Dedupe-Web issue

[Eric van Zanten]

Hey there, I was attempting to respond to a post from bootlear about an issue he was having getting the Spreadsheet Deduper running locally but it looks like someone deleted the thread. Just in case someone out there is still listening, here's my response:

I'm the guy who is mainly responsible for putting together the Spreadsheet Deduper and I think I figured out what's going on. The last time I touched that code was in between the 0.5 and the 0.6 release which means that only some of the API changes were incorporated. I just pushed up a commit this morning that should fix the issue you were running into.

A couple things to note: 

1) The error traceback that you were getting from the "run_queue" process is actually significant and means that, for whatever reason, that process was unable to connect to Redis. 

2) Since you're running this locally, you might want to not have to deal with the limit that we set at 10,000 rows for an uploaded spreadsheet. Frankly, this is a rather arbitrary limit that we set for the version that is deployed at dedupe.datamade.us. If you want to remove that limit, you can comment out these lines: https://github.com/datamade/dedupe-web/blob/master/dedupe_utils.py#L57-L60

Let me know if you're still running into issues after pulling the new commit I made this morning.

Eric