Issue 389 in vim: 'linebreak' breaks c% if the last visual selection was block
71 views
Skip to first unread message
v...@googlecode.com
Jul 30, 2015, 5:21:51 AM7/30/15
to vim...@vim.org
Status: New
Owner: ----
Labels: Type-Defect Priority-Medium
New issue 389 by chris.morganiser: 'linebreak' breaks c% if the last visual
selection was block
https://code.google.com/p/vim/issues/detail?id=389
Précis: when the 'linebreak' option is set and visual block mode was the
last visual mode entered, c% removes the text for deletion in a blockwise
manner rather than the character-wise manner that should be the case.
Detailed steps to reproduce (no special configuration is required;
vim -u NONE exhibits the buggy behaviour):
Enter this demonstrative text:
abcd{ef
ghijklm
no}pqrs
Go to the opening {.
Enter and exit visual block mode (<C-V><C-V>).
At some point no later than this, :set linebreak.
c%<Esc>
The expected contents of the buffer is this:
abcdpqrs
The actual contents of the buffer is this:
abef
ghlm
nors
I haven’t experimented much more, but d% is not buggy.
I am using Vim 7.4.778 on Arch Linux, 64-bit (huge with GTK2 GUI).
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
Owner: ----
Labels: Type-Defect Priority-Medium
New issue 389 by chris.morganiser: 'linebreak' breaks c% if the last visual
selection was block
https://code.google.com/p/vim/issues/detail?id=389
Précis: when the 'linebreak' option is set and visual block mode was the
last visual mode entered, c% removes the text for deletion in a blockwise
manner rather than the character-wise manner that should be the case.
Detailed steps to reproduce (no special configuration is required;
vim -u NONE exhibits the buggy behaviour):
Enter this demonstrative text:
abcd{ef
ghijklm
no}pqrs
Go to the opening {.
Enter and exit visual block mode (<C-V><C-V>).
At some point no later than this, :set linebreak.
c%<Esc>
The expected contents of the buffer is this:
abcdpqrs
The actual contents of the buffer is this:
abef
ghlm
nors
I haven’t experimented much more, but d% is not buggy.
I am using Vim 7.4.778 on Arch Linux, 64-bit (huge with GTK2 GUI).
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
Arnaud Decara
Jul 31, 2015, 9:45:23 AM7/31/15
to vim_dev, vim...@vim.org, codesite...@google.com, v...@googlecode.com, axel....@cip-kommunal.de
Hi all,
Please find attached a fix.
(Both issues with the change operator and 's' should
be fixed; undiscovered similar issues that might have
occured should be fixed too.)
Best,
Arnaud
Christian Brabandt
Jul 31, 2015, 10:03:21 AM7/31/15
to vim...@vim.org
Hi Arnaud!
On Fr, 31 Jul 2015, Arnaud Decara wrote:
> Please find attached a fix.
> (Both issues with the change operator and 's' should
> be fixed; undiscovered similar issues that might have
> occured should be fixed too.)
I believe that breaks test_listlbr.out
Here is a slightly different approach, including a test. The idea is, to
only recalculate the oap->start_vcol and oap->end_vcol columns, if the
cursor is on a wrapped line, else that shouldn't be necessary. However,
I am not completely sure, this is the best solution.
Best,
Christian
--
Der Mensch, der sich vielen Glückes und seiner Schwäche bewußt ist,
wird abergläubisch, flüchtet zum Gebet.
-- Georg Christoph Lichtenberg
(Vermischte Schriften - Beobachtung über den Menschen)
On Fr, 31 Jul 2015, Arnaud Decara wrote:
> Please find attached a fix.
> (Both issues with the change operator and 's' should
> be fixed; undiscovered similar issues that might have
> occured should be fixed too.)
Here is a slightly different approach, including a test. The idea is, to
only recalculate the oap->start_vcol and oap->end_vcol columns, if the
cursor is on a wrapped line, else that shouldn't be necessary. However,
I am not completely sure, this is the best solution.
Best,
Christian
--
Der Mensch, der sich vielen Glückes und seiner Schwäche bewußt ist,
wird abergläubisch, flüchtet zum Gebet.
-- Georg Christoph Lichtenberg
(Vermischte Schriften - Beobachtung über den Menschen)
Arnaud Decara
Jul 31, 2015, 10:33:21 AM7/31/15
to vim_dev, vim...@vim.org, cbl...@256bit.org
Hello Christian!
Maybe I'm not using correctly the tests but test_librl does not seem to be
broken here.
(When I run all the tests via the Makefile, i get "Test results : ALL DONE",
and when I remove test_listlbr.out and rerun the test separately I don't get
any error)
Am I doing something wrong?
(But your idea might be better, I'll look at it)
Arnaud
broken here.
(When I run all the tests via the Makefile, i get "Test results : ALL DONE",
and when I remove test_listlbr.out and rerun the test separately I don't get
any error)
Am I doing something wrong?
(But your idea might be better, I'll look at it)
Arnaud
Christian Brabandt
Jul 31, 2015, 10:37:54 AM7/31/15
to vim_dev, vim...@vim.org
Hi Arnaud!
On Fr, 31 Jul 2015, Arnaud Decara wrote:
> Hello Christian!
>
> On Friday, 31 July 2015 16:03:21 UTC+2, Christian Brabandt wrote:
> > Hi Arnaud!
> >
> > On Fr, 31 Jul 2015, Arnaud Decara wrote:
> >
> > > Please find attached a fix.
> > > (Both issues with the change operator and 's' should
> > > be fixed; undiscovered similar issues that might have
> > > occured should be fixed too.)
> >
> > I believe that breaks test_listlbr.out
> >
> > Here is a slightly different approach, including a test. The idea is, to
> > only recalculate the oap->start_vcol and oap->end_vcol columns, if the
> > cursor is on a wrapped line, else that shouldn't be necessary. However,
> > I am not completely sure, this is the best solution.
> >
On Fr, 31 Jul 2015, Arnaud Decara wrote:
> Hello Christian!
>
> On Friday, 31 July 2015 16:03:21 UTC+2, Christian Brabandt wrote:
> > Hi Arnaud!
> >
> > On Fr, 31 Jul 2015, Arnaud Decara wrote:
> >
> > > Please find attached a fix.
> > > (Both issues with the change operator and 's' should
> > > be fixed; undiscovered similar issues that might have
> > > occured should be fixed too.)
> >
> > I believe that breaks test_listlbr.out
> >
> > Here is a slightly different approach, including a test. The idea is, to
> > only recalculate the oap->start_vcol and oap->end_vcol columns, if the
> > cursor is on a wrapped line, else that shouldn't be necessary. However,
> > I am not completely sure, this is the best solution.
> >
> Maybe I'm not using correctly the tests but test_librl does not seem
> to be broken here. (When I run all the tests via the Makefile, i get
> "Test results : ALL DONE", and when I remove test_listlbr.out and
> rerun the test separately I don't get any error)
> Am I doing something wrong?
That should be okay. I am just wondering, because for me that test
> to be broken here. (When I run all the tests via the Makefile, i get
> "Test results : ALL DONE", and when I remove test_listlbr.out and
> rerun the test separately I don't get any error)
> Am I doing something wrong?
failed (which was my first idea when I saw that problem).
Perhaps this was a problem just for me, so in that case your patch would
be preferable for its simplicity. The test should be included
nevertheless ;)
Best,
Christian
--
Mathematiker-Glücksrad: "Ich kaufe ein Epsilon."
Axel Bender
Aug 5, 2015, 4:54:32 AM8/5/15
to vim_dev, vim...@vim.org, codesite...@google.com, v...@googlecode.com
Will this get fixed, or did I miss something?
Not ok here with 7.4.811.
This is not a push, just a question :()
Christian Brabandt
Aug 5, 2015, 5:08:30 AM8/5/15
to vim...@googlegroups.com
Best,Christian
v...@googlecode.com
Aug 11, 2015, 12:48:17 PM8/11/15
to vim...@vim.org
Updates:
Status: Fixed
Comment #1 on issue 389 by brammool...@gmail.com: 'linebreak' breaks c% if
Fix included in patch 7.4.818
Status: Fixed
Comment #1 on issue 389 by brammool...@gmail.com: 'linebreak' breaks c% if
Fix included in patch 7.4.818
Reply all
Reply to author
Forward
0 new messages