TakeMyMoney

Running instructions

1. Run manager-osx (for Mac) or manager-windows (for Windows)

• Go to Manage Servers
• Ensure that only the PostgreSQL Database is running

2. Run use_djangostack

• For Windows, run use_djangostack.bat
• For Mac, open Spotlight and run use_djangostack

3. While in the use_djangostack terminal:

• Go to the directory containing the manage.py file.
• Run python manage.py runserver. Your server should now be running at localhost:8000.

Tips

Running .sql files without copy-and-pasting

Instead of copying the SQL statements into the psql terminal and running it there, you can also run the following command: psql -U postgres -d djangostack -a -f [.sql filename]

This is especially useful for the database seed files, such as funding.sql, users.sql or projects.sql.

For this command to work, you have to be on the same folder as the .sql file.

NOTE: This command has a small issue where it doesn't update the pid / user_id sequences, so it breaks the registration/project creation flow. See the fix below.

Issues

"The env.py file cannot be imported"

1. Copy env.example.py, paste it in the same folder with the name env.py.
2. Change the PASSWORD string in env.py to match your Djangostack installation details.

"Relation does not exist"

This means that your database tables do not exist yet. To fix this, follow the steps:

1. Run the use_djangostack terminal.
2. Run psql -U postgres djangostack, and type in your password.
3. Copy the relevant content from schema.sql (in this repository) and paste in the terminal, press Enter.

When registering a new user or adding new projects, the database attempts to add a pid/user_id starting from 1 even after inserting the seed data

This happens if we run the seed SQL queries from the bash terminal instead of copy-pasting the queries manually and running them from the psql terminal.

The SERIAL data type is implemented with a backing Postgres sequence - for our database these sequences are called projects_pid_seq and users_user_id_seq.

Whenever we INSERT into the projects table, the pid is automatically set with the next value on the sequence (you can see the next value in the sequence using SELECT nextval('projects_pid_seq')), and the sequence is also automatically advanced.

When we run the seed SQL queries from the bash terminal, for some reason the backing sequence is not automatically advanced. This results in the next value on the sequence still staying at 1 even when we already have IDs up to 20 (for the users table) or 100 (for the projects table).

The fix is to manually set the sequence value to the correct number again, and this is done via the setval call, as such:

SELECT setval('projects_pid_seq', (SELECT MAX(pid) FROM projects));
SELECT setval('users_user_id_seq', (SELECT MAX(user_id) FROM users));