***ijp` is now known as ijp
***Yawar_ is now known as Yawar
<Onslauth>Hi guys, I have a few questions regarding ice-9 and modified-julian-day <Onslauth>Internally it looks like everything is based off of UTC? <Onslauth>(define utc-jdn (date->modified-julian-day (current_date 0))) <Onslauth>(define tz-jdn (date->modified-julian-day (current_date))) <nalaginrut>Onslauth: IMO, it's local time if you don't pass time zone, or it's the time zone specified by you <Onslauth>nalaginrut - I would have expected that too, but it seems to always be UTC <Onslauth>julian-day always seems to be the UTC value, and never includes the timezone. <Onslauth>hi Wingo, I dont suppose you might know an easy way to solve the problem I am having with dates and julian-day? <Onslauth>I am having issues with julian-day not including the timezone when converting from date to julian-day <Onslauth>nalaginrut - You get the gist before you split? <nalaginrut>it's unrelative to timezone, although you specified it <Onslauth>Yeah. Julian day is great for log file rotation. i.e. rotate at midnight <Onslauth>but without the timezone, it rotates at midnight + timezone <nalaginrut>when you convert it to Julian day, they're in same hour, so the difference between utc-jdn and sast-jdn was eliminated <nalaginrut>let me check it out in another Scheme implementation <Onslauth>but if you say its working off utc then its ok. I'll figure out a way of getting the tz into the value. <nalaginrut>I mean the timezone is eliminated during the conversion <Onslauth>But is there no way to keep the tz when converting to julian day? <Onslauth>I would have expected the julian day conversion to take into account the timezone if one is specified in the date? <nalaginrut>as I said, Julian day is +0, so there's no difference even you specified it in current-date <Onslauth>But then to make it useful you have to calculate your tz separately in julian day and add it? <Onslauth>Is it possible to get the timezone from the (current_date) object? <nalaginrut>Onslauth: well, I've no idea about your usage, it's rarely to use Julian day for me ;-) <Onslauth>Ok, so can I get the timezone from the current_date object? <Onslauth>nalaginrut - julian date is great for working with date-time as a decimal value. <didi>Using GOOPS, I have a <base> class with a #:init-value slot `foo' as (list 1 2). I want to write a <derived> class from <base> with a #:init-value slot `foo' as (list 1 2 3). Is it possible to use some incantation so I only have to write in <derived> (list 3) and it generates a `foo' as (list 1 2 3), i.e., the list defined on <derived> appended with the list defined on <base>? <didi>Well, thinking about it, it will not be necessary. For my use case, I can just call (next-method). ***Fuuzetsu is now known as Guest68448
***add^_` is now known as add^_
<taylanub>didi: hidy ho, did you get my notice from sneek ? <didi>taylanub: I did not. :^( <didi>taylanub: Oh, cool. Thank you. ***DerGuteM1 is now known as DerGuteMoritz
<taylanub>didi: BTW don't hesitate to ask any questions about the library. (The docs are quite comprehensive but I'm not sure about their clarity.) ***dunsmoreb_ is now known as dunsmoreb
***Shozan is now known as SHODAN
<mark_weaver>(We should probably just import foof-loop into Guile) <davexunit>ArneBab: where can I read your post? was this a blog post or something I can find on the mailing list? <davexunit>I wish I could write all of my blog posts in org-mode and still keep the neat features of Pelican <davexunit>I wonder how hard it would be to roll my own static site generator that used emacs batch mode to process org-mode files. <ArneBab>davexunit: that should not be that hard. Have a look at org-publish - that might already get you most of the way <davexunit>ArneBab: reading your post about python->scheme and was pleasantly surprised to see guile-2d referenced! <ArneBab>davexunit: It’s one of the really neat examples for people who tried their hands at game programming ☺ <davexunit>the day I learned about coroutines was a good day. <ArneBab>davexunit: and reading its source showed me how deep even short snippets can go ☺ <ArneBab>mark_weaver: also as you can see from the py2guile text, I finally fixed the fontification regexps for wisp <davexunit>ArneBab: I believe that snippet is from wingo. it was posted on the guile mailing list ~2 years ago I think. it took me a long time to understand it, myself. <ArneBab>(how code flows from one to another…) <davexunit>but it's proved to be one of the most useful code snippets ever. ***amoe_ is now known as amoe
***Guest68448 is now known as Fuuzetsu
<lloda>if I use test-begin & test-end from srfi-64, how do I get the result of the test in a variable? I don't have a explicit 'test runner'. <mark_weaver>and (test-runner-current) to get the current test runner. <mark_weaver>ArneBab: I didn't notice the fontification because I use emacs-w3m :) ***Fuuzetsu is now known as Guest56456
***Guest56456 is now known as Fuuzetsu`
***Fuuzetsu` is now known as Fuuzetsu
<lloda>mark_weaver: not sure I get it. <lloda>(test-passed? (test-runner-current)) <mark_weaver>it might make more sense to make a custom test-runner. it's not hard. see test-suite/tests/srfi-64.test in git for an example. <lloda>it's for a wrapper for FFTW. Usually I do C++ wrappers by hand and my own clunky SConstruct/bash test setup, but I'd like to publish this, so I'm using the ffi and srfi-64. But it's my first time using either so I'm probably carrying over some bad habits. <mark_weaver>I don't know why you'd need to check the results of the tests explicitly. The default SRFI-64 test runner will generate a report at the end. <lloda`>mark_weaver: In my current setup I abort the build on test failures. It's easier for me to transition to a proper test suite library if I don't have to do everything differently all at once. <mark_weaver>okay, then I recommend using a custom test runner like the one in test-suite/tests/srfi-64.test that aborts on failure. <didi>Do guile macros guarantee an expression is not evaled more than once? <mark_weaver>if you need that, then you must use a let in the generated code.