X error and vim dies
64 views
Skip to first unread message
Manfred Lotz
May 5, 2020, 1:27:56 AM5/5/20
to vim...@googlegroups.com
This is Fedora 32, Xfce4 4.14.3 and gvim is
VIM - Vi IMproved 8.2 (2019 Dec 12, compiled Apr 17 2020 00:00:00)
Included patches: 1-587
I start gvim from terminal (in order to see the error messages) and do
shade. Then I get:
BadMatch (invalid parameter attributes)
Vim: Got X error
Vim: Finished.
I use CopyQ clipboard manager but it happens as well if CopyQ is
stopped.
guioptions are: egimrLtT!
--
Manfred
VIM - Vi IMproved 8.2 (2019 Dec 12, compiled Apr 17 2020 00:00:00)
Included patches: 1-587
I start gvim from terminal (in order to see the error messages) and do
shade. Then I get:
BadMatch (invalid parameter attributes)
Vim: Got X error
Vim: Finished.
I use CopyQ clipboard manager but it happens as well if CopyQ is
stopped.
guioptions are: egimrLtT!
--
Manfred
Shlomi Fish
May 5, 2020, 1:41:55 AM5/5/20
to Manfred Lotz, vim...@googlegroups.com
Hi Manfred!
I can try to reproduce it on my own fedora 32 x86-64 system. A few things to
try:
1. What is your hardware? (`hwinfo --short` )
2. Does it happen in a new user?
3. Does it happen in icewm and/or jwm?
4. Does it happen with `gvim -u NONE -U NONE`?
--
Shlomi Fish https://www.shlomifish.org/
My Favourite FOSS - https://www.shlomifish.org/open-source/favourite/
<Su-Shee> Botje: “you can’t kill Botje” isn’t Newton’s first law.
<Su-Shee> It’s not even the 5th.
— https://www.shlomifish.org/humour/fortunes/sharp-perl.html
Please reply to list if it's a mailing list post - https://shlom.in/reply .
try:
1. What is your hardware? (`hwinfo --short` )
2. Does it happen in a new user?
3. Does it happen in icewm and/or jwm?
4. Does it happen with `gvim -u NONE -U NONE`?
--
Shlomi Fish https://www.shlomifish.org/
My Favourite FOSS - https://www.shlomifish.org/open-source/favourite/
<Su-Shee> Botje: “you can’t kill Botje” isn’t Newton’s first law.
<Su-Shee> It’s not even the 5th.
— https://www.shlomifish.org/humour/fortunes/sharp-perl.html
Please reply to list if it's a mailing list post - https://shlom.in/reply .
Manfred Lotz
May 5, 2020, 4:20:27 AM5/5/20
to vim...@googlegroups.com
Hi Shlomi,
On Tue, 5 May 2020 08:41:40 +0300
Shlomi Fish <shl...@shlomifish.org> wrote:
> Hi Manfred!
>
> On Tue, 5 May 2020 07:27:43 +0200
> Manfred Lotz <ml_...@posteo.de> wrote:
>
> > This is Fedora 32, Xfce4 4.14.3 and gvim is
> >
> > VIM - Vi IMproved 8.2 (2019 Dec 12, compiled Apr 17 2020 00:00:00)
> > Included patches: 1-587
> >
> >
> > I start gvim from terminal (in order to see the error messages) and
> > do shade. Then I get:
> >
> > BadMatch (invalid parameter attributes)
> > Vim: Got X error
> > Vim: Finished.
> >
> >
> > I use CopyQ clipboard manager but it happens as well if CopyQ is
> > stopped.
> >
> > guioptions are: egimrLtT!
> >
> >
>
> I can try to reproduce it on my own fedora 32 x86-64 system. A few
> things to try:
>
> 1. What is your hardware? (`hwinfo --short` )
>
> 2. Does it happen in a new user?
>
Yes, the same behavior.
> 3. Does it happen in icewm and/or jwm?
>
Don't have them installed but I tested in Gnome, and it doesn't happen
in Gnome!
> 4. Does it happen with `gvim -u NONE -U NONE`?
>
Yes, it happens as well when starting a naked gvim in Xfce4.
So it looks like an issue in Xfce4. Perhaps gtk3 related?
--
Manfred
On Tue, 5 May 2020 08:41:40 +0300
Shlomi Fish <shl...@shlomifish.org> wrote:
> Hi Manfred!
>
> On Tue, 5 May 2020 07:27:43 +0200
> Manfred Lotz <ml_...@posteo.de> wrote:
>
> > This is Fedora 32, Xfce4 4.14.3 and gvim is
> >
> > VIM - Vi IMproved 8.2 (2019 Dec 12, compiled Apr 17 2020 00:00:00)
> > Included patches: 1-587
> >
> >
> > I start gvim from terminal (in order to see the error messages) and
> > do shade. Then I get:
> >
> > BadMatch (invalid parameter attributes)
> > Vim: Got X error
> > Vim: Finished.
> >
> >
> > I use CopyQ clipboard manager but it happens as well if CopyQ is
> > stopped.
> >
> > guioptions are: egimrLtT!
> >
> >
>
> I can try to reproduce it on my own fedora 32 x86-64 system. A few
> things to try:
>
> 1. What is your hardware? (`hwinfo --short` )
>
> 2. Does it happen in a new user?
>
> 3. Does it happen in icewm and/or jwm?
>
in Gnome!
> 4. Does it happen with `gvim -u NONE -U NONE`?
>
So it looks like an issue in Xfce4. Perhaps gtk3 related?
--
Manfred
Shlomi Fish
May 5, 2020, 5:09:25 AM5/5/20
to Manfred Lotz, vim...@googlegroups.com
On Tue, 5 May 2020 10:20:18 +0200
a new user, and it doesn't happen in my normal user in kde plasma5.
my hardware is
```
cpu:
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3710 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3706 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3725 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3770 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3766 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3761 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3742 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3704 MHz
keyboard:
/dev/input/event3 Logitech Keyboard K120
mouse:
/dev/input/mice Logitech M105 Optical Mouse
monitor:
AOC 22E1W
graphics card:
Intel Iris Plus Graphics 655
sound:
Intel Cannon Point-LP High Definition Audio Controller
storage:
Intel Cannon Point-LP SATA Controller [AHCI Mode]
Non-Volatile memory controller
network:
wlp0s20f3 Intel Cannon Point-LP CNVi [Wireless-AC]
eno1 Intel Ethernet Connection (6) I219-V
network interface:
lo Loopback network interface
wlp0s20f3 Ethernet network interface
eno1 Ethernet network interface
disk:
/dev/nvme0n1 Disk
/dev/sda TOSHIBA HDWL110
partition:
/dev/nvme0n1p1 Partition
/dev/sda1 Partition
/dev/sda2 Partition
/dev/sda3 Partition
usb controller:
Intel Cannon Point-LP USB 3.1 xHCI Controller
bios:
BIOS
bridge:
Intel Cannon Point-LP PCI Express Root Port #1
Intel Cannon Point-LP LPC Controller
Intel Cannon Point-LP PCI Express Root Port #15
Intel 8th Gen Core Processor Host Bridge/DRAM Registers
Intel Cannon Point-LP PCI Express Root Port #9
Intel Cannon Point-LP PCI Express Root Port #5
hub:
Linux Foundation 2.0 root hub
Linux Foundation 3.0 root hub
memory:
Main Memory
bluetooth:
Intel Bluetooth Device
unknown:
FPU
DMA controller
PIC
Keyboard controller
Intel Xeon E3-1200 v5/v6 / E3-1500 v5 / 6th/7th Gen Core
Processor Gaussian Mixture Model
Intel Cannon Point-LP MEI Controller #1
Realtek RTS522A PCI Express Card Reader
Intel Cannon Point-LP SPI Controller
Intel Cannon Point-LP Thermal Controller
Intel Cannon Point-LP Shared SRAM
Intel Cannon Point-LP SMBus Controller
/dev/input/event5 Logitech Keyboard K120
```
>
> > 3. Does it happen in icewm and/or jwm?
> >
>
> Don't have them installed but I tested in Gnome, and it doesn't happen
> in Gnome!
>
>
> > 4. Does it happen with `gvim -u NONE -U NONE`?
> >
>
> Yes, it happens as well when starting a naked gvim in Xfce4.
>
>
> So it looks like an issue in Xfce4. Perhaps gtk3 related?
I think gnome3 uses gtk3 too.
Parody of "The Fountainhead" - https://shlom.in/towtf
BYTE editors are men who seperate the wheat from the chaff, and then
carefully print the chaff.
-- based on Lionel Hummel (uiucdcs!hummel), derived from a quote by Adlai
Stevenson, Sr.
Manfred Lotz <ml_...@posteo.de> wrote:
> Hi Shlomi,
>
> On Tue, 5 May 2020 08:41:40 +0300
> Shlomi Fish <shl...@shlomifish.org> wrote:
>
> > Hi Manfred!
> >
> > On Tue, 5 May 2020 07:27:43 +0200
> > Manfred Lotz <ml_...@posteo.de> wrote:
> >
> > > This is Fedora 32, Xfce4 4.14.3 and gvim is
> > >
> > > VIM - Vi IMproved 8.2 (2019 Dec 12, compiled Apr 17 2020 00:00:00)
> > > Included patches: 1-587
> > >
> > >
> > > I start gvim from terminal (in order to see the error messages) and
> > > do shade. Then I get:
> > >
> > > BadMatch (invalid parameter attributes)
> > > Vim: Got X error
> > > Vim: Finished.
> > >
> > >
> > > I use CopyQ clipboard manager but it happens as well if CopyQ is
> > > stopped.
> > >
> > > guioptions are: egimrLtT!
> > >
> > >
> >
> > I can try to reproduce it on my own fedora 32 x86-64 system. A few
> > things to try:
> >
> > 1. What is your hardware? (`hwinfo --short` )
> >
> > 2. Does it happen in a new user?
> >
>
> Yes, the same behavior.
>
Thanks! I can also reproduce this issue in xfce on my fedora 32 x86-64 system in
> Hi Shlomi,
>
> On Tue, 5 May 2020 08:41:40 +0300
> Shlomi Fish <shl...@shlomifish.org> wrote:
>
> > Hi Manfred!
> >
> > On Tue, 5 May 2020 07:27:43 +0200
> > Manfred Lotz <ml_...@posteo.de> wrote:
> >
> > > This is Fedora 32, Xfce4 4.14.3 and gvim is
> > >
> > > VIM - Vi IMproved 8.2 (2019 Dec 12, compiled Apr 17 2020 00:00:00)
> > > Included patches: 1-587
> > >
> > >
> > > I start gvim from terminal (in order to see the error messages) and
> > > do shade. Then I get:
> > >
> > > BadMatch (invalid parameter attributes)
> > > Vim: Got X error
> > > Vim: Finished.
> > >
> > >
> > > I use CopyQ clipboard manager but it happens as well if CopyQ is
> > > stopped.
> > >
> > > guioptions are: egimrLtT!
> > >
> > >
> >
> > I can try to reproduce it on my own fedora 32 x86-64 system. A few
> > things to try:
> >
> > 1. What is your hardware? (`hwinfo --short` )
> >
> > 2. Does it happen in a new user?
> >
>
> Yes, the same behavior.
>
a new user, and it doesn't happen in my normal user in kde plasma5.
my hardware is
```
cpu:
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3710 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3706 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3725 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3770 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3766 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3761 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3742 MHz
Intel(R) Core(TM) i5-8259U CPU @ 2.30GHz, 3704 MHz
keyboard:
/dev/input/event3 Logitech Keyboard K120
mouse:
/dev/input/mice Logitech M105 Optical Mouse
monitor:
AOC 22E1W
graphics card:
Intel Iris Plus Graphics 655
sound:
Intel Cannon Point-LP High Definition Audio Controller
storage:
Intel Cannon Point-LP SATA Controller [AHCI Mode]
Non-Volatile memory controller
network:
wlp0s20f3 Intel Cannon Point-LP CNVi [Wireless-AC]
eno1 Intel Ethernet Connection (6) I219-V
network interface:
lo Loopback network interface
wlp0s20f3 Ethernet network interface
eno1 Ethernet network interface
disk:
/dev/nvme0n1 Disk
/dev/sda TOSHIBA HDWL110
partition:
/dev/nvme0n1p1 Partition
/dev/sda1 Partition
/dev/sda2 Partition
/dev/sda3 Partition
usb controller:
Intel Cannon Point-LP USB 3.1 xHCI Controller
bios:
BIOS
bridge:
Intel Cannon Point-LP PCI Express Root Port #1
Intel Cannon Point-LP LPC Controller
Intel Cannon Point-LP PCI Express Root Port #15
Intel 8th Gen Core Processor Host Bridge/DRAM Registers
Intel Cannon Point-LP PCI Express Root Port #9
Intel Cannon Point-LP PCI Express Root Port #5
hub:
Linux Foundation 2.0 root hub
Linux Foundation 3.0 root hub
memory:
Main Memory
bluetooth:
Intel Bluetooth Device
unknown:
FPU
DMA controller
PIC
Keyboard controller
Intel Xeon E3-1200 v5/v6 / E3-1500 v5 / 6th/7th Gen Core
Processor Gaussian Mixture Model
Intel Cannon Point-LP MEI Controller #1
Realtek RTS522A PCI Express Card Reader
Intel Cannon Point-LP SPI Controller
Intel Cannon Point-LP Thermal Controller
Intel Cannon Point-LP Shared SRAM
Intel Cannon Point-LP SMBus Controller
/dev/input/event5 Logitech Keyboard K120
```
>
> > 3. Does it happen in icewm and/or jwm?
> >
>
> Don't have them installed but I tested in Gnome, and it doesn't happen
> in Gnome!
>
>
> > 4. Does it happen with `gvim -u NONE -U NONE`?
> >
>
> Yes, it happens as well when starting a naked gvim in Xfce4.
>
>
> So it looks like an issue in Xfce4. Perhaps gtk3 related?
Parody of "The Fountainhead" - https://shlom.in/towtf
BYTE editors are men who seperate the wheat from the chaff, and then
carefully print the chaff.
-- based on Lionel Hummel (uiucdcs!hummel), derived from a quote by Adlai
Stevenson, Sr.
Manfred Lotz
May 5, 2020, 5:33:50 AM5/5/20
to vim...@googlegroups.com
On Tue, 5 May 2020 12:05:26 +0300
Very good. :-)
My hardware
cpu:
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800
MHz Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800 MHz
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800
MHz Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800 MHz
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800
MHz Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800 MHz
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800
MHz Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800 MHz
keyboard:
/dev/input/event3 AT Translated Set 2 keyboard
mouse:
/dev/input/mice ELAN0501:00 04F3:303F Mouse
/dev/input/mice ELAN0501:00 04F3:303F Touchpad
monitor:
AUO LCD Monitor
graphics card:
nVidia GP106M [GeForce GTX 1050 Ti Mobile]
Intel HD Graphics 630
sound:
Intel CM238 HD Audio Controller
storage:
Intel HM170/QM170 Chipset SATA Controller [AHCI
Mode] Samsung Electronics NVMe SSD Controller SM961/PM961
network:
enp3s0 Realtek RTL8111/8168/8411 PCI Express Gigabit
Ethernet Controller wlp2s0 Qualcomm Atheros QCA6174
802.11ac Wireless Network Adapter network interface:
wlp2s0 Ethernet network interface
virbr0-nic Ethernet network interface
lo Loopback network interface
virbr0 Ethernet network interface
enp3s0 Ethernet network interface
disk:
/dev/nvme0n1 Samsung Electronics NVMe SSD Controller
SM961/PM961 /dev/sda Crucial_CT1050MX
partition:
/dev/nvme0n1p1 Partition
/dev/nvme0n1p2 Partition
/dev/nvme0n1p3 Partition
/dev/nvme0n1p4 Partition
usb controller:
Intel 100 Series/C230 Series Chipset Family USB
3.0 xHCI Controller bios:
BIOS
bridge:
Intel 100 Series/C230 Series Chipset Family PCI
Express Root Port #3 Intel HM175 Chipset LPC/eSPI Controller
Intel Xeon E3-1200 v5/E3-1500 v5/6th Gen Core
Processor PCIe Controller (x16) Intel 100 Series/C230 Series Chipset
Family PCI Express Root Port #4 Intel Xeon E3-1200 v6/7th Gen Core
Processor Host Bridge/DRAM Registers Intel 100 Series/C230 Series
Chipset Family PCI Express Root Port #9 Intel 100 Series/C230 Series
Chipset Family PCI Express Root Port #5 hub:
Power Management Controller Intel 100 Series/C230 Series Chipset Family
Serial IO I2C Controller #1 Intel 100 Series/C230 Series Chipset Family
MEI Controller #1 Intel 100 Series/C230 Series Chipset Family Serial IO
I2C Controller #2 Intel 100 Series/C230 Series Chipset Family Serial IO
I2C Controller #0 Intel 100 Series/C230 Series Chipset Family Thermal
Subsystem Intel 100 Series/C230 Series Chipset Family SMBus
/dev/input/event15 Chicony Electronics HD WebCam
Realtek RTS5129 Card Reader Controller
Elan Microelectronics ELAN:Fingerprint
> >
> > So it looks like an issue in Xfce4. Perhaps gtk3 related?
>
> I think gnome3 uses gtk3 too.
>
Yes, you may be right.
--
Manfred
cpu:
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800
MHz Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800 MHz
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800
MHz Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800 MHz
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800
MHz Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800 MHz
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800
MHz Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 2800 MHz
keyboard:
/dev/input/event3 AT Translated Set 2 keyboard
mouse:
/dev/input/mice ELAN0501:00 04F3:303F Mouse
/dev/input/mice ELAN0501:00 04F3:303F Touchpad
monitor:
AUO LCD Monitor
graphics card:
nVidia GP106M [GeForce GTX 1050 Ti Mobile]
Intel HD Graphics 630
sound:
Intel CM238 HD Audio Controller
storage:
Intel HM170/QM170 Chipset SATA Controller [AHCI
Mode] Samsung Electronics NVMe SSD Controller SM961/PM961
network:
enp3s0 Realtek RTL8111/8168/8411 PCI Express Gigabit
Ethernet Controller wlp2s0 Qualcomm Atheros QCA6174
802.11ac Wireless Network Adapter network interface:
wlp2s0 Ethernet network interface
virbr0-nic Ethernet network interface
lo Loopback network interface
virbr0 Ethernet network interface
enp3s0 Ethernet network interface
disk:
/dev/nvme0n1 Samsung Electronics NVMe SSD Controller
SM961/PM961 /dev/sda Crucial_CT1050MX
partition:
/dev/nvme0n1p1 Partition
/dev/nvme0n1p2 Partition
/dev/nvme0n1p3 Partition
/dev/nvme0n1p4 Partition
usb controller:
Intel 100 Series/C230 Series Chipset Family USB
3.0 xHCI Controller bios:
BIOS
bridge:
Intel 100 Series/C230 Series Chipset Family PCI
Express Root Port #3 Intel HM175 Chipset LPC/eSPI Controller
Intel Xeon E3-1200 v5/E3-1500 v5/6th Gen Core
Processor PCIe Controller (x16) Intel 100 Series/C230 Series Chipset
Family PCI Express Root Port #4 Intel Xeon E3-1200 v6/7th Gen Core
Processor Host Bridge/DRAM Registers Intel 100 Series/C230 Series
Chipset Family PCI Express Root Port #9 Intel 100 Series/C230 Series
Chipset Family PCI Express Root Port #5 hub:
Linux Foundation 2.0 root hub
Linux Foundation 3.0 root hub
memory:
Main Memory
bluetooth:
Lite-On Bluetooth Device
Linux Foundation 3.0 root hub
memory:
Main Memory
bluetooth:
unknown:
FPU
DMA controller
PIC
Keyboard controller
Intel 100 Series/C230 Series Chipset Family
FPU
DMA controller
PIC
Keyboard controller
Power Management Controller Intel 100 Series/C230 Series Chipset Family
Serial IO I2C Controller #1 Intel 100 Series/C230 Series Chipset Family
MEI Controller #1 Intel 100 Series/C230 Series Chipset Family Serial IO
I2C Controller #2 Intel 100 Series/C230 Series Chipset Family Serial IO
I2C Controller #0 Intel 100 Series/C230 Series Chipset Family Thermal
Subsystem Intel 100 Series/C230 Series Chipset Family SMBus
/dev/input/event15 Chicony Electronics HD WebCam
Realtek RTS5129 Card Reader Controller
Elan Microelectronics ELAN:Fingerprint
> >
> > So it looks like an issue in Xfce4. Perhaps gtk3 related?
>
> I think gnome3 uses gtk3 too.
>
--
Manfred
John Little
May 6, 2020, 2:31:02 AM5/6/20
to vim_use
If this problem has consequences (vim dying in the middle of your work can be bad), if you build your own vim a workaround is to modify os_unix.c in the x_error_handler function, to replace the preserve_exit() call with
g_print("%s", IObuff);
or just comment out the preserve_exit() call. I have used this workaround for a BadWindow error for about a decade, with no ill effect.
Regards, John Little
Manfred Lotz
May 9, 2020, 6:47:48 AM5/9/20
to vim...@googlegroups.com
official vim sources?
--
Manfred
Bram Moolenaar
May 9, 2020, 10:12:10 AM5/9/20
to vim...@googlegroups.com, Manfred Lotz
it might be possible to silently continue.
Let's make it a bit safer by preserving swap files, in case X does
decide to exit or we crash.
In a terminal we can show the message. The comment mentions that in the
GUI, using the X server may fail in the error handler. So we'll not
show the message.
--
hundred-and-one symptoms of being an internet addict:
75. You start wondering whether you could actually upgrade your brain
with a Pentium Pro microprocessor 80. The upgrade works just fine.
/// Bram Moolenaar -- Br...@Moolenaar.net -- http://www.Moolenaar.net \\\
/// sponsor Vim, vote for features -- http://www.Vim.org/sponsor/ \\\
\\\ an exciting new programming language -- http://www.Zimbu.org ///
\\\ help me help AIDS victims -- http://ICCF-Holland.org ///
Manfred Lotz
May 9, 2020, 11:40:25 AM5/9/20
to vim...@googlegroups.com
On Sat, 09 May 2020 16:11:57 +0200
Bram Moolenaar <Br...@moolenaar.net> wrote:
> Manfred Lotz wrote:
>
> > On Tue, 5 May 2020 23:31:02 -0700 (PDT)
> > John Little <john.b...@gmail.com> wrote:
> >
> > > If this problem has consequences (vim dying in the middle of your
> > > work can be bad), if you build your own vim a workaround is to
> > > modify os_unix.c in the x_error_handler function, to replace the
> > > preserve_exit() call with
> > >
> > > g_print("%s", IObuff);
> > >
> > > or just comment out the preserve_exit() call. I have used this
> > > workaround for a BadWindow error for about a decade, with no ill
> > > effect.
> > >
> >
> > Sounds like a good idea. But why isn't this changed this way in the
> > official vim sources?
>
> This is "not supposed to happen". But there was already a comment
> that it might be possible to silently continue.
>
That would be great.
Bram Moolenaar <Br...@moolenaar.net> wrote:
> Manfred Lotz wrote:
>
> > On Tue, 5 May 2020 23:31:02 -0700 (PDT)
> > John Little <john.b...@gmail.com> wrote:
> >
> > > If this problem has consequences (vim dying in the middle of your
> > > work can be bad), if you build your own vim a workaround is to
> > > modify os_unix.c in the x_error_handler function, to replace the
> > > preserve_exit() call with
> > >
> > > g_print("%s", IObuff);
> > >
> > > or just comment out the preserve_exit() call. I have used this
> > > workaround for a BadWindow error for about a decade, with no ill
> > > effect.
> > >
> >
> > Sounds like a good idea. But why isn't this changed this way in the
> > official vim sources?
>
> This is "not supposed to happen". But there was already a comment
> that it might be possible to silently continue.
>
> Let's make it a bit safer by preserving swap files, in case X does
> decide to exit or we crash.
>
> In a terminal we can show the message. The comment mentions that in
> the GUI, using the X server may fail in the error handler. So we'll
> not show the message.
>
see what's going on and got the messages I wanted to see.
--
Manfred
Reply all
Reply to author
Forward
0 new messages