Web Services, Temp Files, and Program Memory

Question

My team is currently developing a resume-parser for a website. Our parser will translate and format the resume into the industry-standard HR-XML. The website will then take the HR-XML-formatted information and pre-populate editable fields so the user can finalize his/her profile on the website.

What would be the best way to port the HR-XML information to the website? Should we store the XML tags in program memory and have the website call a retriever method in our software? Or should we create a temporary file for each resume that is uploaded to the site? If so, where should this file be stored, and how should we go about maintaining our directories so they are not crowded with temp files?

Any insight would be greatly appreciated! Thank you in advance for your time and your help.

Answer 1

There are a lot of things implied by your questions. If I understand, it sounds as if you're going to import/snarf resume content from external sources, then allow users to fill-in-the-blanks to update the information. Additionally, it sounds like this is existing code you have in place that runs in a non-web environment. Please clarify if that's correct, and I'll update my answer further.

Answer 2

A lot depends on the size of the temp files and the volume of resumes you expect to process. I'd recommend writing them to a temp directory that gets cleared at server start and shutdown. This is often useful when diagnosing server misbehavior. Writing them to disk also enables you to run a job run periodically that clears out "old" entries.

Keeping a collection of entries in memory on the server probably makes more sense if volume is moderate and response time is a big factor, but IME container memory is usually more expensive than disk space.