Skip to content



Views are “virtual tables”. Data is presented in rows and columns just like a table, but these rows and columns are calculated on the fly by pulling other data from wherever it is stored.

Views can involve combining data from multiple tables or other views, filtering, sorting, aggregating (grouping), or even creating entirely computed columns.

Under the hood, views are defined by a database (SQL) query.

Views in Mathesar

We expect that Tables will be used for entering simple data quickly and Views will be used for looking at data, creating reports, or editing inter-related data easily.

You should use Views when you’d like to:

  • see data aggregated across multiple tables in one place.
  • bookmark a subset of data in a table (e.g. filtered, sorted, or grouped data)
  • summarize data (e.g. have a column that shows counts of records in a table)
  • compute data (e.g. subtract the value of one column from another column and show the result in a new column)


Tables that are structured to avoid data duplication (i.e. are properly normalized) are not the most useful for seeing the most relevant information in a single place.

Views vs. Tables

To think about the difference between Views and Tables, here’s an example. Imagine you want to track your movie watches in Mathesar. Questions you might be thinking about are:

  • On what day of the week do I usually watch movies?
  • When did I last see Brad Pitt in a movie?
  • How many movies did I watch that were released in the 90s vs. the 2000s.
  • What was the last foreign movie I watched?

Table Structure

In order to track the data necessary to answer these questions, here’s a possible structure for your tables:

ID Title Release Year Primary Language
13 Thelma & Louise 1991 English
22 Meet Joe Black 1998 English
33 Crouching Tiger, Hidden Dragon 2000 Mandarin
41 Crazy Rich Asians 2018 English
.. .. .. ..
ID Name
34 Michelle Yeoh
45 Brad Pitt
71 Geena Davis
83 Zhang Ziyi
.. ..
Movie Actor Map
ID Movie ID Actor ID
67 13 45
68 22 45
69 33 34
70 41 34
89 13 71
97 33 83
.. .. ..
Movie Watch
ID Movie ID Date
91 13 2021-10-13
100 22 2021-10-01
104 33 2021-09-23
190 41 2021-08-12
203 13 2019-01-20
.. .. ..

The Role of Views

As is obvious, none of these tables answer your questions by themselves, even when filters, sorts, or aggregations are applied to an individual table. Also, when you watch a new movie, logging it would involve adding data to four separate tables. The job of views is to simplify that.

Imagine a view with this structure:

ID Title Actors Last Watched Times Watched Language Release Year
13 Thelma & Louise Brad Pitt, Geena Davis 2021-10-13 2 English 1991
22 Meet Joe Black Brad Pitt 2021-10-01 1 English 1998
33 Crouching Tiger, Hidden Dragon Michelle Yeoh, Zhang Ziyi 2021-09-23 1 Mandarin 2000
41 Crazy Rich Asians Michelle Yeoh 2021-08-12 1 English 2018
.. .. .. ..

Once you have set up this view, you could filter and sort to get answers to your questions, or add new movie watches easily by adding all relevant data from one place.

Future Plans

In the future, we will offer alternate display modes for Views (e.g. calendar, map, kanban, etc.).