GetAllTheTexts

A CFC wrapper to the Apache Tika project. Basically - pass a file name to the CFC and it will attempt to get metadata and text from the document.

Getting Started

Create an instance of the CFC. Like:

Then call read() on a file name.

<cfset stuff = gatt.read("/Users/ray/Documents/foo.doc")>

The result is a structure with 2-3 keys. If everything worked well, you get one key called metadata and one key called text. I bet you can guess what is returned in each key.

If things went poorly, you get a key called error with a stack trace.

The file, test1.cfm, shows a simple example of the API. It reads in all the files in a subdirectory called sources and shows the result for each.

Warning

Both myself and other tests have seen out of memory errors when run on large files. Increasing the max memory JVM setting in the CF admin should help with this.

Credit

Obviously credit goes to the Apache Tika team, but also Mark Mandel and his excellent JavaLoader project.