Recipy

A recipe retrieval system which enables users to query recipes by specifying wanted and unwanted ingredients, tools and techniques with considering the unstructured instructions text of recipes.

Installation

1. Download the repository and navigate to the directory that contains requirements.txt file.

2. Create a Python 3.6 virtualenv and activate it.

   python3.6 -m venv recipy_env

   source recipy_env/bin/activate

3. Install the requirements via pip package manager.

   pip install -r requirements.txt

4. Start the development server.

   python recipy/manage.py runserver

5. If above steps were successful, navigate to http://localhost:8000/ on your browser to test the app locally. It will use the remote database. So no database configuration required.

Database Configuration (If you want to use your own database)

1. Download and install PostgreSQL 12 on your machine.
2. Create an empty database on the PostgreSQL.
3. Set the database credential related settings under recipy/recipy/settings.py file.

   DATABASES = {
      'default': {
          'ENGINE': 'django.db.backends.postgresql_psycopg2',
          'NAME': '{DB_NAME}',
          'USER': '{DB_USER}',
          'PASSWORD': '{DB_USER_PASSWORD}',
          'HOST': '{DB_ADDRESS}',
          'PORT': '{DB_PORT}',
      }
   }

4. Migration files are included in the repository so only `migrate command will be sufficient to create tables on the database.

   python recipy/manage.py migrate

5. Insert your own recipe records into the main_recipe table on the database.
6. Recreate indexes and statistical thesaurus.
   1. Open the Python shell of the app.

      python recipy/manage.py shell

   2. Run the scripts over the Python shell of the app.

      from main.utilities.index_creator import *
      
      build_index()
      build_statistical_thesaurus()

   3. Save the indexes as binary files with pickle over the Python shell of the app.

      import pickle
      from main.utilities.index_creator import *
      
      with open('pickles/all_documents.txt', 'wb') as f:
         pickle.dump(ALL_DOCUMENTS, f)
      
      with open('pickles/index.txt', 'wb') as f:
         pickle.dump(INDEX, f)
      
      with open('pickles/thesaurus.txt', 'wb') as f:
         pickle.dump(STATISTICAL_THESAURUS, f)