Fairly simple, MacOS High Sierra's libcurl version is too old and not
have `CURLINFO_CONTENT_LENGTH_DOWNLOAD_T`. This adds a version check to
use the older version of that command which gives back a double instead.
Fixes: #287
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
Fairly simple, the finished entry is free'd and then removed from the
list, but that order results in us accessing the entry's next member
after it has been free'd. Swapping the order of the operations fixes the
issue.
Fixes: #281
The commands _ScreenX and _ScreenY got significantly slower due to the
need to wait for the GLUT thread to wake up and execute the glutGet()
command for them. We've already seen a few programs (including the IDE)
where this behavior completely grinds the program to a halt, so we
definitely can't keep it.
The simple solution here is to not call glutGet() on every _ScreenX/Y
command. Instead every time the idle/timer function runs we get the
current values for the relevant glutGet() variables and store them.
libqb_glut_get() then checks if the value being read is one of the ones
we read in the idle/timer functionand if so just returns the last read
value. By doing it this way the commands no longer has to wait on the
GLUT thread for the result.
Fairly straightfowrad, programs were randomly seg faulting on exit. This
was happening due to GLUT registering a cleanup function via atexit(),
which then gets called when exit() is called.
The issue happens when exit() is called on a thread other than the GLUT
thread, which leads to the exit() call then attempting to cleanup GLUT
while the other thread is still using it, which randomly leads to seg
faults.
Fixing this is slightly annoying. We cannot stop the GLUT thread, as
the basic GLUT API (which is used on Mac OS) simply does not offer a way
to exit the glutMainLoop() call. Thus the simplest solution is to simply
make sure we call exit() on the GLUT thread, which we can fairly easily
due via the message queue.
That being the case, a new libqb_exit() API was added, which simply
regsiters the GLUT exit message and then waits for the program to end.
The atexit() handler then runs on the GLUT thread and everything works
out fine.
In the future we probably should redo the exit logic a bit so that all
the threads are actually stopped/joined to ensure the exit process is
consistent, however this is good enough for now. Also, there's plenty of
error states which call exit() which I did not address.
This fixes all the code so that all the calls to glut functions
happen on the same thread that is running GLUT.
We achieve this by creating a queue of GLUT commands to execute.
Commands can be added to the queue anywhere in the code, and then the
queue is processed on the GLUT thread via it's idle func or timer func.
The command is run and if necessary the result is provided in the
message queue object. Each object contains a completion which can be
waited on to block until the GLUT thread has processed the command.
Fixes: #66
This adds the libqb_http API, which is designed to support HTTP and
HTTPS usage from QB64-PE source.
The design consists of a single thread which services all the HTTP(s)
connections. There are then various libqb_http APIs exposed that allow
interacting with this thread to create a new connection, query
connection status, read data, or close the connection.
Internally the thread makes use of the curl_multi interface to allow a
single thread to service multiple HTTP(s) connections in parallel. This
means you can _OPENCLIENT() multiple HTTP connection in a row and all of
them will be serviced at the same time in whatever order data is
available.
HTTP is optional and selected via a Makefile setting. A stub is swapped
in if HTTP support is not used, which avoids need to add another build
flag to libqb.cpp.
Changing midiaudio.h will make it harder to incorporate new versions
into QB64-PE as they come out. To fix that I have reverted all the
changes to midiaudio.h and moved the few private parts we were using
into a separate 'filepath' API that's part of libqb.
We started defining `_WIN32_WINNT` a little while ago to express that we
require Windows Vista or above for support. This enables us to access
some Windows Vista-only APIs. The fact that `WINVER` also needs to be
defined was missed, and it seems that defining one means the other no
longer gets defined automatically as it did before. Thus we're simplying
now also defining `WINVER` the same as `_WIN32_WINNT`.
This fixes High-DPI awareness and a few other things that were gated
behind WINVER checks.
We shouldn't allow mutex lock/unlock to silently do nothing if NULL is
passed, as that is very likely a bug. Beyond that the Windows version
doesn't do this, so it's inconsistent as well.
This moves a lot of the preprocessor flags for what compiler/platform
we're using into a libqb-common.h header inside the libqb/include
folder. This gets included at the top of every libqb .cpp file, and is
intended to be fairly small, providing only necessary things like
_WIN32_WINNT (which needs to be defined before including <windows.h> or
friends).
The Buffer API implements an append-only buffer, where you can write to
the end or read from the beginning. Data that is read is discarded from
the buffer, and you can query the buffer to get the current amount of
data inside.
Internally the buffer API is implemented as a chain of separate buffers.
This is done to avoid having to resize the existing buffer, which is
expensive. We keep track of where the reading currently is, and discard
the internal buffers after all the data in them is read.
Completions are basically a oneshot flag, which provide a `wait()` call
that blocks until 'finish()' has been called on the completion.
The nice aspect of completions is that because it is a oneshot the order
does not matter - if 'finish()' is called before 'wait()' then 'wait()'
returns immediately. It makes the logic for waiting until a thread is
done finishing up some work easy to implement.
This adds generic APIs to libqb for handling thread's, mutex's, and
condition variables. On Linux and OSX these are implemented via the ones
provided by pthreads. On Windows they're implemented via the ones
provided by the Win32 API.
For compiling, the code itself is not conditional, but the Makefile
includes logic to decide which implementation to pick.
Note that it would have been nice to simply use std::thread and friends
from C++11, however using them on MinGW appears to be a bit messy. Since
using the Windows ones directly isn't that hard this was an easy compromise.
This sets up a few different flags we'll need for the conditional
compiling, and also sets the C++ standard to gnu++11, which effectively
just matches what we were implicitly using before.
Mostly old build scripts and helper files that are now covered by the
Makefile.
A notable deletion is the glew dll and lib files. These are unnecessary
because we compile `glew.c` directly rather than link against the dll or
lib copies we have.
If a user has Homebrew installed, calling g++ won't use Apple's Xcode version of the tool, which is required by QB64. This fix continues on with the patch introduced in 4dbdddc89e.
Extra "osx" references replaced with "macOS", but nothing substantial (almost aesthetic).
Implements changes to allow programs to run on ARMBI-v7 Android devices which also provides a massive performance boost to but runtime and startup time.
Implements changes to allow programs to run on x86-based Android devices.
Fix to ensure variables allocated in the "stack" are aligned on appropriate boundaries which would have otherwise crashed on Android.
Fix to _DEVICES to ensure variables are aligned on appropriate boundaries which would have otherwise crashed on Android (includes some refactoring).
Added permissions to allow reading from/writing to external storage. Use the _DIR$(...) function to help locate appropriate paths.
The scroll wheel now behaves correctly on Linux. Reports suggest that the fix does not have any effect
on OSX though. The licensing documentation is updated to reflect the fact that there are no GPL components
included in a QB64 program. Some redundant code was removed from the Windows setup script. OSX setup and
run scripts were missing the executable bit - hopefully the repository will preserve that change.
Fixes gamepad support dependency / thread of call
Allows libqb.cpp to use Objective C++
Fixes tracking/handling of SHIFT, ALT, CTRL & CAPS LOCK
Fixes serious bug in internal getticks() which caused _LIMIT, _DELAY, SLEEP, TIMER, etc to malfunction and IDE to be unusably slow
