<ijp>ArneBab: it's an object that represents the guile vm ***micro` is now known as Guest13987
***sajith is now known as Guest60752
***ft_ is now known as ft
***jvalleroy_ is now known as jvalleroy
<mekeor>i'm running a web-server inside the guile-interpreter on a raspberry pi. why can't i access the page through a browser on another device inside my home-network? – code: <mekeor>(use-modules (web server)) (run-server (lambda (request request-body) (values '((content-type . (text/plain))) "Hello World!")) <mekeor>`netstat -tulpen` says "tcp 0 0 127.0.0.1:8080 0.0.0.0:* LISTEN 1001 14144 6045/guile"... <mekeor>in the browser, i was trying to access <ip-of-my-raspberry-pi>:8080 <mekeor>when i run that web-server on my desktop computer, it's accessable through localhost:8080 <mekeor>mekeor: for sake of completeness: the solution was: (run-server handler 'http `(#:addr ,INADDR_ANY)) <ArneBab_>ijp: so I can run the code if I start by importing the runtime.js? <ArneBab_>(I know that they are not the same, but still the small size suggests good startup times) ***handsome1pirate is now known as handsome_pirate
<sneek>catonano, you have 2 messages. <catonano>sneek later tell paroneayea Thanks ! The one in Vimeo is the one I remembered. I saw it on youtbe but the deffinition was quite poor. On Vimeo it's way better ! <catonano>sneek later tell paroneayea it's so better that it looks like a totally diffferent footage ! <catonano>sneek later tell paroneayea I took a glimpse at his generics functionality that he uses in Structure and Interpretation of Classical Mechanics. It's really interesting and it'd be fun to work on that in Guile <catonano>sneek later tell paroneayea I am preparing a war plan to work on several diferent guile based proects in parallel, starting next autumn ***eena is now known as meena
***dsmith-work is now known as dsmith
***dsmith is now known as dsmith-work
<sneek>Welcome back paroneay`, you have 4 messages. <sneek>paroneay`, catonano says: Thanks ! The one in Vimeo is the one I remembered. I saw it on youtbe but the deffinition was quite poor. On Vimeo it's way better ! <sneek>paroneay`, catonano says: it's so better that it looks like a totally diffferent footage ! <sneek>paroneay`, catonano says: I took a glimpse at his generics functionality that he uses in Structure and Interpretation of Classical Mechanics. It's really interesting and it'd be fun to work on that in Guile <sneek>paroneay`, catonano says: I am preparing a war plan to work on several diferent guile based proects in parallel, starting next autumn <paroneayea>catonano: Guile has generics through GOOPS, but it's also easy to define your own generics system; for Various Reasons (actually because GOOPS doesn't support composite types) I couldn't use GOOPS for my ActivityStreams implementaiton <paroneayea>but I wrote my own generics system and it was less than 100 lines of code, pretty easy <catonano>paroneayea: I don't know whhat composite types are, but it's ok, I'll find tat out <catonano>I read that Racket has a way to reconcile car, cdr with stream-car and stream-cdr. Does anyone here know more ? What is this reconciliation about and jow is it achieved ? <paroneayea>catonano: so in json-ld and other linked data systems you can have a type, like <ijp>ArneBab_: yes, but there is a catch. This is just the vm (well, the parts I had implemented previously). In particular it does not contain boot-9 which is going to be significantly bigger <ArneBab_>ijp: is it possible to include only the parts which are needed? <ijp>maybe? module-ref could make it tricky <ijp>I think if I can get rid of just the syntax-case stuff that we keep around, that'll cut down the size quite a bit <ijp>wingo: is there a list of primitives that don't return values, e.g. box-set!, wind, unwind, handle-interrupts, push-dynamic-state, etc <ijp>Also, does anyone know when the #<syntax> representation was changed? <ijp>I want to read the discussion of the change from the special vector lteral <ijp>is there anything we can do to break boot-9 into something more manageable <ijp>looks like the syntax type is just a simple struct (make-syntax expression wrap module) <ijp>but I can't pattern match on it, can I? <civodul>ijp: the new "syntax" struct was introduced in 2.2.2 <civodul>essentially to have unforgeable syntax objects <civodul>as opposed to the vectors that psyntax used before that <ijp>that was roughly my assumption, but I'm surprised the change is so recent <civodul>yeah this is to support the sandboxing feature used by 'guix potluck' <ijp>civodul: do you know how the change to a struct effects compilation? <civodul>you mean how structs are serialized? <ijp>serialised and deserialised, yes <civodul>ACTION checks 64c5cc58fced3092f17639bbbddb46c1bae974c8 <civodul>looks like it's just an extra case next to pairs, strings, vectors, etc. <ijp>okay, that's good to know <ijp>(or don't, it's like 3MB, and this is "minimised") <ijp>I'm surprised uglifyjs doesn't barf on indenting it <ijp>I think I might look into when I can elide the syntax objects from the output, since they are hilarious to look at ***profan_ is now known as profan