|
This channel is intended for people just starting with the Raku Programming Language (raku.org). Logs are available at irclogs.raku.org/raku-beginner/live.html Set by lizmat on 8 June 2022. |
|||
|
00:58
deoac joined
01:52
deoac left
07:47
Manifest0 joined
08:07
dakkar joined
11:35
teatwo left
11:36
teatwo joined
11:43
tea3po joined
11:47
teatwo left
13:03
wafflus joined
|
|||
| rcmlz | m: for 0..6 -> \i { say "sin(i × π) = " ~ sin(i × π) } | 13:56 | |
| Raku eval | sin(i × π) = 0 sin(i × π) = 1.2246467991473532e-16 sin(i × π) = -2.4492935982947064e-16 sin(i × π) = 3.6739403974420594e-16 sin(i × π) = -4.898587196589413e-16 sin(i × π) = 6.123233995736766e-16 sin(i × π) = -7.347880794884119e-16 | ||
| rcmlz | I was wondering why that is not always 0. | 13:57 | |
| nemokosch | Because pi is not accurately represented (and it cannot really be, probably) | 13:58 | |
| so the mistake probably adds up | |||
| rcmlz | Yes, but I remember a discussion about log10(1000) not beeing 3 | 13:59 | |
| nemokosch | the difference is, 10 and 1000 can be accurately represented | 14:00 | |
| rcmlz | and technically sin(2pi) == 0 | ||
| nemokosch | I mean, it could probably be hardcoded... to some extent I guess? But don't forget that Raku wouldn't become a symbolic computation system either way, so you could just multiply the value by some large number and the hardcoding would make the sin function utterly not continuous | 14:01 | |
| rcmlz | ok, understood. Thanks. | ||
| nemokosch | basically you have to make this choice. Either special values are somehow detected (which is not even easy if they are just floating point numbers) and then you get accurate values for them, or the function can be continuous | 14:02 | |
| but I think "doing it right" would only really be possible if pi wasn't just a floating point number | 14:03 | ||
|
14:03
wafflus left
|
|||
| I wonder if somebody had this idea to create some symbols and symbolic expressions in Raku, with versions of pi and e that are not bare numbers | 14:06 | ||
| @librasteve maybe you know more about this | 14:07 | ||
| antononcube | @rcmlz One would need "full blown" (or maybe just "partially blown") mathematical symbolic system for that. | 15:19 | |
| rcmlz | ok. | 15:36 | |
|
17:06
dakkar left
|
|||
| drudgesentinel | What would be a correct way to strip whitespace from a file using a oneliner? I was trying raku -ne '.print.subst(/\s+/, "", :g);' file.example # no changes in printed output Which I believe to substitute any whitespace characters including newlines/tabs with nothing. I also tried s/// but I'm not sure of whether that works with $_, as in the examples everything is reassigned and uses the smart | 18:01 | |
| match operator | |||
| I ended up just using tr since my use case was so simple but was curious as to what I was doing wrong | 18:03 | ||
| nemokosch | All whitespace? | 18:23 | |
| scullucs | Swap the 'print' and 'subst': raku -ne '.subst(/\s+/, "", :g).print;' file.example | 18:34 | |
| drudgesentinel | Is it expected to not be able to use shell scripts (e.g., qx/VARNAME=VALUE ) to define environment variables? my $pod = "foobar" my $env-output = qx/export TFEPOD=$pod/; Probably another wrong tool for the job situation | 19:34 | |
| nevermind, I didn't source things so this is a shell behavior | 19:40 | ||
| nemokosch | You can define variables that processes launched from your Raku script can see | 20:26 | |
| But (thankfully) you cannot leak variables without the caller agreeing to that; that would be hell | 20:27 | ||
|
22:49
Manifest0 left
23:35
teatwo joined
23:36
teatwo left
23:37
teatwo joined
23:38
tea3po left
|
|||