This repository was archived by the owner on Nov 7, 2018. It is now read-only.
Tags: 18F/open-data-maker
Tags
API documentation now at API.md Disable sessions (therefore CSRF protection) to permit caching Migrate from Unicorn to Puma with a reasonable prod config Increase unicorn workers from 1 to 5 increase in unicorn workers requires more memory Set Cloud Foundry instances and env var defaults version 1.0 -- note this is a code version which is de-coupled from API version but we're calling 1.0 since we expect to be final for production (!)
Use `_source: false` to limit JSON coming back from ES
exclude fields starting with _
when the whole source is returned
when we’re not specifying fields
we need to explicitly exclude _names
Log ES query time, and show with "debug" option
Also record "took" MS value from ES result
also the "oj" gem for faster JSON
Only allow up to MAX_PAGE_SIZE per page
autocomplete type alpha sort
option parameters should now be underscore-prefixed
autocomplete type for closer to what people are doing with search
using the ‘common’ query with ‘and’ operator
we can reduce the noise of the word “university”
indexes lowercase, by 3-25 letters in any word
searches lowercase, ignoring “stop words”
word boundaries determined by unicode standard text segmentation
in dictionary calculate expresssion, 0 -> false
In ruby, 0 evaluates to true, but that’s not expected out of context
handles integers and floats
API error handling, part 1
docs & improvements to developer experience
add some brief explanation to the README, plus
added additional related docs that were missing
new dictionary_only option added to sample-data
so new open source devs get a good experience
CloudFoundry service names configurable with environment vars
Proper REST errors from the controller
PreviousNext