-
Notifications
You must be signed in to change notification settings - Fork 63
problem with the size of the first window with wxwidgets ! (+valgrind / callgrind) #1471
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
I forgot to mention that to use valgrind + callgrind we need to compile : |
I succeed to reproduce on a Debian10 . Obviously some uninitialized size somewhere. Indded, must be solved before 1.0.2 |
This is purely an effect of the gnome window manager. |
But I am using XFCE ?! (xfwm4) Please note that Valgrind give several mentions to non initialized variables when calling Window or Plot |
See #1473 |
created an issue for wxWidgets --- the size is not constant through platforms. |
This is a continuation of Issue #1461 since this issue (no first drawing of plot) is now solved in current Git version (to be the 1.0.2). To my knowledge this is the last real basic bug for a 1.0.3 delivery. As long as SHADE_SURF is not solve it is not smart to have a 1.1, isn't it ?
Call 1 : plot, indgen(10) & Window, 2
Call 2 : Window, 0 & plot, indgen(10) & Window, 2
Both windows should have the same size.
Call 1 and Call 2 give the same output, and the size of both windows (0 & 2) are the same
Call 1 and call 2 give the same output but the size the the 2 windows are different
When running that remotely, it is absolutely clear that the first window is plotted in two steps, the first one has the good size (but with a white/grey background) then is resized to a bad size and a black background.
This problem is maybe related to a side effect within the window manager. I do use
xfwm4
and when I double the size of the top bar, the size of the first window is altered by the same factor.About Callgrind
I also try to use
valgrind --tool=callgrind ./install/bin/gdl
(yes I stored the outputs, so called callgrind.out.1234, 1234 is the pid of the run as usual withvalgrind
. To be explore withkcachegrind
. Since these files are big (5 Mo each) I cannot put them here ) and it seems, following the graph of the internal calls, that Call 1 and Call 2 don't follow the same path to reach the same final status (in both cases, X11 & wx). I do see that as a risk of bugs and future problems for the maintenance of the code.The text was updated successfully, but these errors were encountered: