Germany
https://herrhotzenplotz.de/gcli
FreeBSD guy
From Nico Sonack to ~herrhotzenplotz/gcli-devel
lhr@disroot.org wrote: > Escapes spaces in .Nm so eg. `gcli pulls' is one discrete unit. groff_mdoc(7) > doesn't like multiple arguments to .Nm; previously subsequent references > to .Nm (without arguments) would just show `gcli' rather than eg. `gcli pulls'. Thanks! What a silly mistake. I never noticed this bug since groff_mdoc isn't the default preprocessor on FreeBSD (which is using mandoc from OpenBSD). However when manually preprocessing the page with groff_mdoc the issue becomes very visible :D I applied the patch and pushed it. Nico
From Nico Sonack to ~herrhotzenplotz/gcli-devel
We used to successfully configure even if the compiler set in CC wasn't found. This patch now checks that both the host compiler and the build compiler exist. With this I also moved around some code such that cross-compilation works properly with the checks. Fixes: https://gitlab.com/herrhotzenplotz/gcli/issues/244 Signed-off-by: Nico Sonack <nsonack@herrhotzenplotz.de> --- configure | 34 +++++++++++++++------------------- 1 file changed, 15 insertions(+), 19 deletions(-) diff --git a/configure b/configure [message trimmed]
From Nico Sonack to ~herrhotzenplotz/gcli-devel
We otherwise see compilation failures on NetBSD due to va_list not being defined. Signed-off-by: Nico Sonack <nsonack@herrhotzenplotz.de> --- thirdparty/sn/sn.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/thirdparty/sn/sn.h b/thirdparty/sn/sn.h index 37bc2fca..be220505 100644 --- a/thirdparty/sn/sn.h +++ b/thirdparty/sn/sn.h @@ -1,5 +1,5 @@ /* [message trimmed]
From Nico Sonack to ~herrhotzenplotz/gcli-devel
remph <lhr@disroot.org> wrote: > This is the local branch I've been running, and tweaking as I encounter > bugs. Sent in one thread to avoid spamming the mailing list. > Also, thanks for the cool program! Thank you for your contribution! :-) > Misc. small patches: I have split this patch into its three parts for better granularity and bisectability. Other than that all applied and pushed.
From Nico Sonack to ~herrhotzenplotz/gcli-devel
This adds assertions in various places where we wish to check that JSON scoping is correct. These are only in effect if not in release mode (NDEBUG is not defined). Reference: 3147ffa4 (Fix uninitialised jsongen structs, 2024-12-24) Signed-off-by: Nico Sonack <nsonack@herrhotzenplotz.de> --- src/json_gen.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) diff --git a/src/json_gen.c b/src/json_gen.c index 51ee1ae2..051a34b6 100644 [message trimmed]
From Nico Sonack to ~herrhotzenplotz/gcli-devel
This will disable assertions on most libc's. Signed-off-by: Nico Sonack <nsonack@herrhotzenplotz.de> --- Makefile.in | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/Makefile.in b/Makefile.in index 77e17fff..140133bf 100644 --- a/Makefile.in +++ b/Makefile.in @@ -24,13 +24,13 @@ SHELL = @SHELL@ ############## Optimiser / Warning flags ##### [message trimmed]
From Nico Sonack to ~herrhotzenplotz/gcli-devel
> - Uninitialised jsongen in gcli-forks causes an infinite loop in fit(), where > capacity is doubled until a string fits, but capacity is 0 so doubling does > nothing > - Also change what looks to me like unbalanced gcli_jsongen_{begin,end}_* > calls in src/github/issues.c Actually, no that I am thinking about this... maybe I should change the return type of the jsongen* functions to void and simply assert that whatever we pop off the scope stack is of the correct type. Another solution is to simply add assertions in various places (which are removed if compiled in non-debug mode). I'll work on a patch for likely solution #2.
From Nico Sonack to ~herrhotzenplotz/gcli-devel
> - Uninitialised jsongen in gcli-forks causes an infinite loop in fit(), where > capacity is doubled until a string fits, but capacity is 0 so doubling does > nothing > - Also change what looks to me like unbalanced gcli_jsongen_{begin,end}_* > calls in src/github/issues.c Good catch! Thanks for the patch, I have applied it. I'm gonna update the Changelog later as well. Nico
From Nico Sonack to ~herrhotzenplotz/gcli-devel
Content-Type: multipart/mixed; boundary="----_=_55e9418470a4e71e44f4dd10_=_" This is a multipart message in MIME format. ------_=_55e9418470a4e71e44f4dd10_=_ Content-Type: text/plain; charset=UTF-8 Content-Disposition: inline
From Nico Sonack to ~mawww/kakoune
On Tue, Dec 10, 2024 at 04:42:32PM UTC, mawww@kakoune.org wrote: > From a quick look at that test script, I think there is a race condition as > we do not wait for the correct refresh call, we get the one that immediately > follows the draw_status instead of the one triggered by the keystroke. > > Does this diff fix the issue ? I guess this is already obsolete...but no. It does unfortunately not fix the issue. > > ``` > diff --git a/test/tools/git/blame-in-diff/script > b/test/tools/git/blame-in-diff/script > index 0cee0ca7..ba257e6a 100644