It turns out the '-m pdb' workaround for the readline crash does work for me running unit tests.
I decided I wanted to follow the good practice of adding a failing unit test which exposes the bug and then show after the fix that same test passes. I discovered there are already unit tests for other group undo commands like convertAllBlanks and convertAllTabs and I wondered how those tests were passing. It turns out all the undo/redo tests start with a fresh undo stack. The off-by-one error of this bug causes the bead index to be -1 instead of 0. But for a single element list array, indices -1 and 0 point to the same list element. So the group redo will always function properly when it is the only item on the list. In order to duplicate the bug, I will need to write a test containing multiple commands. I might be out of time for this for a few days, but I plan to tackle it when I get a chance. On Wed, Aug 21, 2019 at 7:45 AM Brian Theado <[email protected]> wrote: > Sure, I'll do that or figure out the segfault issue sometime this week. > > On Tue, Aug 20, 2019 at 7:26 AM Edward K. Ream <[email protected]> > wrote: > >> Unfortunately, I'm not able to run unit tests without getting a segfault. >>> >> Alright. Create a new branch, and I'll run the unit tests there. >> > -- 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 view this discussion on the web visit https://groups.google.com/d/msgid/leo-editor/CAO5X8Cxs_P3MHsYWw1eKC9tSNZug9XexBPezsOk0aR%2BqBt7mVQ%40mail.gmail.com.
