<nckx>sneek: later tell dsmith-work ...nothing, I just want a ping when they wake up :) <nckx>sneek: infinite botsnack ***dostoevsky4 is now known as dostoevsky
***pkill9_ is now known as pkill9
***pkill9 is now known as tedt
***tedt is now known as pkill9
<sneek>Welcome back dsmith-work, you have 1 message! <sneek>dsmith-work, nckx says: ...nothing, I just want a ping when they wake up :) <nckx> #whereiseveryone - a Guix-adjacent channel, I guess you'd call it, but not affiliated. <nckx>Good people but I'd certainly understand if you don't want to take sneek that global. <nckx>Fame is a cruel mistress, something something drugs. <janneke>dongcarl: found it. somewhere between binutils-2.35.2 and binutils-2.36.1, using -flto creates broken dlls <dongcarl>janneke: This is some next-level bisecting you're doing :0 <attila_lendvai>is there a way to load a module-relative file programmatically? currently i'm using (current-filename), but it returns #f in some circumstances (i guess when the file is not compiled, only loaded) <civodul>janneke: interesting use of Guix + Wine :-) <civodul>attila_lendvai: hi! current-filename returns #f only if the thing is not in an actual file, like at the REPL or Geiser <civodul>that said, you can get the file name of a module by calling module-filename, if it helps <attila_lendvai>civodul, module-filename seems to work, thanks! i'll run it through a pull and reconfigure now. <rekado>I’m trying to capture stderr output; the manual says that the process started with open-pipe* inherits current-error-port when it is a file port. <rekado>I used (tmpfile) to open a temporary file port, parameterize current-error-port and then (call-with-new-thread (lambda () (open-pipe* …) …)), reading from the pipe to drain it. <rekado>I then read from the port that tmpfile returned <civodul>rekado: parameterizing current-error-port should do the trick; it must be a file output port of course <civodul>so then you need to open that file and read it <rekado>ah, I see what my mistake was: I stopped reading as soon as I hit EOF, but the output file starts out empty so that’s the first thing I’ll get when reading from it. <rekado>I guess I’ll just discard EOF and wait until the other thread has died. Or maybe I could use fibers and communicate the data over a channel.