IRC channel logs

<rlb>ok

<old>what are we talking about here? Something cool getting merge? :D

<old>ArneBab: Thanks for the merge for fixing meta describe command btw

<rlb>old: pr 22 (win64 bits --- long vs intptr_t, etc.)

<sneek>wb tohoyn!!

<tohoyn>sneek: botsnack

<sneek>:)

<tohoyn>sneek: Merry Christmas

<sneek>wb tohoyn!!

<tohoyn>mwette: function g_dbus_connection_call_sync returns a GVariant. See end of the page https://docs.gtk.org/gio/method.DBusConnection.call_sync.html.

<Codeko>I'd like to automatically call a procedure after i set! a specific variable called frame-resize-pixelwise. Can i somehow achieve this with guile?

<tohoyn>Codeko: Perhaps https://www.gnu.org/software/guile/manual/html_node/Identifier-Macros.html

<Codeko>I've already thought of changing the behaviour of set! before for implementing buffer-local variables but ended up doing it by implementing setq. I guess i should also handle this in setq and state that any use of set! in my program is wrong and should not be done. I can't make a decision.

<Codeko>i can probably do better than just state you shouldn't use it, i could make set! print "don't do that >:(" instead of setting, or undefine it ?

<tohoyn>Codeko: it seems that identifier-syntax is what you need for your first question.

<mwette>tohoyn: got it

<ArneBab>old: thank you for the fix! That fix may look small but the effect on programming experience is pretty big.

<ArneBab>rlb: do I understand it correctly that with https://codeberg.org/guile/guile/commit/758349e64ee9b2718de7125b4ab4afab6ada6f5e we can close https://codeberg.org/guile/guile/pulls/42 ?

<ArneBab>Can someone merge https://codeberg.org/guile/guile/pulls/15 ? I think I shouldn’t merge my own PRs.

<rlb>I'm not positive right now, but I think #42 has already been handled directly.

<ArneBab>rlb: Thank you! I now closed it with a thank you.

<rlb>Regarding srfi-134, for my part, I don't know much about it atm, and don't know if I'll have time to delve soon --- ideally would also like at least a top-level look from Ludovic or Andy...

<ArneBab>No problem -- thank you. Maybe I’ll ping them in the new year :-)

<ArneBab>ACTION needs to start cooking

<old>ArneBab: np. These kind of issues ought to be copied from debbugs to codeberg. these are low hanging fruits that ought to be fixed easily by anyone that has the time

<rlb>ArneBab: was there some significant reason the win64 pr changed the C code in hash.c like "if (sizeof (..) === 8" to an #if, i.e. was that just preference, or some belief that the compiler wouldn't optimize it out?

<rlb>I tend to prefer the C code (otherwise) because it indents normally, and is less noisy...

<rlb>And I'm only asking because I'm having to deal with all the rebase conflicts wrt the utf8 branch commits, and so I have to pick again.

<rlb>i.e. in the new code it's now "#if SCM_SIZEOF_INTPTR_T == 8 ..." vs previous "if (...) ...". Not critical, just wondering.

<rlb>...and also, why is that even checking SCM_SIZEOF_INTPTR_T vs either SIZEOF_UINTPTR_T or preferably(?) sizeof (scm_t_hash), which is the actual type?

IRC channel logs

2025-12-24.log