Mirror of forum.nim-lang.org

2691 :: (string + string) and (string * int)

• Allosteric (orginal) [2016-12-31T17:58:29+01:00] view original

  Couldn't render post #16649.

---

• Stefan_Salewski (orginal) [2016-12-31T18:54:12+01:00] view original

  You can define them for yourself when you like it. But for most decisions Andreas had good reasons. For not using + for string concatenation there is one, but I have some problems to remember. May it be that we may like to write

  "1".toInt + "2".toInt

  Something similar it is, but I really am not sure. Maybe mentioned in Dom's book.

  And for repetition there is stringutils.repeat(). Maybe * is not used for the same reason as +. And indeed & is fine for concatenation. For + some people would say that it is bad for strings, because concatenation is very different from mathematical addition.

---

• Araq (orginal) [2016-12-31T20:26:16+01:00] view original

  Reason:

  @[1, 2, 3] + @[1, 1, 1] == @[2, 3, 4]
  @[1, 2, 3] & @[1, 1, 1] == @[1, 2, 3, 1, 1, 1]
  Run

  I prefer repeat over * since I want the string to be repeated default to the space character and that would look alien with a multiplication operator. Multiplication should also be commutative for consistency so that would be two procs instead of one.

---

• Allosteric (orginal) [2017-01-01T01:54:29+01:00] view original

  I see the point. Now it all seems a well-reasoned decision to me too. Thanks a lot!

---

• leepysham (orginal) [2017-01-01T02:53:41+01:00] view original

  Since Stefan_Salewski mentioned math, I thought I'd point out that in math, + is almost always taken to be commutative, and string concatenation is not commutative. Usually when talking about string concatenation in math, we use *, and for repetition, we use s^n. But since there's no implicit conversion from the literal "1" to the empty string, & is okay ;)

---