IRC channel logs

<rlb>lechner: I don't recall if we have a specialized primitive there, if not, perhaps just a bytevector-length bytevector-u8-ref loop of some flavor.

<rlb>let loop, or degenerate do, or...

<rlb>(...suppose you could also use list functions if performance doesn't matter, e.g. srfi-1, after converting to a list)

<old>lechner: I would read chunck of 64-bit in the bytevector and apply masks to the integer to see if there's a 0 bytes

<old>ofc, for the rest that does not fit 64-bit, you iterate individually

<lechner>old / rlb / thanks for your suggestions! being in a proof-of-concept stage, I went with the online solution to a college problem set, but i'll re-implement. As a side note, I am also likely to look at pread64 over PTRACE_PEEKUSR following this suggestion https://lkml.iu.edu/0302.3/0112.html

<lechner>mwette / Hi, I use Nyacc to import kernel constants. Does the order of files in the #:include field in the FFI declaration matter?

<sneek>wb dsmith :D

<dsmith>sneek, botsnack

<sneek>:)

<mwette>lechner: includes are processed in order presented; if not, it could certainly make a difference

<mwette>and if the order is not working; that's a bug

<identity>i have noticed that (let () (use-modules (ice-9 receive)) (receive (unbound) car unbound)) returns an error of unbound variable of the expected name, and it does not happen if i replace "let ()" with "begin". from the expansion it looks like the let expands in a way that puts use-modules inside an additional begin form, which is probably the reason. am i holding it the wrong way or is this a bug?

<mwette>I wonder if use-modules needs to be top-level (begin counts as top-level). You could try (let ((receive (@ (ice-9 receive) receive))) (receive ...))

<mwette>let introduces a new scope; begin does not

IRC channel logs

2025-05-06.log