Rounds off the coordinates of the feature to the specified number of decimal places.

Jump to Configuration

Typical Uses

  • Generalizing features

  • Altering coordinates to match a schema or dataset

  • Rounding coordinates where precision exceeds accuracy

How does it work?

The CoordinateRounder receives features and rounds their x, y, and/or z coordinates to a specified precision.

X, Y, and Z Precision parameters are specified as integers, and may be entered directly or set to an expression, user parameter, attribute, or conditional value. Positive numbers represent the number of decimal places to round to, and negative numbers produce rounding to multiples of ten (10). A precision of zero (0) produces integers - coordinates with no decimal places.

If any of the Precision parameters are left blank, those coordinates will pass through unaltered.

For arcs, only the start and end points are rounded, as these are the only coordinates which lie on the path of the arc. The center point is not rounded, nor are any other of the arc's parameters.

Rasters are passed through unaltered.

Rounding Errors and Floating-Point Processing

The CoordinateRounder may not always behave as expected, due to known limitations in floating-point computing.

Representing decimal (base 10) numbers in binary (base 2) can result in infinitely repeating values - much like 1 divided by 3 (that is, ⅓) results in 0.3333 (repeating to infinity) in the decimal number system.

For example, 1 divided by 5 (that is, ⅕), which produces 0.2 in the decimal system, recurs infinitely when represented as a binary value:

0.0111 1111 1001 0011 0011 0011 0011 0011... (0011 repeating)

When this value is converted back to a decimal, it produces 0.200000000000000011102230246252.

As a result, rounding can sometimes be unpredictable, rounding in an unexpected direction.

Some numbers like 1.1 cannot be stored precisely in binary and so end up as 1.1000000000000001, regardless of rounding. Rounding to 3 decimal places also results in 1.1000000000000001. Similarly, a value of 3.3333 is stored as 3.3332999999999999, and when rounded to 2 decimal places ends up as 3.3300000000000002.

Note that the same behavior may not be seen when values are handled as strings, as in the AttributeRounder, but they will be affected whenever the value is used as floating point.

Coordinates in FME are typically logged with full floating-point precision, including any floating-point processing precision error. Note that many formats have a way of controlling decimal precision based on data types or precision as set on the writer.


Usage Notes


Input Ports

Output Ports


Editing Transformer Parameters

Using a set of menu options, transformer parameters can be assigned by referencing other elements in the workspace. More advanced functions, such as an advanced editor and an arithmetic editor, are also available in some transformers. To access a menu of these options, click beside the applicable parameter. For more information, see Transformer Parameter Menu Options.

Defining Values

There are several ways to define a value for use in a Transformer. The simplest is to simply type in a value or string, which can include functions of various types such as attribute references, math and string functions, and workspace parameters. There are a number of tools and shortcuts that can assist in constructing values, generally available from the drop-down context menu adjacent to the value field.

Dialog Options - Tables

Transformers with table-style parameters have additional tools for populating and manipulating values.


Processing Behavior


Feature Holding


Dependencies None

FME Community

The FME Community is the place for demos, how-tos, articles, FAQs, and more. Get answers to your questions, learn from other users, and suggest, vote, and comment on new features.

Search for all results about the CoordinateRounder on the FME Community.


Examples may contain information licensed under the Open Government Licence – Vancouver and/or the Open Government Licence – Canada.