Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Patrick Smith
Jordan, I tried a similar patch, and yes it does work for me. My message reporting that may have been delayed by moderation or something, I'm not sure. One drawback to this is that seems to disable the user's input method. If, for example, I do printf 'あ\nお' | ./dmenu (that contains Japanese hir

Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Patrick Smith
I tried patching the dmenu 4.9 source code similarly to https://www.mail-archive.com/fltk-bugs@easysw.com/msg01159.html. Result: printf 'foo\nbar' | ./dmenu now works correctly. And I do not need to move the mouse cursor before pressing the enter key. Of course, I don't know whether that would be

Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Jordan Timmerman
I believe I have a patch to fix this issue. It is attached. Taken from: https://www.mail-archive.com/fltk-bugs@easysw.com/msg01159.html The link was pointed to by the Servo issue that Patrick Smith linked to. (Thanks, Patrick!) Patrick, can you please try (from a clean 4.9 dmenu sourcetree) `git

Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Jordan Timmerman
Patrick, Can you also try reverting 377bd37e212b1ec4c03a481245603c6560d0be22 ? It certainly sounds like the same issue I experience; if so, reverting that commit will cause dmenu to work correctly again. This may be able to help guide your search if you continue debugging. On Sun, Feb 10, 2019 a

Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Patrick Smith
Another point of information... after seeing https://github.com/servo/servo/issues/9938, I tried these two commands with my instrumented dmenu: printf 'foo\nbar' | ./dmenu Calling XOpenIM Back from XOpenIM, xim=0x0 Calling XCreateIC Back from XCreateIC, xic=0x0 xic = 0x0, event type = 2 calli

Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Jordan Timmerman
With a little finagling, I've managed to produce a backtrace by using Quentin's suggested flag modifications to config.mk. This is attached. It seems to me only to confirm what we've already learned: the failure occurs in XmbLookupString. Perhaps with the additional diagnostic information Patrick w

Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Patrick Smith
I'm experiencing a problem that appears very similar to this one, and I can add a bit of information. Summary: When I have scim+anthy installed as an input method, the call to XOpenIM fails, returning a null pointer. But I don't know much about the internals of X, so can't be 100% sure this is the

Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Jordan Timmerman
The attachment is missing output. Forgot to `make clean` first. My bad. I've attached the corrected output. On Sun, Feb 10, 2019 at 5:35 PM Jordan Timmerman wrote: > > Quentin, > > dmenu does not build if I set the flags as you've suggested. I have > attached the output I receive; it fails to res

Re: [dev] [dmenu] 4.9 segfault on input

2019-02-10 Thread Jordan Timmerman
Quentin, dmenu does not build if I set the flags as you've suggested. I have attached the output I receive; it fails to resolve `#include `. It builds if run `make` by itself. On Sat, Feb 9, 2019 at 2:49 AM Quentin Rameau wrote: > > Hello Jordan, > > > Per the hacking page, I have attached the

[dev] [st] IBus is not working properly in st

2019-02-10 Thread Wolf
Greetings, I've decided to give a st a try but I've hit a problem quite quickly. It seems that IBus input is not handled completely correctly. While trying to input japanese inside rxvt-unicode (using IBus with Anthy), it shows the prompt around the cursor (urxvt.png). While for st, the prompt is d