Match last instance of pattern

techie411 has asked for the wisdom of the Perl Monks concerning the following question:

I'm trying to do a Perl one-liner to replace ONLY the last </TABLE> Tag and replace it with something. I tried the following, but it doesn't work:

perl -pi -e 's/.*<\/TABLE>/REPLACE/' test.htm

The following doesn't just match the last one but also replaces another </TABLE> tag. Any assistance appreciated. Thanks.

Comment on Match last instance of pattern Download Code

Replies are listed 'Best First'.
Re: Match last instance of pattern by JStrom (Pilgrim) on Nov 05, 2008 at 22:22 UTC
No need to involve the regex engine. `perl -0777 -pe "substr $_, rindex($_,'</TABLE>'), 8, 'REPLACE'" x.html`	[reply] [d/l]
Re: Match last instance of pattern by moritz (Cardinal) on Nov 05, 2008 at 21:59 UTC
If you run perl 5.10.0 or later, you can use this regex: `s/.\K<\/TABLE>/REPLACE/` If you're not that fortunate, you could use captures: `s/(.)<\/TABLE>/${1}REPLACE/`	[reply] [d/l] [select]
Re: Match last instance of pattern by GrandFather (Saint) on Nov 05, 2008 at 22:41 UTC
Is there only one line in the file you are processing? If not you have something of a problem because the substitution is applied to each line in the file. Perl reduces RSI - it saves typing	[reply]
Re: Match last instance of pattern by ccn (Vicar) on Nov 05, 2008 at 21:59 UTC
`perl -0777 -pe '$_=reverse;s{>ELBAT/<}{ECALPER};$_=reverse' test.htm` [download]	[reply] [d/l]
Re^2: Match last instance of pattern by lostjimmy (Chaplain) on Nov 05, 2008 at 22:09 UTC
Are the two reverses (and backward words) better to use than just matching everything before the last </table>? `perl -0777 -pe 's!(.*)</table>!${1}REPLACE!s;' test.htm`	[reply] [d/l]
Re^3: Match last instance of pattern by ccn (Vicar) on Nov 05, 2008 at 22:29 UTC
It depends... How do you compare one JAPH with another?	[reply]