Awesome

ER board

Aggregate ER state liveboard

• grafana: http://er.mohw.g0v.tw
• map demo: http://jsbin.com/mequnuyuqiso/2/edit

CC0 1.0 Universal

To the extent possible under law, Chia-liang Kao has waived all copyright and related or neighboring rights to hackfoldr.

This work is published from Taiwan.
http://creativecommons.org/publicdomain/zero/1.0

Overview

this project contains three parts:

• influxdb, database for store time serial json data from scripts.
• grafana, in gh-pages branch, frontend website.
• crawlers, submodules in this branch(master), to crawling data from hospitals.

Install influxdb

1. we suggest to use v0.8.3(git:fbf9a47) influxdb (this repo built on it, if you use newest version, please check the update information.)
2. Continuous Query: write a Query select * from ER into ER.[hospital_sn] after your database created.

Install crawlers

dependencies

• Task runner (one of below)
  • nodejs v0.10.x or v4.2.3LTS. (npm also required, it usually installed with nodejs)
    (remember install npm modules from package.json. (npm install or npm i))
  • python3.5+ and influxdb library
• python2 and requests library.
• php v5.x and php5-curl. (didn't test on php7, if you do that, PR welcome.)

submodules

• use git clone --recursive https://github.com/g0v/er.mohw to clone submodules in the same time.
• or run git submodule init && git submodule update --init to clone submodules. (cd to repo directory)

Running

Please make sure you already installed influxdb and crawlers.

1. Create a backup directory to save backup.
2. Choice a Task runner to grab data from submodule:

# For nodejs runner
$ node twer.js twer.csv --influxHost [yourHost] --influxDb [yourdatabase] --influxUser [youraccount] --influxPass [yourpass] > temp && ./backup.sh

# For python3 runner
$ python3 runner.py twer.csv -host [yourHost] -db [yourdatabase] -user [youraccount] -pass [yourpassword] > temp && ./backup.sh

Notice:

• replace [yourHost], [yourdatabase], [youraccount] and [yourpass] to yours.
• the node ... command will ONLY grab data once, if you need continuous grab data into influxdb, please use crontab to schedule it. (we suggest 20mins or more)

grafana frontend website

• clone gh-pages branch to another folder.
• modify influxdb:{...} content in config.js to fit your influxdb settings.
• run a website service to provide frontend website.

Other usages

Purify backup files.

python3 clerify.py [--dir <dir path>] [--output <output filename>]

The default dir is backup, and output name is dataset.json. it be fixed to dump a json file, so you don't need add .json.

Dataset scheme

for the raw backup file, will be named as "yyyy-mm-dd_HH-MM", for the <date> below, means yyyy-mm-dd, and <timestamp> means file name to be parse to timestamp.

[{
  "<date>": {
    "<timestamp>": {"<each hospital data as a pydict>"},
  }
}]

Add/update parser

1. write down new parser in gist.
2. insert new row into twer.csv
3. execute runner command:

# update one 
$ node twer.js twer.csv --update --only <hospital_sn>
$ python3 runner.py twer.csv --update --only <hospital_sn>
# or update all
$ node twer.js twer.csv --update
$ python3 runner.py twer.csv --update

influx-db docker images wtih usages

1. docker pull t0mst0ne/influx-er
2. sudo docker run --restart=always -d -p 8083:8083 -p 8086:8086 --expose 8090 --expose 8099 -v /somewhere/data:/data t0mst0ne/influx-er
3. edit crontab with the following lines , assume the repo er.mohw was at /somewhere/
4. */20 * * * * cd /somewhere/er.mohw && /usr/bin/node /somewhere/er.mohw/twer.js /somewhere/er.mohw/twer.csv --influxHost x.x.x.x (ip address) --influxDb twer --influxUser guest --influxPass guest > temp && ./backup.sh