IFSC search is web interface for scrapping, filling up database and then searching on it using an HTML interface.

Production Preview

You need to have the underlying technology installed that run the project

• Phoenix Framwork
• Postgres
• LibreOffice(Optional)

You'll want to clone this repository with git [clone] (https://github.com/vishaldeepak/ifsc_search.git). If you plan on contributing, you'll want to fork it.

enter the root project directory

cd ifsc_search

You can now fetch your dependencies, compile, and run the server:

mix deps.get

Create the database. Make sure you setup your development database name, username/password in dev.exs:

mix ecto.create && mix ecto.migrate

Install Node.js dependencies with

cd assets && npm install

You can now compile and run the server(from the project root directory):

mix phx.server

Now you can visit localhost:4000 from your browser.

For convinence you can simply download or clone this repo. Once done move IFSC_data folder of the downloaded repo into build_data/ of the root project of this repo(IFSC-search).

If you prefer you can setup this on your own but it takes a little more work.

In the root project directory hit the command. This should download all files in build_data/IFSC_data

 mix ifsc_scrapper

Some files downloaded will be in xls format. This project needs all files to be in xlsx.To do that I used libreOffice command line tool.

cd build_data/IFSC_data
soffice --headless --convert-to xlsx *.xls

Now process these files to build data to database. Make sure to go back to the root project directory.

# Add States
mix parse_states

# Add Banks and Branches
mix parse_ifsc

NOTE - The parse_ifsc can take more than a couple of minutes to complete. You can also check for any errors in log/error.log. The project is currently able to parse more than 95% of the rows succesfully. Almost all erros are due to inconsitent data.

Feel free to raise an Issue