commands have consistent behavior that can be
relied upon.
Fixes: #234
In order to allow $DEBUG to work with programs that call CLEAR, the connection handle used to connect to the IDE is locked by default and cannot be CLOSEd. With this change, the debuggee itself can now unlock the handle and close the link.
This reverts commit a94b738407.
It turns out that change stops the SHELL function returning the
child exit code, which is worse than having zombie processes.