Overview

This is a scraper for the website seloger.com, built using Scrapy.

Setup

• Install virtualenv pip install virtualenv
• Create a new virtual workspace virtualenv my_workspace && cd my_workspace
• Clone the project into your workspace folder:
  git clone https://github.com/falcononrails/scraper-seloger.git and navigate to it cd scraper-seloger
• Install the required packages pip install -r requirements.txt

Usage

Go to the project folder

cd ~/my_workspace/scraper-seloger/simple_seloger

Run the spider with the URL of your search query on seloger.com

scrapy crawl seloger -a search_url="https://www.seloger.com/list.htm?tri=initial&enterprise=0&idtypebien=2,1&pxMax=1000000&idtt=2,5&naturebien=1,2,4&ci=910377"

You can use the -o option to specify an output file (JSON or CSV):

scrapy crawl seloger -o annonces.csv -a search_url="https://www.seloger.com/list.htm?tri=initial&enterprise=0&idtypebien=2,1&pxMax=1000000&idtt=2,5&naturebien=1,2,4&ci=910377"

Exporting the data to MongoDB

• Make sure you have MongoDB installed and its deamon running.
• Change MONGO_URI and MONGO_DB in the settings.py file of the project.
  A typical scenario is MONGO_URI = mongodb://localhost:27017 and MONGO_DB = seloger
• You can use Robo 3T to see your database and manipulate the data.

Deploying on Heroku

The repo has all the files you need to deploy to Heroku, I'll clarify the steps below.

• Install Heroku CLI
• Create a new Heroku app heroku create seloger-demo
• Add the new app as a remote heroku git:remote -a seloger-demo
• Change the url argument under the [deploy:local] section in the scrapy.cfg file to url = https://seloger-demo.herokuapp.com/
• Add git add . and commit everything git commit -m "first commit"
• Finally push to Heroku and watch your app deploy git push heroku master

The scrapyd interface is now accessible through https://seloger-demo.herokuapp.com/

To start a job through scrapyd, run the following from your terminal:

curl https://seloger-demo.herokuapp.com/schedule.json -F project=default -F spider=seloger 
-F search_url="https://www.seloger.com/list.htm?tri=initial&enterprise=0&idtypebien=2,1&pxMax=1000000&idtt=2,5&naturebien=1,2,4&ci=910377"

Adding MongoDB Lab module through Heroku

• Add the add-on to your existing app heroku addons:create mongolab:sandbox --app seloger-demo
• Get the mLab URI heroku config:get MONGODB_URI --app seloger-demo
• Replace MONGO_URI and MONGO_DB in the settings.py file of the project with the values returned in your terminal.
• Example:

MONGO_URI = 'mongodb://heroku_v10nm298:[email protected]:37740/heroku_v10nm298'
MONGO_DB = 'heroku_v10nm298'

The data you scrape will now be saved in an external cloud MongoDB database linked with your Heroku app.