Quick Start

Welcome to Tad!

If you are reading this message, Tad has been succesfully installed.

Launching Tad

There are three ways to launch Tad:

1. Application Icon

   You can double click on the Tad application icon to open Tad. Use File...Open and the standard file open dialog to open a .csv or .tad file.

2. Context Menu or Drag and Drop

The installation process registers Tad as a viewer for .csv files. On any .csv file in the Finder (macOS) or Explorer (Windows), use Open With... and select Tad to open the file in Tad.

3. From the Command Line

   Once Tad is installed and available on your PATH, simply type:

   $ tad somefile.csv
   

   to launch Tad to explore somefile.csv. You can also open .tad files (previously saved Tad view configurations) via the command line.

Basic Usage

You can explore the basics of Tad by opening this example CSV file (from the IMDB 5000 Movie Dataset on Kaggle) included in the Tad distribution:

Open movie_metadata.csv

When you open a CSV file in Tad, Tad automatically imports your CSV file into an internal in-memory SQLite database. You can see the result of this import process by clicking on the gear icon on the left of the Tad window:

This will open the sidebar:

Tad scans your data file during the import process makes a conservative guess at the type of each column. For now the only supported types are text, integer and real, and there is no facility for manually overriding the inferred types.

The name and type of each column are shown in the main Column Selector above. The additional controls in the column selector enable you to quickly build a pivot table to explore your data set:

• Use the checkboxes under Pivot in the main column selector to choose columns for the pivot hierarchy.
• Use the checkboxes under Show in the main column selector to determine what columns are shown / hidden.
• Set the sort columns using either the checkboxes under Sort the main column selector or by clicking on column headers in the main grid.

Additional controls for setting the order of columns, pivots and the sort key, and controlling aggregation and formatting rules are available under Additional Properties at the bottom of the side bar.

Filtering

Tad provides rudimentary facilities for constructing a filter to select a subset of data. A filter consists of a collection of simple predicates that compare a value in that row (identified by column name) with a constant.

Activate the filter pane by clicking the Filter link at the bottom of the window:

When opened, a (populated) filter dialog looks like this:

Construct the filter by selecting the logical operator (AND or OR) (1) and editing individual predicate rows (2). Add more rows by pressing the + button (3), and delete predicates by clicking the x button (4) next to that predicate.

Discard, test or commit your changes using the three buttons in the lower right of the pane:

Data Sources

If you work in data science, chances are you have no shortage of CSV files of your own to explore. In addition to the movie_metadata.csv sample file included with Tad, a few of my favorite data sources on the Internet are:

• Kaggle Datasets. The movie_metadata.csv file used for many Tad examples is from the IMDB 5000 Movie Dataset on Kaggle.
• Met Museum Open Access Initiative. The Metropolitan Museum of Art recently made 420,000 artworks available online for unrestricted use. The catalog file MetObjects.csv available at the linked github repository is a good example for testing Tad's ability to handle larger CSV files. It's a 228MB CSV file with 420,000 rows of data. It takes about 13 sec to load and display this file in Tad on my late 2013 MacBook Pro.
• data.world is a site for sharing and exploring data on a variety of topics.
• The FiveThirtyEight github repository has many CSV files providing data behind the stories and interactives on the FiveThirtyEight web site.

Going Further (Developers and Data Hackers)

If for some reason you want to see the gnarly details of the SQL queries that Tad is constructing and executing, you can use the -f option on the command line to keep Tad in the foreground and --show-queries to get Tad to print the generated SQL queries like so:

$ tad -f --show-queries somefile.csv

Not for the faint-hearted.

And Finally...

Have fun! I hope you find Tad a useful utility for viewing and analyzing data. Feedback welcome.