IRC channel logs

2020-06-13.log

back to list of logs

<anadon>Looks like there may be something wrong in the guix packages. Time to bounce there.
***terpri_ is now known as terpri
<tohoyn>sneek, botsnack
<sneek>:)
***apteryx_ is now known as apteryx
<msiism>What is Guile's implementation language?
<RhodiumToad>the parts of guile that aren't written in guile are in C
<msiism>I see, thanks.
<gjanssens>Hi, for gnucash on Windows we are currently pulling in guile 2.2.7. It has a few issues for which we carry local patches.
<gjanssens>One in particular has to do with a missing mktime_internal implementation.
<gjanssens>Searching the guile bug database I ran across this one: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=24681
<gjanssens>That reports the same issue on sparc solaris.
<gjanssens>The bug suggests this is due to a change in gnulib.
<gjanssens>And the fix should go into 2.0.15
<gjanssens>We're more than three years later and the bug is still open.
<gjanssens>Is this oversight or are there technical hurdles that prevent this from being fixed ?
<janneke>gjanssens: you aren't using the 2.2.3-based wip-mingw-guile-2.2 (similar questions apply)?
<wingo>sneek: later tell civodul i don't thing the conclusion about scm_call_n is right fwiw, or that it applies to anything beyond the case you are considering
<sneek>Okay.
<wingo>i.e. unrelated to after-gc-hook
<wingo>sneek: later tell civodul important to avoid jumping to conclusions imo to avoid effort misfocussing
<sneek>Okay.
<gjanssens>janneke: no we're not using wip-mingw-guile-2.2 I wasn't even aware that was a thing...
<gjanssens>that branch hasn't been touched since 2018. How much of it has been merged in the main development branch ?
*janneke starts to doubt...
<janneke>i remember that stable-2.2 didn't build for me and i started looking for patches, found wip-mingw-guile-2.2
<janneke>...interesting
<gjanssens>For gnucash we now have a working guile 2.2.7. But our patches are absolutely minimal and aimed at getting guile to work for us.
<gjanssens>It's a lot of duplicated effort though to keep on maintaining this out of tree.
<gjanssens>That's why I started looking around for other work in this area. And the first thing I ran into was the bug I linked to before.
<civodul>hey wingo
<civodul>wingo: i probably kind of jumped to the conclusions
<civodul>that's how i interpreted the perf report
<civodul>but i'm happy to hear your opinion
***terpri_ is now known as terpri