You could run a command on the files returned by the find command.
For example with the following command I list all the files starting from the current directory and on each I run the grep command to search for the string 'getUri'
find . -name '*.*' -exec grep --color 'getURI' {} +
This works perfectly on my Ubuntu 12.04
Anyway, I do not think that grep command is able to search within binary documents such as PDF.
Also using the command above on a large directory tree could be very burdensome in terms of computational time.
Another aspect to consider is that you can not search for a string in a raster PDF, in suck kind of scenario a document management system as LogicalDoc could help.