Infohub (Flow)How-to

T06: Format Strings

Control how every value displays, and use the three-section pattern to turn a numeric measure into Pass or Fail.

Updated 15 Oct 2026Infohub Docs

What you will build: correct formats across your model, and a status widget that displays Pass or Fail from a numeric measure.

Time: about 20 minutes.

You need: T05 completed.

Step 1 — See the problem#

Find a measure with no format set and look at it on a chart or table. You will see full floating-point precision — something like 3.4179999828338623.

That is what an unset format looks like, and it appears everywhere the measure does: charts, tables, widgets, forms, reports, notifications.

Step 2 — The formats you will actually use#

Infohub passes the format string straight to .NET's number formatting, so the whole standard specification is available.

Set these on your existing measures:

Measure	Format	Renders `1234.567` as
`Steam Flow Hourly`	`0.0`	`1234.6`
`Steam Flow Daily`	`0.0`	`1234.6`
`Water Used Daily`	`#,##0`	`1,235`
`Bottles Daily`	`#,##0`	`1,235`
`Energy per 1000 Bottles`	`0.00`	`1234.57`
`Production vs Target Daily`	`0.0 '%'`	`1234.6 %`

Step 3 — `0` and `#` are different#

This is the distinction that matters for tables.

0 forces a digit. 0.00 renders 5 as 5.00.
# omits it when not needed. #.## renders 5 as 5.

Put 0.0 and #.# on two measures and look at them side by side in a table. The 0.0 column has aligned decimal points and can be scanned down; the #.# column is ragged.

Prefer 0 for operational reporting. Use # only where a trailing zero would actively mislead.

Step 4 — The percentage trap#

Try 0.0% on Production vs Target Daily.

It displays something like 9420.0%.

Important

% multiplies by 100. In a .NET format string, % is not decoration — it scales the number. Your measure holds 94.2 meaning 94.2%, so % makes it 9420.

Two consistent options:

Store the percentage (94.2) and format with a quoted literal: 0.0 '%'
Store the ratio (0.942) and format with 0.0%

Pick one convention for the whole model. Mixing them is a reliable source of wrong dashboards.

Set it back to 0.0 '%'.

Step 5 — The three-section pattern#

A format string can carry three sections separated by semicolons:

positive ; negative ; zero

Ordinary use, accounting style with a dash for zero:

#,##0.0;(#,##0.0);—

1234.5 renders 1,234.5, -1234.5 renders (1,234.5), and 0 renders as an em dash so empty table cells do not clutter.

Step 6 — Now the useful trick#

Because a section can be a quoted literal, a numeric measure can display as text.

Create a quality check. In your Quality folder, add a calculated measure:

Field	Value
Name	`Brix In Spec`
Retrieval type	Calculated
Interval	Day(s) — or Shift(s)
Unit	`ratio`
Dependencies	`Brix`

Expression — returns 1 if in specification, 0 if not:

(Measure_<brixId>.Value ?? 0) >= 10.5 && (Measure_<brixId>.Value ?? 0) <= 12.5 ? 1 : 0

Now set the format to:

'Pass';'Fail';'Fail'

Put it on a Widget. It displays Pass or Fail — no enumeration, no special chart type, no extra configuration.

Add a limit to the measure so the widget also colours itself green or red, and you have a complete status indicator built from a calculation, a format string, and a limit.

Other useful variants:

Format	Positive	Negative	Zero
`'Running';'Fault';'Stopped'`	Running	Fault	Stopped
`'Yes';'No';'No'`	Yes	No	No
`'Above';'Below';'On target'`	Above	Below	On target

Step 7 — Variance with a sign#

For a measure that can go either way, show the sign explicitly.

Add a variance calculation in Efficiency:

Field	Value
Name	`Production Variance Daily`
Interval	Day(s)
Unit	`bottles`
Format	`+#,##0;-#,##0;0`

(Measure_<bottlesDailyId>.Value ?? 0) - (Measure_<targetDailyId>.Value ?? 0)

It renders +12,500, -12,500, or 0. On a variance column that is much clearer than relying on the reader to spot a minus sign.

Step 8 — Scaling in the format#

A trailing comma divides by a thousand:

Format	`1234567` renders as
`#,##0`	`1,234,567`
`#,##0,`	`1,235`
`#,##0,,`	`1`

Useful for large numbers on a small widget — and dangerous.

Important

If you scale, change the unit. Displaying kWh divided by a thousand while the measure is still labelled kWh is exactly how a report ends up wrong by three orders of magnitude. If you scale in the format, say so in the unit: MWh, or kWh (thousands).

Step 9 — Test the awkward cases#

For each format you have set, check five values rather than one:

A normal value.
Zero.
A negative value, if the measure can go negative.
A very large value — does the separator appear where you expect?
A very small value — does it round to zero and look broken?

Formats are cheap to change and easy to get subtly wrong. Thirty seconds of checking is worth it.

What you have learned#

Formats are .NET format strings. The whole specification is available.

0 forces digits, # omits them. Use 0 for aligned columns.

% multiplies by 100. Use 0.0 '%' for a value already in percent.

Three sections give you positive, negative, and zero — and a quoted literal turns a number into a word.

A trailing comma scales by a thousand. Change the unit if you use it.

Formatting belongs on the measure, not in the expression. A calculation must return a double?; formatting there breaks the contract and defeats the measure's own format.

Try this before moving on#

Audit your whole model's formats. Every measure should have one, and the set should be internally consistent — the same quantity formatted the same way everywhere.

While you are there, check the units too. Infohub stores the unit as free text and validates nothing, so m3 and m³, or t and ton, are different strings to every grouping you will ever run. Pick one spelling per quantity and hold to it.

Next: T07: Event Schemes — capturing what was happening, not just what the value was.