Fix unicode.split with seperators examples

[zetashift]

Fixes: #17174
https://nim-lang.org/docs/unicode.html#split.i%2Cstring%2CRune%2Cint didn't work with the set constructor {} so replaced it with a string that's converted to an openArray compatible type containing Runes.

If there is a better/more idiomatic way to get Runes from a set/string, then please do tell, first time interacting with Unicode module.

[zetashift]

@timotheecour using sugar.cycle is hard(or I am using it wrong) because toRunes returns a seq[Rune] and the collect macro packs that into a seq[seq[Rune]] should I just use a string as a separators list?

import sugar

let splittedBySequence = newSeq[string]()
let separators = collect(newSeq):
  for separator in @[";", ":", "$"]:
    separator.toRunes

for word in split("this:is;an$example", separators):
  splittedBySequence.add(word)

doAssert splittedBySequence = @["this", "is", "an", "example"]

[zetashift]

I currently have this, is this a decent example?

    # And the following code splits a string using a sequence of Runes.
    import sugar

    var splittedBySequence = newSeq[string]()
    let separators = collect(newSeq):
      for separator in ";:$".runes:
        separator

    for word in split("this:is;an$example", separators):
      splittedBySequence.add(word)

    doAssert splittedBySequence == @["this", "is", "an", "example"]

[ringabout]

Looks good!

nit:

import std/sugar


var splittedBySequence = newSeq[string]()
let separators = collect:
  for separator in ";:$".runes:
    separator

for word in split("this:is;an$example", separators):
  splittedBySequence.add(word)

# use assert 
assert splittedBySequence == @["this", "is", "an", "example"]

Or even one line

let separators = collect(for separator in ";:$".runes: separator)

[zetashift]

I didn't know you could use collect like that, the examples didn't mention it! :D, awesome thanks.

[zetashift]

Last question is it better to group the imports?

    import std/sequtils

    assert toSeq("hÃllo\lthis\lis an\texample\l是".split) ==
      @["hÃllo", "this", "is", "an", "example", "是"]

    # And the following code splits the same string using a sequence of Runes.
    import std/sugar

    var splittedBySequence = newSeq[string]()
    let separators = collect(for separator in ";:$".runes: separator)

Turn the imports into a top level:
import std/[sequtils, sugar]
Not sure which makes it better :(

[timotheecour]

Last question is it better to group the imports?

yes but turns out you don't even need collect here, in all your examples, toSeq is enough.
(Note: nim-lang/RFCs#76 would allow the more explicit: from std/[sugar, sequtils] import toSeq, collect but again, that's not needed here)

[timotheecour]

LGTM, thanks!
further cleanup PR's welcome (see also https://github.com/nim-lang/Nim/pulls?q=is%3Apr+author%3Akonsumlamm+sort%3Aupdated-desc for similar ones, in particular changing code-block (often broken since not run in CI, more verbose, etc) to runnableExamples is an easy way to get started, starting with the most commonly used modules that still rely on code-block

[zetashift]

I would love to do more of these! I am gonna ask a lot of questions though since I'm not familiar with all of the stdlib modules, if that's okay haha!