Data-modeling clipf

Copyright (C) 2016 Dominic Walden.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license can be found here: https://gnu.org/licenses/fdl.html.

I want to get better at data modeling in software testing. I am using the command-line personal finance manager clipf. I will try to test it with respect to the data it processes.

I installed clipf via Debian repositories (version 0.4). I have not looked to see if this is the latest version.

I read the man page which told me that it takes one argument which is the directory of the configuration files. If this is not passed it uses ~/.clipf/ which is will create if there is not one already.

I called clipf from Emacs' ansi-term, and saw that ~/.clipf/ is created with 3 empty files and one file which contains its version number. Apparently, there is a "default configuration" but clipf.conf is empty, so perhaps stored in the program itself or elsewhere. EDIT: default appears to be stored in /etc/clipf.conf, but is all commented out.

From reading man and config files, I can:

• Change default account
• Control max lines of output (over this output in 'less')
• Change encoding of terminal
• Change encoding of db files
• Change colour output options
• Set aliases for commands

According to the manual, I can also set default date and "reporting period".

Options can be set in config files or while using the program.

Objects include:

• Accounts
• Items
  • Has propery "income" or "expense"
  • Is either an item or an item group
  • Item groups can have subitems (how deeply nested?)
  • Item codes are supposed to be unique
• Item Groups (treated differently to Items?)
• Operations
  • Has a date
  • Is associated with an account
  • Is categorised by an item (how many?)
  • Can have Tags (how many?)
  • Can have a note (how large?)

1. item code does not have to be unique (but program appears to only use most recently created item with particular code.)
2. When deleting items, prod rm <code> will delete all items whose item code begins <code>. Don't appear to be able to not do this. This is how it is described as working in the manual, but doesn't seem very desirable.