Mirror of forum.nim-lang.org

2666 :: unescape \n \r etc in string

• Krux02 (orginal) [2016-12-15T01:29:58+01:00] view original

  I have a macro with a string argument. The string is partially preserved, but all n from the string are now litterally n and t. so how do I unescape thesevalues to create a string object out of them? All my ideas right now feel like a lot of work with some uncertenty if I did everything correct, so I thought I might better ask, even though the implementation is propably trivial I do not think that the correct implementation is.

---

• cblake (orginal) [2016-12-15T07:39:48+01:00] view original

  Perhaps you want strutils.escape? I.e.:

  import strutils
  echo escape("\n\t")
  Run

  Output is "\x0A\x09"

  Its output is designed to be as portable an 'input' to de-escapers/escape-interpreters as possible. So you will see things like \x0A and \x09 instead of the \n \t because the former syntax is accepted by more string parsers.

---

• Krux02 (orginal) [2016-12-15T16:53:42+01:00] view original

  no, I am asking for the inverse of that function. There is unescape, but it doesn't do what I want to do:

  import strutils
  
  let str = r"\n\r"
  
  echo str
  assert unescape(str, "", "") == "\n\r"
  Run

---

• vlad1777d (orginal) [2016-12-15T21:15:38+01:00] view original

  @Krux02, maybe this is a bug, you could post an issue on GitHub.

---

• cblake (orginal) [2016-12-16T00:12:24+01:00] view original

  Looking at the implementation of strutils.unescape, it seems to only interpret the xHH syntax that escape outputs. Of course, the compiler proper is often changing those raw/quoted string forms into special characters. So, maybe there is some other approach/trick that could work..sort of like an "eval" in interpreted languages.

---