議員投票指南
Hackpad 開發討論區
Hackpad 意見回饋

• crawler
  各縣市議會的crawler，各crawler名稱與功用如下：

  councilors: 現任議員資料
  councilors_terms: 歷屆議員資料（不一定包含現任的資料）
  bills: 議案資料
  meeting_minutes: 議事錄資料（開會出缺席、表決）

• data
  由上述crawler產出的各縣市原始JSON

  util/prettyjson.py: 產出indent好讀版的JSON, README
  pretty_format: 放置上述產出的各縣市好讀版JSON
  hashlist_meeting_minutes-v141001.json: links map, 存放由meeting_minutes cralwer抓下的binaries detail
  candidates_2014.xlsx: 中選會公告的議員候選人

• parser
  將上述data下的JSON標準化後放入database（如果你只是需要完整的database，可直接跳至Restore DB）

  councilors/councilors.py: 處理現任和歷屆議員資料
  councilors/candidates.py: 處理候選人資料
  bills/bills.py: 處理議案資料
  votes/: 出缺席、表決資料，各縣市、各屆分開處理

• voter_guide
  Web application using Django, Enviroment Setup

Scrapy offcial install doc

sudo apt-get install libxml2-dev libxslt1-dev python-dev libffi-dev python-pip
sudo pip install lxml
sudo pip install Scrapy
sudo pip install requests

After install scrapy, you can run commands to test, below using tcc(臺北市議會) for example:

cd crawler/tcc
scrapy crawl bills
scrapy crawl councilors
scrapy crawl councilors_terms
scrapy crawl meeting

If you want to output json file:

cd crawler/tcc
scrapy crawl bills -o bills.json -t json
scrapy crawl councilors -o bills.json -t json
scrapy crawl councilors_terms -o bills.json -t json
scrapy crawl meeting -o bills.json -t json

install basic tools

sudo apt-get update
sudo apt-get upgrade
sudo reboot
sudo apt-get install git python-pip python-dev python-setuptools postgresql libpq-dev
sudo easy_install virtualenv

It is quite big now. please be patient. don't use command like git --depth

git clone https://github.com/g0v/councilor-voter-guide.git       
cd councilor-voter-guide/voter_guide/

(if you don' mind packages installed into your local environment, just pip install -r requirements.txt)

virtualenv --no-site-packages venv      
source venv/bin/activate        
pip install -r requirements.txt     

We use SQLite as the default database, if you want to use another database, please set your database engine in local_settings.py.

Create Table

python manage.py syncdb --noinput 

This step may take some time, be patient.

python manage.py loaddata db.json

python manage.py dumpdata --exclude auth.permission --exclude contenttypes > db.json

python manage.py runserver

Now you should able to see the web page at http://localhost:8000

###Prepare Compiler

There are some python package written in C or C++ such as lxml. so a compiler is required. you can install a compiler via the following command:

xcode-select --install

###Prepare PostgreSQL

You can install the packaged app here. put the app in your Application folder and click it to start.

And please add the following line to your ~/.bash_profile

export PATH=/Applications/Postgres.app/Contents/Versions/9.3/bin/:$PATH

please change the version number 9.3 if you download a different version of PostgreSQL.

after you add the PATH environment variable, source it.

source ~/.bash_profile

if you don't add the PATH variable, installation of psycopg2 will not success.

git clone https://github.com/c3h3/g0v-cvg-pgdata.git && cd g0v-cvg-pgdata && tar xfzv 47821274c242ce68f2d8d18d4bb0d050d6481311.tar.gz

• After that, you will get pgdata dir.
• Assume pgdata's absolute path is "your_pgdata"

docker run --name pgdb -v your_pgdata:/var/lib/postgresql/data postgres:9.3

If you want to use pgadmin connect with your db, you could also forwarding the port out ... with command ...

docker run --name pgdb -p 5432:5432 -v your_pgdata:/var/lib/postgresql/data postgres:9.3

• "your_pgdata" is pgdata's absolute path in previous step.

docker run --name g0v-cvg-web --link pgdb:postgres -p port_on_host:8000 -d c3h3/g0v-cvg-web

• "port_on_host" is the port forwarding out to your host, which you could find your web on http://localhost:port_on_host

docker run --name g0v -p forward_port:6800 -v outside_items:/items -v outside_logs:/logs -d c3h3/g0v-cvg-crawler

• "forward_port" is the port you want to forward into docker image (EXPOSE 6800)
• "outside_items" is the directory you want to mount into docker image as /items
• "outside_logs" is the directory you want to mount into docker image as /logs

docker run --link g0v:g0v -it c3h3/g0v-cvg-crawler /bin/bash

in a running docker instance which linked with g0v (Scarpy Server), you can use the following command to deploy tcc crawler to server:

cd /tmp/g0v-cvg/crawler/tcc && python deploy.py

in a running docker instance which linked with g0v (Scarpy Server), you can use the following command to deploy tcc crawler to server:

cd /tmp/g0v-cvg/crawler/bin && python crawl_tcc_bills.py

CC0 1.0 Universal
This work is published from Taiwan.