Mirror of forum.nim-lang.org

8755 :: diff lib?

• oyster (orginal) [2021-12-30T03:52:58+01:00] view original

  I have checked diff, but the result, at least for me, is useless since I can't read on which place the delete/change/insert operation occurs

  import experimental/diff
  var
      x = "apple"
      y = "Apple!"
  
  echo x
  echo y
  
  echo diff.diffText(x, y)
  Run

  outputs

  
  @[(startA: 0, startB: 0, deletedA: 1, insertedB: 1)]
  Run

  so is there something like https://github.com/edg2s/google-diff-match-patch or https://docs.python.org/3/library/difflib.html

  BTW, https://github.com/edg2s/google-diff-match-patch abandoned C++ version, which requires the Qt library, long time ago.

  thanks

---

• Yardanico (orginal) [2021-12-30T04:57:25+01:00] view original

  There's https://github.com/mark-summerfield/diff which is a Nim implementation of the Python difflib

---

• cblake (orginal) [2021-12-30T22:33:07+01:00] view original

  I think my port of that same Py difflib algo is a bit tidier and also quite a bit faster: https://github.com/c-blake/hldiff

  In the near future (whenever duplicate keys in std/tables.Table move from deprecated to gone - a move I still consider better resolved through documentation improvements), this will grow a package dependency on https://github.com/c-blake/adix

---

• oyster (orginal) [2022-01-05T15:38:40+01:00] view original

  thanks. but how to take space into account. For

  
  a = "the quick brown fox jumped over the lazy dogs"
  b = "quic k brown fox  jumped over the very busy dogs"
  Run

  please note that there are 2 spaces between fox and jumped in b. what I expected is that space is treated too. so the result I expected is something like, where I use ␣ for space display, and to be clear, I do not show aStart/aEnd/bStart/bEnd, on the other hand I use text directly

  
  delete a: the␣
  
  equal: quic
  
  insert b:  ␣
  
  euqal: k␣brown␣fox␣
  
  insert b:  ␣
  
  equal: jumped␣over␣the
  
  replace a: laz
  replace b: very␣bus
  
  equal: y dogs
  Run

---

• oyster (orginal) [2022-01-05T15:42:01+01:00] view original

  yes, I have raise an issue

---