[Octave-bug-tracker] [bug #45084] regexptranslate incorrectly escapes "<" and ">", maybe others

[Colin Macdonald]

URL:
  <http://savannah.gnu.org/bugs/?45084>

                 Summary: regexptranslate incorrectly escapes "<" and ">",
maybe others
                 Project: GNU Octave
            Submitted by: cbm
            Submitted on: Tue 12 May 2015 01:35:32 PM GMT
                Category: Octave Function
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Incorrect Result
                  Status: None
             Assigned to: None
         Originator Name: 
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any
                 Release: dev
        Operating System: Any

    _______________________________________________________

Details:

regexptranslate has an "escape" mode which escapes characters.

The docs say: "special meaning for regular expressions are escaped so that
they are treated literally."

But some characters without special meaning are incorrectly escaped:

>> s = '<><>'
s = <><>
>> re = regexptranslate ('escape', s)
re = \<\>\<\>
>> regexp(s, re)
ans = [](1x0)


('re' is just 's' again on Matlab and thus 're' does indeed match 's'.  I
believe this is the correct behaviour).

I looked at the implementation and it does look wrong in that it escapes all
non-word characters:


  elseif (strcmp ("escape", op))
    y = regexprep (s, '([^\w])', '\\$1');


This is https://github.com/catch22/octave-doctest/issues/42.

If someone confirms this, I will try to write a patch which explicitly lists
all characters to be translated.




    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?45084>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/