Column Filters

Table of contents

Overview

Column filters narrow the displayed results by conditions on individual columns. Filters apply to the data already loaded in the results grid — they don’t re-run the SQL or fetch new data. The filter popover is value-picker-first: check the values you want, or expand the advanced section for operator-based conditions.

Opening a Filter

Hover over a column header and click the funnel icon at the right edge of the type row. Columns with an active filter show a filled, accent-colored funnel at all times. Filtered results are also reflected in the action-bar status text (“V of T rows … • K filters”).

The Value Picker

The popover opens with a checklist of the column’s distinct values:

  • Search — the field at the top narrows which values are listed (it doesn’t change what’s checked).
  • (Select All) — a tri-state checkbox that checks or unchecks all currently listed values.
  • (Blanks) — NULLs and empty values are collapsed into a single “(Blanks)” row.
  • Counts — each value shows how many rows have it. When other columns’ filters are active, counts show filtered/total, cascading the way spreadsheet filters do.
  • Sort — the Value/Count control in the header orders the list by value (ascending by default) or by count (descending by default); click the active segment again to flip direction.
  • Resize — drag the grip in the bottom-right corner to enlarge the popover.

Click Apply to keep only rows matching the checked values (checking everything clears the filter). Clear resets the column’s filter. When more rows remain unloaded on the server, a footer notes that counts cover only the loaded rows.

Advanced Text Filter

Expand Advanced text filter at the bottom of the popover for operator-based conditions. Applying an advanced filter takes precedence over the checklist.

Operators by Type

String (text, varchar, char, uuid, inet, cidr, …): contains, does not contain, contains any of, does not contain any of, starts with, ends with, equals, does not equal, is null, is not null.

Numeric (integer, bigint, numeric, double precision, money, …): equals, does not equal, less than, less than or equal, greater than, greater than or equal, between, contains any of, is null, is not null.

Boolean: is true, is false, is null, is not null.

Temporal (date, time, timestamp, timestamptz, interval, …): equals, less than, less than or equal, greater than, greater than or equal, between, is null, is not null.

JSON and Array: contains, equals, contains any of, does not contain any of, is null, is not null.

Input Controls

The value input adapts to the column type and operator:

  • Text / numeric — a text field; press Return to apply
  • Date — a calendar date picker
  • Timestamp — a calendar picker plus an HH:MM:SS time field
  • Time — a stepper-style time picker
  • Interval — separate day/hour/minute/second fields, compared by total seconds
  • “any of” operators — a token field; comma or Return adds each token
  • Between — two fields joined by “and”; both bounds inclusive

Filter Behavior

  • Filters are client-side, on the loaded result set. To filter at the database level, add a WHERE clause (or use a chart drill-down in server mode).
  • Filters on different columns combine with AND.
  • Text comparisons are case-insensitive; between is inclusive; temporal values compare chronologically; intervals compare by total seconds.
  • NULLs never match value-based operators — use is null / is not null, or the “(Blanks)” checkbox.
  • Committing a chart selection in client mode applies the same kind of column filters; they show up with a “Filtered by chart” chip and can be cleared from there.
  • Reset Filters in the action bar clears all column filters at once.

Pharos — A native macOS PostgreSQL client built with Swift and Rust.

This site uses Just the Docs, a documentation theme for Jekyll.