IRC channel logs

2017-09-04.log

back to list of logs

<ArneBab_>amz3: I can’t say much about the content, but the design looks like a neat book
<mwette>amz3: I sent new info on libgit2.scm. I am working to chase down the pointer issue.
<sneek>mwette, you have 1 message.
<sneek>mwette, janneke says: congrats on the release! i'll be interested to see what asm does...mescc has been using its own construct to implement asm which looked like a function call :-)
<mwette>janneke: thanks, I don't remember what I return. I may be just ignoring, but I can add it in.
<mwette>janneke: I just checked. The parser returns `(asm-expr ...)'
<mwette>amz3: if you are here, I am trying debug the pointer issue.
<mwette>It is in unwrap~pointer. I am trying to support both Guile foriegn pointers and bytestructure pointers. Converting these back and forth must be broken.
<mwette>and wrap/unwrap should be only needed withing libgit2.scm. They would only be needed by other ffi-modules using types defined in companion ffi-modules
<mwette>amz3: got it working. In ffi-help-rt.scm I changed ...
<mwette>
<lloda>for my --listen port encoding question the other day, I have a solution
<lloda>call scm_setlocale(scm_from_int(LC_ALL), scm_from_locale_string("")); before entering the Guile shell
<lloda>the problem was only when embedding, $GUILE --listen was always ok
<lloda>kind of surprised it's needed otherwise, oh well
<cmaloney>morning
<dustyweb>moin moin
<dustyweb>meow meow
<happy_gnu[m]>hello \\o/
<happy_gnu[m]>I've been reading '(the land of lisp)
<happy_gnu[m]>it is really different from scheme :/
<happy_gnu[m]>I feel like I am reading a python book
<zacts>hi guilers
<zacts>happy_gnu[m]: have you tried the Realm of Racket book?
<zacts>racket is a scheme variant
<zacts>it's inspired by Land of Lisp
<happy_gnu[m]>zacts: no I haven't yet :) but I have it already
<happy_gnu[m]>I will do that book after this one
<zacts>cool
<zacts>I own both books as well
<zacts>I need to get through them too
<zacts>I've just been caught up with learning Ruby lately
<dustyweb>happy_gnu[m]: Land of Lisp is fun
<zacts>I much prefer scheme/lisp over Ruby
<zacts>but much existing codebases and frameworks are using Ruby
<zacts>so I may need it for a job
<dustyweb>happy_gnu[m]: probably a lot of the difference you're seeing is that it uses an imperative style until towards the end of the book
<dustyweb>at which point it introduces functional programming
<zacts>I do like Ruby syntax though
<zacts>but I like lispy syntax actually better due to its practical features, and it's very elegant. direct code is already an AST
<happy_gnu[m]>dustyweb: ohh I see
<happy_gnu[m]>Yeah I like the games it is fun
<dustyweb>I really liked the mini-genetic-programming game
<happy_gnu[m]>maybe the author thought more on imperative programmers, as there are more and schemers probably didn't need any help lol
<dustyweb>well it isn't really genetic programming
<happy_gnu[m]>I am at half more or less
<dustyweb>as much as it is a slight amount of evolution
<happy_gnu[m]>on the orc battle
<happy_gnu[m]>one friend told me to do a chess board
<dustyweb>happy_gnu[m]: it was useful for me, because I was an imperative programmer with a background in python, and the functional programming part of the book finally brought some clarity to me by what was meant by that and why it was valuable
<happy_gnu[m]>we are thinking on how to do it with guile
<happy_gnu[m]>we are not plannig an AI for now, just a board and check if movements are correct
<happy_gnu[m]>but it is really harder than it looks :/
<happy_gnu[m]>dustyweb: I see, I don't know much python or javascript, I barely know about objects, so I found scheme way simpler
<happy_gnu[m]>But I want to do fun complicated stuff :) problem is, is hard :/
<dustyweb>happy_gnu[m]: figuring out how to turn such concepts into running code does take time, and practice
<jonh>and coffee or *insert fav drink here*
<happy_gnu[m]>jonh: lol :)
<amz3`>that is good project: https://github.com/aosabook/500lines/
<ijp>amz3`: what a coincidence, I was going to bring it up myself
<pmikkelsen>happy_gnu[m]: speaking of chess I made this for an exam at my school. it wasn't a programming exam, but an exam in electronic circuts, so the code is a bit of a mess. we made a moving chess board, but never finished it ;) https://gitlab.com/pmikkelsen/guile-chess
<pmikkelsen>is was my first "big" project in guile and boy, i love the language
<amz3`>why is that?
<amz3`>do you feel englightenened already?
<pmikkelsen>i have been a haskell user for quite some time, and then i got into lisp and then scheme, and it just seems so "simple" if you know what i mean
<pmikkelsen>the one thing i really want to learn though is macros
<amz3`>I've doing guile for something like 4 use, I only know define-macro and define-syntax-rule
<amz3`>I already had a look at the mighty define-syntax but no much
<pmikkelsen>to me they are magic ;)
<mwette>pmikkelsen: search for paper by Dybvig called "Writing Hygienic Macros in Scheme with Syntax-Case"
<mwette>He also has explanaions and examples in his book "The Scheme Programming Language"
<pmikkelsen>thanks mwette, I will give it a read tomorrow
<amz3`>tx mwette for the quick reply, I did not try it yet, but I will do it tomorrow hopefully
<amz3`>I hope to have at the current test suite passing before december
<amz3`>mwette: btw, the thow other bindings I am interested in are cairo and wiredtiger
<amz3`>I have tests for wiredtige
<amz3`>but it can wait, I libgit2
<amz3`>I also have a program running using termbox so I can test it with that
<amz3`>which are very simple
<amz3`>but you prolly havea lot on your plate :)
<mwette>cairo is in the 0.81.0 nyacc release, in examples/nyacc/lang/c99/ffi.d. I am working on glib, which is providing a few new challenges (e.g., bit-fields)
<amz3`>oh
<amz3`>glib? really?
<amz3`>I will surely include your ffi helper in my book as soon as a get something like libgit2 working
<amz3`>or prolly start with a simple example like cairo
<mwette>I wanted rsvg which relies on glib, gobject, ... so I want to add #:use-ffi-module (glib)
<amz3`>oh ok