Hacker News new | past | comments | ask | show | jobs | submit login
Bloodhound – Elasticsearch client and DSL for Haskell (github.com/bitemyapp)
85 points by coolsunglasses on May 4, 2014 | hide | past | favorite | 25 comments



Why you never document your code? You didn't document your Clojure libs and now you're continuing the trend with Haskell. If you're going to make libs, please provide the adequate documentation with examples of usage in the code. I'm not gonna keep going back to the README. I'm going to use Hoogle and Hackage. All those comments in the README could be looking all pretty on Hackage right now.


huh, Chris definitely should add some comments to his code. I'll tease him about it for the next week!


Is there a particular Clojure library that you need more documentation for?

I'll transfer the README docs over to Haddock/lhs when I do the next non-trivial release.


that is frustrated me when entering functional language world. rarely good document compared to other language. In Perl, many modules will have their own tutorial


As far as I know bool filters are preferred over and/or filters (see http://www.elasticsearch.org/blog/all-about-elasticsearch-fi...)

Seminearring and Monoid Filter could encourage usage of the bool filters, instead of and/or.

Thank you otherwise, I was longing for something similar when I was editing JSON queries. Going to try that the next time I have to do some ES work.


This is great advice, I read the follow-up links in that post and they're very informative! I'll jot down a task for figuring out how to flip the Seminearring over to bool.

The reason for the original design was that they mated to nested And/Or filters which I knew to be less than ideal but figured it work as a first pass.

Thanks!


Looks really, really good. My Haskell app is just up to the point that search seems a logic next step and I will surely use ES and your lib for that.


Ping me if you need any help or have suggestions for improvements!


as a contributor to the other ElasticSearch client - kudos. This is much nicer than ours in many ways :)


Thank you very much and thanks to everyone that helped me build this. You, carter, and the others in the Haskell/Haskell.au IRC channels were a big help. The thank-you list would be huge if I tried to list everybody that made a library I used or whose blog post helped me along.


Other ES client? In haskell?


yeah - worked on it a bit with Ollie Charles.

https://github.com/ocharles/Elasticsearch

it's not very principled, tbh. uses error etc rather than proper sum types for returning errors.


To be fair, errors are implicitly encoded in the HTTP statuses and I'm not doing anything about Conduit's exceptions right now. This is something I'd like to fix later.


Ah right, I thought there might be another one I didn't know about! ocharles tends to be a lot more principled these days if it's any comfort :-)


Naming is the same as Statamic Bloodhound plugin. Are those 2 projects related or just coincidence?


Unfortunate coincidence. I've been on a "projects starting with letter B" streak[1] and Bloodhound is search related. Also I like dogs.

[1]: http://bitemyapp.com/projects.html


Bear starts with a B.


Naming a project Bear would be a little too megalomaniacal of me: https://twitter.com/bitemyapp/ :)


But honey is elastic ..


Also Apache Bloodhound.


How popular is elasticsearch?


It's spreading pretty rapidly in my circles. It's used by a lot of popular sites, such as Github, Soundcloud, Stumbleupon, Quora, Foursquare, Etsy, etc.

They have some case studies:

http://www.elasticsearch.org/case-study/

eg. http://www.elasticsearch.org/case-study/github/ http://www.elasticsearch.org/case-study/soundcloud/ http://www.elasticsearch.org/case-study/stumbleupon/


For what it's worth a gg-tends comparison between solr (the main ES competitor I know of) and ES itself.

http://www.google.com/trends/explore#q=elasticsearch%2C%20so...

TLDR; they are roughly equal in popularity according to gg-tends, ES growing very fast.


My impression is that Elasticsearch's growth is coming from being more popular outside of the Java ecosystem.


I run the Elasticsearch Berlin UG and we have a steady and fast influx of new people. I've rarely seen communities grow that fast. The community is great and active and Elasticsearch is caring a lot for that.

http://www.meetup.com/Elasticsearch-UG-Berlin/




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: