pisql

What is pisql ?

pisql is a mix of a command line tool and a small python library to interact with a Sybase ASE database.
It so happens, that I have to work with a Sybase ASE database and I wanted to have a tool to interact with it.

But why ?

The reason I built this is fairly simple although frustrating:

• My company has tight control over what gets installed on the workstations
• My company hasn't bought the SQL Anywhere drivers for ASE on top of ASE itself as of today
• The only way I have to interact with the database is through a very old and ugly piece of software called "sqlDbx".
• I wanted to have scripting and automation capabilities that the interface couldn't provide, but that bash/powershell/python could.

What is it ?

pisql is, for the CLI part, a rich CLI wrapper around the barebones isql command line tool that comes with the ASE installation every time.

For the python part, it's a small library that allows you to interact with the database through python code. You can turn .sql files into dataframes (pandas or polars), and further manipulate them. You can use every tool you have in python to interact with the data, vizualize it, etc.

How do I use it ?

Installation

You can install pisql through pip:

    pip install pisql

Although I recommend pipx:

    pipx install pisql

This works on Windows, Linux and MacOS.

Usage

CLI

The CLI is fairly simple to use.

To execute a single .sql file, you can just use the exec command:

    pisql exec my_file.sql

which is also aliased to pisql e my_file.sql and pisql x my_file.sql.

To execute multiple .sql files, you can use what I call the "query mode" or "run mode", using either
the symbols q, run or ::. Once this more is activated, you can chain multiple executions together
by using the ++ or // commands:

    pisql q ++ file_one.sql file_two.sql file_three.sql

An important feature of this mode, is that you can list both files and directories.

    pisql :: // file.sql some_dir other_file.sql

What happens then is that pisql will execute sequentially:

• file.sql first
• then all the .sql files in some_dir second
• then other_file.sql last

One nice feature is the presence of rich progress bars like so (give example)

NB: I haven't had the time to implement further recursion, so if you have a directory in some_dir, it will be ignored.

Python

Will explain in the next few days.

What's next ?

I'm currently working on a few things:

• Implementing a config file subcommand to set the default database, user, etc.
• Give more freedom to users to change the storage of the dataframes and config files
• Build a semi ORM to interact with the database through python. Comes to mind the SELECT, WHERE, JOIN, etc. clauses.
• Have a little templating feature, but nothing too fancy. I don't want to reinvent the wheel here.
• Use Agg-Grid to have a web interface to interact with the database. I'm not sure how to do this yet, but I'll figure it out.

Here's most of it for now ! I'll update this as I go.