On Mon, 4 May 2015 05:33:47 -0700 (PDT) "Edward K. Ream" <[email protected]> wrote:
> First, saying no-to-all just hard crashed Leo on Windows. This is > not ok with me. We should either fix this or get rid of the "to all" > options. I know reproducibility is a problem here, but last time we worked on this the bug was not with 'yes to all', it was in code quite distant from it, and could have been manifested by clicking yes every time. I think it had something to do with assumptions about things being initialized already when loading outlines after the first that were invalidated if loading the first had not proceeded as usual. All the yes to all code does is manage a tri-state bool that is None (i.e. ask) or yes or no (don't ask). It doesn't change program flow vs. just clicking yes every time. I typically have 4+, sometimes 10 outlines open, and I sometimes kill Leo unintentionally (virtual machines, vnc sessions) - having to click 8-20 times to start up, instead of two, would be painful. So I'm hoping that either there's a way to make this reproducible and debuggable, or at least you can audit (which in my experience typically means re-write ;-) the yes-to-all code to see that it's innocent here. > Second, it appears that in PyQt5 the correct call is > aLayout.setContentsMargins(QtCore.QMargins(0,0,0,0)) rather than > aLayout.setMargin(0). I just made this change in the python_terminal > plugin. Not sure why this hasn't caused problems in the Qt gui code. > Do you have any objections to me changing it? Sure, do whatever you want to it - it's Jake's code, not mine ;-) :-) Cheers -Terry > Edward > -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/leo-editor. For more options, visit https://groups.google.com/d/optout.
