Replace every instance of a string with another string in bash

[siriusfox]

Does anyone know of a way, within bash (linux), to within every file in a directory, replace every instance of a string with another. I've got a very poorly written web script that I need to do some modifications to, but the files within said directory statically rely on the name of the directory.

So within directory abc lets say, there are a bunch of PHP, config, xml, and some static HTML files. I need a script or command, that will go through all of those files and replace every instance of abc with abc_beta or something of that nature.

Any help would be great.

-Siriusfox

[soulbleed]

find . -name '*.html' -print0 | xargs -0 perl -pi -e 's/oldstring/newstring/g'

[puredistortion]

Here is an example all done in bash so you can use on system with no perl.

${string/replace/parttern/}

With that pattern is just a standard REGEX I .have included an example that will clear out the 1 for a string 1234.

[bash]$ a=1234
[bash]$ echo ${a//[^2-4]/}
234
[bash]$

[cooper]

find . -name '*.html' -exec sed -i 's/oldstring/newstring/g' {} ;

Note that the slashes before the accolades and the semicolon, as well as the space before the escaped semicolon are required. FYI, that bit get replaced by every matching file that find is able to find.