Pirate Redmine Api Reverse Proxy Cache

early beta (big bordel) stage

merging, filtering, sorting and autoupdating ability

Source data - Czech Pirate Party Redmine project management server JSON API
https://redmine.pirati.cz/projects/snemovna/issues.json

Output purpose - Czech Pirate Party webpage media results application (subpage)
https://pirati.cz/vysledky

Based on hapi.js 18.1.0 - Server Framework for Node.js

Utilizes Wreck library for Redmine API requests
Requires Node.js > 8.12

Handles requests from web application.
Merges and process requests to redmine API.
Caching data in catbox-memory.

Reason

1. Very slow response of Redmine API (e.g. more than 3 seconds for 100 records)
2. Paginated output of Redmine API (max. 100 records per request)
3. Data filtering
4. Data sorting

Win/Win

1. Now the response is sorted and filtered (smaller payload by aprox. 25%)
2. Server keeps cached data in memory and automaticly reload them every 30 minutes
3. Response time is below 200ms

Install

git clone https://github.com/madbeyk/pirate-redmine-api-reverse-cache.git
yarn install

Set-up

process.env.HOST - IP adrress of server (default 0.0.0.0)
process.env.PORT - running port (default 80)

export PORT=3001
export HOST=your.host.ip

Run

yarn run start2

Todo

• Caching
• Filtering
• Sorting by resorts and internal priority of each issue
• Merge paginated requests to redmine (max. 100 results in one API response)
• Automatic data update every 30 minutes
• Advanced external API error handling (outtages ...)
• Image processing (caching, resizing, minimizing ...)

];()~~~~~