https://bugs.documentfoundation.org/show_bug.cgi?id=146366

Miklos Vajna <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|regression                  |implementationError
      Regression By|Miklos Vajna                |

--- Comment #3 from Miklos Vajna <[email protected]> ---
I would argue that this is indeed a bug, but it's not a regression. For one,
this only "worked" for a short period of time, commit
fa39e7970496537258eaad1f5351db2d675225b6 was a fix for bug 72486, which was
itself a regression, so this only seemed to work after the initial work for
select-all vs tables at doc start and that commit.

For another, here are steps to reproduce this before the above commit:

1) Take e.g. linux-50max.git, check out
f9d087f66b13035ff036ec04cf650f468227af32, i.e. the commit before core.git
fa39e7970496537258eaad1f5351db2d675225b6.

2) Before you do ctrl-a (step 2 above), press enter to create an empty
paragraph before the table, then press the down arrow to enter to the
Table11:A1.

3) Then continue with the same steps: some unwanted layout frames will still
present after undo, even before core.git
fa39e7970496537258eaad1f5351db2d675225b6.

So yes, this is a bug, it is worth fixing, but it's not a regression from the
above commit, I would say. Adjusting keywords accordingly.

I looked at the actual problem a bit, but it's tricky, somehow it's around how
the undo for SwXText::convertToTable() works on an already existing document,
so RTF import and ODF paste is not affected, just RTF paste. Would need more
time to find out what's going on there.

-- 
You are receiving this mail because:
You are the assignee for the bug.

Reply via email to