| Age | Commit message (Collapse) | Author |
|---|
|
|
|
Closes #534.
See: https://source.winehq.org/git/wine.git/blob/refs/heads/stable:/dlls/kernelbase/console.c#l520
|
|
This should fix the badly-rendered progress message when run in
Terminal.app.
|
|
Sometimes the viewport srWindow may report an invalid rectangle where
the top row is below the bottom one.
|
|
Somehow I forgot to save after copy-pasting some code and changing it.
|
|
The cursor must be restored after the line is printed, not before.
Take into account the visible viewport to correctly compute the terminal
size.
|
|
In order to update the printed progress string the code tried to move
the cursor N cells to the left, where N is the number of written bytes,
and then clear the remaining part of the line.
This strategy has two main issues:
- Is only valid if the number of characters is equal to the number of
written bytes,
- Is only valid if the line doesn't get too long.
The second point is the main motivation for this change, when the line
becomes too long the terminal wraps it to a new physical line. This
means that moving the cursor to the left won't be enough anymore as once
the left border is reached it cannot move anymore.
The wrapped line is still stored by the terminal as a single line,
despite now taking more than a single one when displayed. If you try to
resize the terminal you'll notice how the contents are reflowed and are
essentially illegible.
Querying the cursor position on non-Windows systems (plot twist,
Microsoft suggests using VT escape sequences on newer systems) is
extremely cumbersome so let's do something different.
Before printing anything let's save the cursor position and clear the
screen below the cursor, this way we ensure there's absolutely no trace
of stale data on screen, and after the message is printed we simply
restore it.
|
|
|
|
|
|
This bug caused the compiler to deadlock when multiple c objects
were build in parallel.
Thanks @kprotty for finding this bug!
|
|
* move concurrency primitives that always operate on kernel threads to
the std.Thread namespace
* remove std.SpinLock. Nobody should use this in a non-freestanding
environment; the other primitives are always preferable. In
freestanding, it will be necessary to put custom spin logic in there,
so there are no use cases for a std lib version.
* move some std lib files to the top level fields convention
* add std.Thread.spinLoopHint
* add std.Thread.Condition
* add std.Thread.Semaphore
* new implementation of std.Thread.Mutex for Windows and non-pthreads Linux
* add std.Thread.RwLock
Implementations provided by @kprotty
|
|
Related: #4917
|
|
|
|
|
|
thx king protty
|
|
We generally get away with atomic primitives, however a lock is required
around the refresh function since it traverses the Node graph, and we
need to be sure no references to Nodes remain after end() is called.
|
|
If this errdefer did get run it would constitute a race condition. So I
deleted the dead code for clarity.
|
