Page 1 of 1
How to find last occurrence of character in string?
Posted: Fri Jul 03, 2009 8:47 am
by mathman99
All,
Is it possible to locate the last occurrence of a particular character in a string,
An example : if the string was "C:\Program Files\GIMP-2.0\etc\fonts", how would you find the position of the last "\" ?
Is it possible to do it purely with functions, i.e. without going into OO Basic ?
regards,
Andrew
Re: How to find last occurrence of character in string?
Posted: Fri Jul 03, 2009 10:44 am
by Robert Tucker
The regex:
will find "\fonts"
Search for:
Replace with:
will replace the "\" with "/" for example.
Re: How to find last occurrence of character in string?
Posted: Fri Jul 03, 2009 5:08 pm
by acknak
Or, something like =SEARCH("\\[^\\]*$"; B2) if you need the position as a number value.
Re: How to find last occurrence of character in string?
Posted: Fri Jul 03, 2009 11:28 pm
by mathman99
Thanks for this everyone, but could someone please break this regular expression down. It is making my head spin.
Also, acknak's version is different than Robert Tucker's, in that the first \\ does not have round brackets around it.
Re: How to find last occurrence of character in string?
Posted: Sat Jul 04, 2009 12:14 am
by acknak
Have you looked at this?
Regular Expressions in Calc [OOo wiki]
Ignore the differences--they're beside the point. The parentheses are only needed if you're using Find & Replace, so I left them out. Robert's regex would work just the same if you used it in the SEARCH example.
Walk through it one element at a time:
- \\ A single, literal backslash character. You have to use two in a row because the backslash is a special character in regular expressions; two in a row means a plain, non-special backslash character.
[^\\] A class (or set) of characters, any of which will count as a match. In this case, the set has only one character, our literal backslash friend again. The caret ^ causes the meaning of the set to be inverted: instead of matching any character in the class, it matches any character not in the class. So this means: "match any character other than a backslash".
* This is a repetition marker; it means the preceding pattern element should match zero or more times. So that's any character other than a backslash, any number of times, including none at all.
$ This matches only at the end of the input.
So all together, it matches a backslash, followed by any number of non-backslashes, followed by the end of the input. That's a long way of saying "the last backslash in the input".