Changes: Alternate delimiters for the replace command

Latest revision as of 05:28, 13 July 2012

Tip 406 Printable Monobook Previous Next

created 2003 · complexity basic · author Raj Kiran · version 5.7

You don't need to use the slash character as the expression delimiter. Whatever character follows the :s is defined to be the delimiter character. You can use most non-alphanumeric characters (but not \, " or |). This is very handy when working with Unix filenames, as in the following example:

:s#/usr/local/#/opt/#

If you insist on using the normal slash, you have to escape each slash in the pattern with a backslash:

:s/\/usr\/local\//\/opt\//

As you can see, the escaping method is much less readable, so consider carefully your choice of delimiter character!

Comments[]

What a great tip. Very nice when trying to comment out Java or JS. e.g. :5,8 s#^#//# which looks much nicer than :5,8 s/^/\/\//

The percent ('%') character is also a frequently used alternative. Both percent and the octothorpe ("pound sign") ('#') characters have '/' embedded in them, to visually cue you why they are there as delimiters.

Other common choices are the hypen and underscore. (Protip: This tip can be applied in a lot of other places, especially when writing in a language (e.g. JavaScript, Perl).) SheeEttin 04:29, July 17, 2010 (UTC)

A comma delimiter works well and seems less cluttery. This is particularly handy for text, HTML/XML or Ruby.

@@ Line 1: / Line 1: @@
-{{review}}
 {{TipImported
 |id=406
 |previous=405
 |next=407
-|created=January 17, 2003
+|created=2003
 |complexity=basic
 |author=Raj Kiran
 |version=5.7
 |rating=56/20
+|category1=
+|category2=
 }}
+You don't need to use the slash character as the expression delimiter. Whatever character follows the <code>:s</code> is defined to be the delimiter character. You can use most non-alphanumeric characters (but not \, " or |). This is very handy when working with Unix filenames, as in the following example:
-This text is adapted from http://www.troubleshooters.com/lpm/200212/200212.htm
+<pre>
-You don't need to use the slash character as the expression delimiter. You can use most non-alphanumeric characters (but not \, " or |). This is very handy when working with Unix filenames, as in the following example:
+:s#/usr/local/#/opt/#
+</pre>
+If you insist on using the normal slash, you have to escape each slash in the pattern with a backslash:
- :s+/usr/local/+/opt/+
+<pre>
-Whatever character follows the <tt>:s</tt> is defined to be the delimiter character.
+:s/\/usr\/local\//\/opt\//
+</pre>
+As you can see, the escaping method is much less readable, so consider carefully your choice of delimiter character!
-The alternative, is to use the normal slash, but escape each slash in the pattern with a backslash:
- :s/\/usr\/local\//\/opt\//
-As you can see, the escaping method is much less readable, so if you can use alternative delimiter characters, it's a good idea.
 ==Comments==
+What a great tip. Very nice when trying to comment out Java or JS. e.g. :5,8 s#^#//# which looks much nicer than :5,8 s/^/\/\//
+The percent ('%') character is also a frequently used alternative.
-<pre>
+Both percent and the octothorpe ("pound sign") ('#') characters
-:'a,'bs#str1#str2#gci
+have '/' embedded in them, to visually cue you why they are there as delimiters.
-:1,23s=str1=str2=gci
-</pre>
+Other common choices are the hypen and underscore. (Protip: This tip can be applied in a lot of other places, especially when writing in a language (e.g. JavaScript, Perl).) [[User:SheeEttin|SheeEttin]] 04:29, July 17, 2010 (UTC)
-I use one of the above, that is, either <tt>=</tt> or <tt>#</tt>
+A comma delimiter works well and seems less cluttery. This is particularly handy for text, HTML/XML or Ruby.
-----