Structure of a NiFi Expression
The NiFi Expression Language always begins with the start delimiter ${
and ends with the end delimiter }
. Between the start and end delimiters is the text of the Expression itself. In its most basic form, the Expression can consist of just an attribute name. For example, ${filename}
will return the value of the filename
attribute.
In a slightly more complex example, we can instead return a manipulation of this value. We
can, for example, return an all upper-case version of the filename by calling the
toUpper
function: ${filename:toUpper()}
. In this case,
we reference the filename
attribute and then manipulate this value by
using the toUpper
function. A function call consists of 5 elements. First,
there is a function call delimiter :
. Second is the name of the function -
in this case, toUpper
. Next is an open parenthesis ((
),
followed by the function arguments. The arguments necessary are dependent upon which
function is being called. In this example, we are using the toUpper
function, which does not have any arguments, so this element is omitted. Finally, the
closing parenthesis ()
) indicates the end of the function call. There are
many different functions that are supported by the Expression Language to achieve many
different goals. Some functions provide String (text) manipulation, such as the
toUpper
function. Others, such as the equals
and
matches
functions, provide comparison functionality. Functions also
exist for manipulating dates and times and for performing mathematical operations. Each of
these functions is described below, in the Functions section, with an
explanation of what the function does, the arguments that it requires, and the type of
information that it returns.
When we perform a function call on an attribute, as above, we refer to the attribute as the subject of the function, as the attribute is the entity on which the function is operating. We can then chain together multiple function calls, where the return value of the first function becomes the subject of the second function and its return value becomes the subject of the third function and so on. Continuing with our example, we can chain together multiple functions by using the expression ${filename:toUpper():equals('HELLO.TXT')}
. There is no limit to the number of functions that can be chained together.
Any FlowFile attribute can be referenced using the Expression Language. However, if the attribute name contains a "special character", the attribute name must be escaped by quoting it. The following characters are each considered "special characters":
-
$ (dollar sign)
-
| (pipe)
-
{ (open brace)
-
} (close brace)
-
( (open parenthesis)
-
) (close parenthesis)
-
[ (open bracket)
-
] (close bracket)
-
, (comma)
-
: (colon)
-
; (semicolon)
-
/ (forward slash)
-
* (asterisk)
-
' (single quote)
-
(space)
-
\t (tab)
-
\r (carriage return)
-
\n (new-line)
Additionally, a number is considered a "special character" if it is the first character of the attribute name. If any of these special characters is present in an attribute is quoted by using either single or double quotes. The Expression Language allows single quotes and double quotes to be used interchangeably. For example, the following can be used to escape an attribute named my attribute
: ${"my attribute"}
or ${'my attribute'}
.
In this example, the value to be returned is the value of the "my attribute" attribute, if it exists. The Expression Language will search through a hierarchy for a matching property. See Expression Language Hierarchy for a description of the hierarchy.
There also exist some functions that expect to have no subject. These functions are invoked
simply by calling the function at the beginning of the Expression, such as
${hostname()}
. These functions can then be changed together, as well.
For example, ${hostname():toUpper()}
. Attempting to evaluate the function
with subject will result in an error. In the Functions section below, these
functions will clearly indicate in their descriptions that they do not require a
subject.
Often times, we will need to compare the values of two different attributes to each other. We are able to accomplish this by using embedded Expressions. We can, for example, check if the filename
attribute is the same as the uuid
attribute: ${filename:equals( ${uuid} )}
. Notice here, also, that we have a space between the opening parenthesis for the equals
method and the embedded Expression. This is not necessary and does not affect how the Expression is evaluated in any way. Rather, it is intended to make the Expression easier to read. White space is ignored by the Expression Language between delimiters. Therefore, we can use the Expression ${ filename : equals(${ uuid}) }
or ${filename:equals(${uuid})}
and both Expressions mean the same thing. We cannot, however, use ${file name:equals(${uuid})}
, because this results in file
and name
being interpreted as different tokens, rather than a single token, filename
.