Re: tlug: Re: search & replace (was: Grep for subdirectories?)

[Rex Walters <rex@example.com>]

On Mon, Oct 05, 1998 at 02:30:16PM +0900, Scott Perlman wrote:
> actually, just in the spirit of the subdirectories it should probably be 
> find . -name '*.html' -e 'perl -pi.bak -e "s/match/replace/" {} ;'
> 
> (all that is is wrapping The provided perl witha find)

Actually, to be pedantic, Darren wrote:

> > > I've a similar question: How do I do a search and replace operation on
> > > files?

Thus I didn't specifically include any recursion into sub-directories.

But if you really wanted to do the search and replace on all the .html
files in an entire directory *tree*, you might be better off doing it
differently from how you've described unless you're quite certain that
there are only a few .html files.

The problem is that yours will fire off a new perl process for each and
every .html file found.  This can be pretty slow on a under-powered
(under-memoried) machine.  With only 200 html files it probably doesn't
matter (but if they are *big* html files ... ;-).

Better would be to do the recursion/inode-walk within perl (the find
module is useful for this) or to feed the filenames into stdin.  Both
options are sufficiently ugly as a one liner that I would probably break
down and write a real script.  :-)

Regards,
-- 
Rex
---------------------------------------------------------------
Next Meeting: 10 October, 12:30 Tokyo Station Yaesu central gate
Featuring the IMASY Eng. Team on "IPv6 - The Next Generation IP"
Next Nomikai: 20 November, 19:30  Tengu TokyoEkiMae 03-3275-3691
---------------------------------------------------------------
Sponsor: PHT, makers of TurboLinux http://www.pht.co.jp