Upsource - why does it index?

I am rather confused as to what indexing does. My Mercurial Repo is about 1GB in size. After indexing 100 days worth an additional 16GB of disk space has been consumed. There is then the huge length of time it takes to index new revisions and the fact that I can't quite see what indexing gives me before / after my 100 day limit other than a little icon telling me the revision hasn't been indexed. So what exactly does indexing add to Upsource? It feels like a process / memory / disk space intensive distraction.

Good work BTW - just confused when the likes of Fogcreek's Kiln and Atlassian's Crucible seem to manage without.

Comment actions Permalink

Hi Jo,

The key features that indexing provides are reflected on that page -

If to summarize, they are:

Yes, indexing process is greedy for system resources, but we are continuesly working to improve it in future releases.

Please let me if it answers on your questions.


Comment actions Permalink

Good response. Thanks. In future realeases it would be really nice to be able to switch some of those features off. Not working in Java and having a rather ungainly code base, code analysis is not required. Niether are analytics which I can see managers loving but for software devs, they are just a distraction discussed over a coffee and then promptly ignored. Really good work though. I like what I am seeing so far and look forward to future releases.

Kind regards

Jon Dittman

(Not sure how Jo got in there but ho hum! :))


Please sign in to leave a comment.