IRC channel logs

2026-01-27.log

back to list of logs

<mwette>Sorry about all that. I have other challenges, like gobject use of non-C in macro args: GOBJECT_DEPRECATED_FOR('G_TYPE_VARIANT')
<janneke>mwette: sorry, i really meant that friendly-funny; you're doing grand work!
<janneke>also, providing multi-char readback and (slightly) patching tcc ave both pretty viable
<jelleke>Howdy folks, last year I read on the Ada forum about some Ada bootstrapping efforts. Very exciting stuff! Last week I released the source for the Ada- compiler. This might be of help. Read more about this and the community efforts on reviving this large Ada subset from the 80s: https://forum.ada-lang.io/t/source-for-the-ada-compiler/4137
<jelleke>Anyway, just dropping in to let you folks know. I really appreciate the efforts of this community. Bootstrappability is extremely important imo <3
<jelleke>Last thing I heard there were some plans for using Ada/Ed for bootstrapping, but I think that Ada- might be easier to handle, since the c port of Ada/Ed really shows its roots in SETL and is a bit hard to comprehend (at least to me)
<jelleke>Hah, I see that both Irvise and efraim are here, so I bet you're already up to date if this relevant :-D
<jelleke>In that case, please don't mind the rude drive by ;)
<janneke>jelleke: ooh, that's great; thanks and congrats!
<jelleke>You're welcome! It's incredibly difficult to try to get the source for all these 80s Ada compilers and I was incredibly happy that Jan van Katwijk was so kind
<janneke>i can imagine, and that's lovely to hear
<janneke>possibly it helps that we have a, imho, pretty noble cause and lofty goal :)
<jelleke>yes, it sure does
<jelleke>is there a status update on the Ada bootstrapping effort somewhere?
<jelleke>I wonder how that's coming along
<jelleke>my personal motivations are along the lines of permacomputing. When things are bootstrappable, they are also way easier to port to old hardware or uncommon operating systems
<lanodan>Permacomputing bit reminds me that bootstrapping ends up pointing out bits that are hard to rebuild or can't be. And well: "That which cannot be repaired is already broken."
<jelleke>yeah, exactly
<jelleke>without proper bootstrapping, permacomputing becomes impossible
<jelleke>so I imagine that folks from #permacomputing are very happy with your efforts :-D
<mwette>janneke: Do you want c99 to avoid using include-from-path of pre-generated mach.d/ code? That can be done by using (use-modules (nyacc lang c99 mach)) instead. (That is, instead of including generated scm code for tables, just get it from the mach module directly.) Same for cppmach.
<mwette>^ for mes, via cond-expand
<stikonas>^^ CC: fossy