<aw->
but I can parse the file just fine with my JSON decoder
<aw->
it's only ht:Read that fails
<aw->
it seems PicoLisp handles it correctly
<aw->
or.. does ht:Read actually 'parse' or tokenize the file?
<Regenaxer>
PicoLisp gets confused if a byte looks like the first char of an UTF-8 sequence but ist
<Regenaxer>
result may be unpredictable
<Regenaxer>
if the byte looks like a 3~byte char, it counts 3 but only 1 byte folluws perhaps
<Regenaxer>
you need a conversion
<aw->
hmmm
<aw->
i guess curl is confused as well
<Regenaxer>
(in '(iconv "-f" "ISO-8859-15" "file.txt") (ht:Read ..
<aw->
oh, the file is UTF-8 actually
<aw->
file test2.json
<aw->
test2.json: UTF-8 Unicode text
<Regenaxer>
hmm, perhaps it contains a mixture of charsets?
<Regenaxer>
(in (list "/usr/bin/iconv" "-f" "ISO-8859-15" "-t" "UTF-8" "test2.json") (ht:...
<aw->
maybe
<aw->
oh i converted and the filesize changed
<Regenaxer>
The question is what to use for -f
<Regenaxer>
yeah
<aw->
wow that worked
<Regenaxer>
With iconv?
<aw->
haha yeah
<Regenaxer>
Perhaps just by chance
alexshendi has quit [Read error: Connection reset by peer]
<Regenaxer>
if the source charset is not clear, the result may be wrong
<aw->
right
<aw->
ok next question, how can I make ht:Read not fail in that situation?
<Regenaxer>
I think there is no general way
<aw->
yeah.. if you don't know the charset, it's hard to do anything haha
<Regenaxer>
If you convert first, then 'info', ht:Read will not fail
<Regenaxer>
just some chars may be wrong :)
<aw->
right, but if the doc is already utf-8 then i shouldn't convert
<Regenaxer>
depends if that is really a problem
<aw->
and in this case, the doc said it was utf-8 even though it wasn't 100%
<Regenaxer>
true
<aw->
so I can't trust that
<aw->
argh, i guess this is the eternal problem with computers
<aw->
shoganai
<Regenaxer>
passing always through iconv first is too expensive?
<Regenaxer>
:)
<Regenaxer>
daijobu
<aw->
yes, not worth the effort
<aw->
thanks for the help!
<Regenaxer>
np
_whitelogger has joined #picolisp
orivej has quit [Ping timeout: 260 seconds]
arievw has joined #picolisp
<arievw>
Hi!
<Regenaxer>
Hi arievw!
<arievw>
Downloaded Wiki sources from here: https://picolisp.com/wiki/?documentation as wiki.tgz but using this unchanged gives different results from the official version. Is this the current version??
<Regenaxer>
I thought the link points to software-lab.de (?)
<Regenaxer>
and 8080 is not needed. Default port is 5000
<arievw>
I use 5000 indeed.
<Regenaxer>
yes, the command line is good then
<arievw>
OK. Iĺl just redo neatly and retry first ...
<Regenaxer>
I use a local ./pil as the global one from Debian is always very old
<Regenaxer>
Perhaps just remove DB
<Regenaxer>
$ rm -r db/wiki/
<Regenaxer>
Should initialize automatically then
<arievw>
pil -version shows 18.2.17
<Regenaxer>
ah, good
<Regenaxer>
should be fine
<arievw>
I´ll remove the db. Pls wait ...
<Regenaxer>
sure
<arievw>
Should the db folder be in the wiki folder itself?
<Regenaxer>
it is created in the current directory
<Regenaxer>
I just tried here on my tablet
<Regenaxer>
did not have the wiki installed
<Regenaxer>
Works fine
<Regenaxer>
Login is admin/admin (see wiki/init.l)
<arievw>
Here problem stays the same. So, first a retry install then.
<Regenaxer>
What problem?
<Regenaxer>
There cannot be a bad install I think
<Regenaxer>
just tar xvfz wiki.tgz
<Regenaxer>
BTW, Arie, some details are indeed much better discussed here. I observed that if too many emails about minor details without general interest flood the mailing list, some people unsubscribe due to the noise
<arievw>
Well, Iĺl install firefox on Windows and try :)
<Regenaxer>
Not bad or wrong, it is rather how the menu items are arranged, right?
<Regenaxer>
Perhaps a matter of font sizes and available space
<arievw>
Well, in the Ubuntu version I miss e.g. the documentation link ...
<Regenaxer>
hmm, yes, this is strange - the data should be the same
<Regenaxer>
No, wait!
<Regenaxer>
You look at different DBs
<Regenaxer>
Impressum etc. is only in the "real" DB
<Regenaxer>
yes
<Regenaxer>
it is picolisp.com!
<arievw>
Also the page contents are different. In Opera it start with ¨programming simplified¨ but on Ubuntu the first page starts with ¨PicoLisp: The Scalpel of Programming¨
<Regenaxer>
is this a replacenment for wiki/lib.css?
<Regenaxer>
or wiki/wiki.css?
<Regenaxer>
seems wiki.css
<Regenaxer>
Looks good, doesn't it? Is it ready to be installed on picolisp.com?
<arievw>
Yes, wiki.css. I tried tot test it (also with different resolutions). Because of the mark-up currently used things may look a bit off for now. In the next days I´ll first make that in order.
<arievw>
So I think it can be installed safely.
<Regenaxer>
cool!
<Regenaxer>
Did it, can you try?
<arievw>
Yep.
<Regenaxer>
if ok, I also release a new TGZ
<arievw>
I think I have to make a few adjustments first. Pls wait :)
<Regenaxer>
ok, no hurry
<arievw>
OK. I adjusted the sizes a bit. It was a bit too large. Could you please use this wiki.css: https://pastebin.com/fA0J3X6c
<Regenaxer>
Why did you remov 23jun18avw ?
<Regenaxer>
Shall I put it back in?
<Regenaxer>
Ah, there is also "edg" - who is that?
<arievw>
He! I´ll Have a look
<Regenaxer>
:)
<arievw>
Baffled! Code seems OK afaict. So, yes, please re-add that line :)
<Regenaxer>
So we have two editors today? edg and avw?
<arievw>
Also I now use the same browser in both Windows and Linux to make comparison more honest.
<arievw>
Ah. That was it! No edg was earlier. I´ll have a look :)
<Regenaxer>
I see, it was 26apr18edg
<Regenaxer>
I fix it
<Regenaxer>
edg is Erik Gustafson
<arievw>
Should be: line 1 ¨* 23jun18avw¨ and line 2 ¨26apr18edg¨
<Regenaxer>
Installed it
<Regenaxer>
yes
<arievw>
OK. Thx.
<Regenaxer>
fixed already
<Regenaxer>
I release a TGZ
<Regenaxer>
Released, you could download if you like to cross-check
<Regenaxer>
oops, wait
<Regenaxer>
ok, done
<arievw>
OK. Thx. Next days i´ll rework stuff a bit and people can have a look if everything is fine.
<Regenaxer>
Thank you!
<arievw>
The stuff Karl-Heinz asked about is a PicoLisp code generation thing. Maybe he can get involved in that part :)
<Regenaxer>
Sure!
jibanes has quit [Quit: Changing server]
jibanes has joined #picolisp
yunfan[Cr] has quit [Ping timeout: 268 seconds]
jibanes has quit [Quit: leaving]
jibanes has joined #picolisp
jibanes has quit [Client Quit]
jibanes has joined #picolisp
orivej has joined #picolisp
arievw has quit [Quit: Leaving]
andyjpb has joined #picolisp
pierpal has quit [Quit: Poof]
pierpal has joined #picolisp
andyjpb has quit [Ping timeout: 276 seconds]
alexshendi has joined #picolisp
pierpal has quit [Quit: Poof]
pierpal has joined #picolisp
pierpal has quit [Remote host closed the connection]
pierpal has joined #picolisp
styx has joined #picolisp
pierpa has joined #picolisp
reed has joined #picolisp
<reed>
Hi All. I'm trying to understand "use". If I do
<reed>
(push 'Result 5) then (de f () (use (Result) (push 'Result 10) (show 'Result)))
<reed>
Then execute (f)
<reed>
I get "Result (10 5)"
<reed>
How can I prevent the code in 'f' from being "contaminated" by the fact that Result was used outside of 'f'