Shell Programming and Scripting

Hi,

I'm new to scripting. I understand the concepts and syntax of some commands but have difficulty with others and combining actions to achieve what I'm trying to do so hope someone on here can help.

A long while back I inherited a website with 1000's of pages most of which were created by a tool I'm unable to use to recreate them from fresh. A lot of these pages have a hard coded link to the home page of the domain where it was originally hosted.

Having changed the domain to something more suitable I'm now stuck with 1000's of pages within perhaps 50 directories that have incorrect and links to the home page of a site that no longer exists.

I understand that sed can easily be used to replace all occurances of a string with another

Code:

's/{old value}/{new value}/'

but I'm having difficulty in extending this to recursively act on a whole site going into all directories and replacing all the old url's with new links.

Some pages may be up to 3 directories deep from the home page but having a clue as to how to execute the command for all files within a single directory would be a start.

A while back someone did something similar to this for me but due to some memory limit on how it was executed any file over 8k in size lost everything after the 8k size mark.

I have access to a unix box where I can place the site to run the command /script and then use the output to replace the existing site.

Any suggestion as to where to start would be much appreciated.

Cheers

Bob

Code:

find /path/to/root/of/docs -type f -exec sed -i 's%http://%https://%g' {} \;

Something along those lines. sed -i might not be available to you, but this same question gets answered here almost daily, so searching along these lines will hopefully help you find something which works for you.

Many thanks for your time. I've had a tinker around, escaped the slashes in the URL's and it works a treat.

Cheers again,

Bob