***didi`` is now known as didi
***Fuuzetsu` is now known as Fuuzetsu
<pallagun>beginner question: I made a smob (just a pointer to double) that I allocated with malloc instead of the scm_gc_malloc. I have a bunch of existing code that uses malloc, so I wanted to test/figure it out. The example in the guile doc uses scm_gc_malloc. Is there an example somewhere that uses normal malloc? <pallagun>also, if I made a smob and set up a scm_set_smob_free function, should that get called when I run ",gc" ? *nalaginrut just fixed heartbleed for his server <didi>It doesn't look like a good idea. <b4283>nalaginrut: my old debian is vulnerable :( <zacts>b4283: I thought that debian squeeze was not vulnerable to heartbleed <b4283>dist-upgrade shows libssl is required for update <zacts>I wonder how difficult it would be to implement the SSL/TLS protocol in guile scheme? <nalaginrut>b4283: if yours is lower than 1.0.x, and it's fine <zacts>I guess gnutls + guile works <zacts>but it's just a curiosity, I do feel that more code should be implemented in languages other than C.. <zacts>C is starting to show its cruft with bugs like these, it seems.. but I'm still a newbie <zacts>hm.. I don't mean for a library for servers and browsers, but for guile scheme applications <nalaginrut>I think the better solution is to use guile-gnutls <zacts>but I guess we should probably stick with gnutls as it is more tested <nalaginrut>if you need a secure guile server, you'd better based on guile-gnutls *nalaginrut Artanis will do this someday... <zacts>I want to use (scheme) instead of perl. <nalaginrut>Artanis' main site is generated by hyde of Chicken <nalaginrut>the only problem is lacking of tutorial, only a manual *nalaginrut is considering to post a tutorial on his blog <nalaginrut>zacts: BTW, you need Chicken-Scheme rather than Guile on this case <didi>b4283: Change your keys. <didi>Revoke the certificates. <b4283>i guess i'll just have to change it <b4283>nalaginrut: you need to put it in the new certificate's revoke list <nalaginrut>oh, my new server hasn't published the certificate <b4283>nalaginrut: how do you publish a cert ? <nalaginrut>I mean this is my own server, so I didn't share the pub key to others <nalaginrut>I wonder if it's the same thing we're talking about *wingo checks out gdb from git so he can write a guile extension for it, whee <nalaginrut>My actual question is, can I replace pmatch with match? <nalaginrut>taylanub: I found it in (system base pmatch), and used in (language ecmascript compile-tree-il) <nalaginrut>well, I guess pmatch is a simpler implementation of 'match' *wingo can't remember his nick *wingo poking at gdb extensions <wingo>very :) a couple of rough edges tho, like exposing the name "gsmob" <wingo>was this already discussed, do you know? <civodul>i didn't really like insisting on SMOBs in the doc <civodul>i thought most of it was gone, maybe no <wingo>i also wonder why `object-property' is insufficient <wingo>yeah, i guess it's an obvious name for what it is :P <wingo>civodul: were you able to get gdb's test suite to run? fails for me for some reason <wingo>civodul: speaking of which, hydra needs a kick to flush queued jobs or something ***Fuuzetsu is now known as Guest21808
***Guest21808 is now known as Fuuzetsu`
***Fuuzetsu` is now known as Fuuzetsu
<davexunit>taylanub: guile-wm is very neat. I go through phases of using it instead of gnome 3. *wingo forgets how to send a patchset over email <wingo>seems i was missing git-send-email <Fulax>for file in *.patch; do mail <address>; done ? :) <taylanub>I use ratpoison since a long time, and actually only ever use full-screen windows, the only actual functionality I make use of being a couple Xorg-global keybindings, and the little echo area it can show on the top right. Perhaps those things shouldn't be done by the WM anyway ... <Fulax>I missed mail <address> <$file ***dsmith-w` is now known as dsmith-work
<dsmith-work>wingo: Got another master test case not happy. cse.test is looking for the (language tree-il cse) module. <civodul>wingo: i was able to run the Guile part of GDB's test suite <civodul>but when i switched to libgc 7.4, weird things started happening <taylanub>Hrm, with GNU MP 6.0.0, or something else related to my 2.0.11 build on OS X, #i1@-0 evals to 1.0+0.0i instead of 1.0-0.0i. <taylanub>Makes numbers.test fail at the string->number test. <taylanub>Wrote it in my e-mail about building 2.0.11 on OS X 10.9 / Xcode 5.1, I guess I should make a separate topic for it, don't think it's OS X-related. <taylanub>Hehehe, GMP 6.0.0 release notes: "This release will not work reliably on FreeBSD 10.0 for i386 or amd64 using the 32-bit ABI. The reason is bugs in the compiler 'clang'." Not related to my issue since I built with GCC, but it's interesting to see more clang-related issues ... <didi>civodul: Congratulations on the release.