-
Notifications
You must be signed in to change notification settings - Fork 295
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
Bizarre invalid selection / selection just gives up #336
Comments
I am observing this too. |
It's not just me! For reference this issue has never gone away, it just comes and goes sometimes. |
@EvanKomp Are you also observing this only on Mac? |
In my case, I have an object named |
I'm using Fedora 39 with Python 3.10.16 and PyMOL 3.1.3. |
I ran into this exact same issue. If I repeatedly paste the following into the command line on PyMOL 3.1.3 (incentive version) on macOS Sonoma, it will give a selection not found error every four times:
I think it may have something to do with the undo feature. The problem goes away by disabling undo support:
|
I can confirm that I see the error about every four times using that short script repeatedly and it does disappear after disabling undo. This is using incentive PyMOL 3.1.3.1 and macOS Sequoia 15.3. Disabling Undo before running the original script in the thread also fixes the issue. I'll create an internal ticket and investigate the issue, though this may take some time if it's some undefined behavior with Undo. Once I confirm this is definitely related to Undo, I'll close this ticket since it's an incentive only feature, but I will update this thread once there's a fix version. Thank you everyone for reporting this and helping to narrow down a potential cause! |
After the problem rears its head, it also appears to interfere with the "Action" button at least on objects where the selection is invalid. Pressing that button gives the following error:
If it's going to take a while to fix, I'd suggest putting a note somewhere obvious for incentive users that the "Undo" feature in the incentive version is buggy and should be disabled if users encounter problems. |
Here a excerpt from the console when I load the attached .pdb file without doing
|
I can confirm that the "undo" button seems to explain my problem as well. I had noticed that this problem went away a few months ago on it's own, but I think it's just because With undo disabled, I don't have this bug. With undo enabled, I have this bug. |
Hi,
I'm been using pymol for a long time, using original PyMOL 1 and PyMOL 2 on El Capitan. I recently had to update my mac to Monterey, and now, pymol crashes in bizarre ways.
Here's a minimal test snippet that will crash for me (regardless of what protein is displayed (even none)):
Notably, this doesn't crash on i == 0. It crashes on i == 11.
This snippet crashes in the gui (whether you 1-line it for the command prompt or put it in a script and call run) in both open source and closed source versions. I've tried both 2.5.7 and 2.5.8.
Once you get into this state. There's a broken selection called "test" that displays in the gui (and displays in cmd.get_names()), but clearly is broken because pymol throws an error when you click on the "A" button.

I do not get this crash or the glitchy selection when running these snippets from the commandline (using ipython and import pymol). It also does not crash when running in my .pymolrc and launching the gui.
Debugging
This snippet does not crash:
Then, it gets weirder. If you run that bad line multiple times, you have to fix it multiple times.
Here's a table with the minimum values of B I could get this to NOT crash at for a given value of A. These values are the same in both the closed and open source versions.
But once you call that command 15 times, it just works. (i.e. if A is >= 15, B can be 0)
Anyways, I'm happy to help get this bug fixed. Or, please just tell me how to remove all traces of pymol from my computer so I can reinstall. I've tested with some other people and these bugs don't exist. I don't know what's wrong with my install.
The text was updated successfully, but these errors were encountered: