Mirror of forum.nim-lang.org

606 :: proc match always returns false

• drifter (orginal) [2014-10-29T02:15:18+01:00] view original

  Hello,

  I am having a hard time using the match proc to catch occurences of a regular expression in a string. It is the same regex I use in Perl, but it does not seem to match here:

  
  import parseutils
  import strutils
  import re
  import pcre
  import unicode
  
  type TMatches = array[0..1, string]
  var matches: TMatches
  var currentline = "some <bold<text> and <italic<text>.  Also some <cmd 1, cmd 2, cmd 3<text>."
  var regex: TRegex = re"<.*?<"
  
  if match(currentline,regex,matches):
    echo(matches[0])
    echo(matches[1])
  else:
    echo "No matches"
  Run

  Anyone who could point out where I am going wrong would be appreciated.

  Thanks!

---

• Stefan_Salewski (orginal) [2014-10-29T13:31:25+01:00] view original

  Have tested a simplified example

  import re
  
  type Matches = array[0..1, string]
  var matches: Matches
  let currentline = "Hallo <Stefan>"
  let regex = re"(Hallo)\s(<\w*>)"
  
  if match(currentline,regex,matches):
    echo(matches[0])
    echo(matches[1])
  else:
    echo "No matches"

  Output for bigbreak is as exspected: Hallo <Stefan>

  Currently I have no idea what is wrong with your example, maybe I can do more testing this evening. (I have only experience with Ruby pattern matching, Nim libraries may work not exactly the same...)

---

• Jehan (orginal) [2014-10-29T13:56:55+01:00] view original

  re.match requires that the entire string matches the regular expression. You probably want to use re.find(...) >= 0 for your example instead, which looks for a substring matching the regular expression (or pad the regular expression at the begin and end to match the remaining characters).

---

• will (orginal) [2014-10-29T13:56:59+01:00] view original

  I believe your pattern must match the whole line. You also don't have any capturing groups to populate the matches array.

---