On Mo, 25 Feb 2019, Robert Webb wrote:
> [Note: a third patch below]
>
> Hmm, hard to see how those changes relate to either of those issues. The
> description of their solutions don't directly relate either. Maybe other
> changes were what fixed these things?
>
> The change to set_fraction() for 7.4.280 looks a bit nonsensical too.
> w_fraction is an integer, scaled up by a large number FRACTION_MULT to allow
> it to act like a fixed point float. At the time I wrote that, there were no
> floats used anywhere in vim, and it seemed risky
> somehow to start using them just for this, cross-platform-wise. Are floats
> used elsewhere now? If so we could make this a float instead which would
> simplify the math.
>
> Anyway, the scaling looks wrong now (it is mixing the scaled-up ints with
> ints that haven't been scaled up, and treating them the same). Could
> probably revert the change in set_fraction() and it would be better.
>
> In fact I do still see a bug with the w_fraction stuff. Resize down to two
> lines, move cursor from top line to bottom line, resize larger, then resize
> back down to 2 lines. Now cursor is on top line instead of bottom where it
> started. I suspect reverting set_fraction()
> will fix that...
> Yep, tried it and it fixed it. Patch below (please combine with previous
> patch).
>
> The other changes in the fix seem separate, and are probably what's fixing
> the 7.4.280 issue.
>
> The changes for 7.4.365 don't appear related. They change the indenting on
> the two calls to set_topline() in win_new_height(), but didn't move them
> inside the two if-branches. That must have happened earlier.
> Ah, it was 7.4.309. I think it's the added "if (sline > 0)" check that fixes
> that, not moving the call to set_topline() which I think is an error.
>
> So I suspect my three patches (including the one below) should be safe. Are
> there tests for 7.4.280 and 7.4.309 to confirm?
Sorry, I am a bit confused now. It is probably best to submit a squashed
complete patch. If you can, try to create a test case for those fixed
issues (or at least describe precisely how to reproduce the issue
starting from vim --clean, then we can probably make a new-style test
for it to make sure it won't regress).
Best,
Christian
--
Ein Meinungsaustausch ist, wenn ein Beamter mit seiner Meinung zu
seinem Vorgesetzten geht und mit dessen Meinung zurückkommt.
-- Andrej Gromyko
--
--
You received this message from the "vim_dev" maillist.
Do not top-post! Type your reply below the text you are replying to.
For more information, visit http://www.vim.org/maillist.php
---
You received this message because you are subscribed to the Google Groups
"vim_dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
