Description:

This processor updates the attributes of a FlowFile using properties or rules that are added by the user. There are three ways to use this processor to add or modify attributes. One way is the "Basic Usage"; this allows you to set default attribute changes that affect every FlowFile going through the processor. The second way is the "Advanced Usage"; this allows you to make conditional attribute changes that only affect a FlowFile if it meets certain conditions. It is possible to use both methods in the same processor at the same time. The third way is the "Delete Attributes Expression"; this allows you to provide a regular expression and any attributes with a matching name will be deleted.

Please note that "Delete Attributes Expression" supersedes any updates that occur. If an existing attribute matches the "Delete Attributes Expression", it will be removed whether it was updated or not. That said, the "Delete Attributes Expression" only applies to attributes that exist in the input FlowFile, if it is added by this processor, the "Delete Attributes Expression" will not detect it.

Properties:

The properties in this processor are added by the user. The expression language is supported in user-added properties for this processor. See the NiFi Expression Language Guide to learn how to formulate proper expression language statements to perform the desired functions.

If an Attribute is added with the name alternate.identifier and that attribute's value is a URI, an ADD_INFO Provenance Event will be registered, correlating the FlowFile with the given alternate identifier.

Relationships:

Basic Usage

For basic usage, changes are made by adding a new processor property and referencing as its name the attribute you want to change. Then enter the desired attribute value as the Value. The Value can be as simple as any text string or it can be a NiFi Expression Language statement that specifies how to formulate the value. (See the NiFi Expression Language Usage Guide for details on crafting NiFi Expression Language statements.)

As an example, to alter the standard "filename" attribute so that it has ".txt" appended to the end of it, add a new property and make the property name "filename" (to reference the desired attribute), and as the value, use the NiFi Expression Language statement shown below:

The preceding example illustrates how to modify an existing attribute. If an attribute does not already exist, this processor can also be used to add a new attribute. For example, the following property could be added to create a new attribute called myAttribute that has the value myValue:

In this example, all FlowFiles passing through this processor will receive an additional FlowFile attribute called myAttribute with the value myValue. This type of configuration might be used in a flow where you want to tag every FlowFile with an attribute so that it can be used later in the flow, such as for routing in a RouteOnAttribute processor.

Advanced Usage

The preceding examples illustrate how to make changes to every FlowFile that goes through the processor. However, the UpdateAttribute processor may also be used to make conditional changes.

To change attributes based on some condition, use the Advanced User Interface (UI) in the processor by clicking the Advanced button in the lower right corner.

Clicking the Advanced button displays the Advanced UI. In the Advanced UI, Conditions and their associated Actions are entered as "Rules". Each rule basically says, "If these conditions are met, then do this action." One or more conditions may be used in a given rule, and they all must be met in order for the designated action(s) to be taken.

Adding Rules

To add rules and their associated conditions and actions, click on the buttons with the plus symbol located to the right of the "Rules", "Conditions", and "Actions" labels.

Upon adding a rule with its condition(s) and action(s), it is important to save it by clicking the Save button in the lower right corner. If you do not do so and attempt to add or navigate to another rule, an error message will appear, asking you if you want to save your changes.

Example Rules

This example has two rules: CheckForLargeFiles and CheckForGiantFiles. The CheckForLargeFiles rule has these conditions:

Then it has this action for the filename attribute:

Taken together, this rule says:

Adding another Rule

Continuing with this example, we can add another rule to check for files that are larger than one gigabyte. When we add this second rule, we can use the previous rule as a template, so to speak, by taking advantage of the "Copy from existing rule" option in the New Rule window. Simply start typing the name of an existing rule, and it will show up in a dropdown menu below the entry field.

In this example, the CheckForGiantFiles rule has these conditions:

Then it has this action for the filename attribute:

Taken together, this rule says:

Combining the Basic Usage with the Advanced Usage

The UpdateAttribute processor allows you to make both basic usage changes (i.e., to every FlowFile) and advanced usage changes (i.e., conditional) at the same time; however, if they both affect the same attribute(s), then the conditional changes take precedence. This has the added benefit of supporting a type of "else" construct. In other words, if none of the rules match for the attribute, then the basic usage changes will be made.

Deleting Attributes

Deleting attributes is a simple as providing a regular expression for attribute names to be deleted. This can be a simple regular expression that will match a single attribute or more complex regular expression to match a group of similarly named attributes or even several individual attribute names.

The delete attributes function does not produce a Provenance Event if the alternate.identified Attribute is deleted.

FlowFile Policy

Another setting in the Advanced UI is the FlowFile Policy. It is located in the upper-left corner of the UI, and it defines the processor's behavior when multiple rules match. It may be changed using the dropdown menu. By default, the FlowFile Policy is set to "use clone".

If the FlowFile policy is set to "use clone", and multiple rules match, then a copy of the incoming FlowFile is created, such that the number of outgoing FlowFiles is equal to the number of rules that match. In other words, if two rules (A and B) both match, then there will be two outgoing FlowFiles, one for Rule A and one for Rule B. This can be useful in situations where you want to add an attribute to use as a flag for routing later. In this example, there will be two copies of the file available, one to route for the A path, and one to route for the B path.

If the FlowFile policy is set to "use original", then all matching rules are applied to the same incoming FlowFile, and there is only one outgoing FlowFile with all the attribute changes applied. In this case, the order of the rules matters and the action for each rule that matches will be applied in that order. If multiple rules contain actions that update the same attribute, the action from the last matching rule will take precedence. Notably, you can drag and drop the rules into a certain order within the Rules list.

Filtering Rules

The Advanced UI supports the creation of an arbitrarily large number of rules. In order to manage large rule sets, the listing of rules may be filtered using the Filter mechanism in the lower left corner. Rules may be filtered by any text in the name, condition, or action.

Closing the Advanced UI

Once all changes have been saved in the Advanced UI, the UI can be closed using the X in the top right corner.

Stateful Usage

By selecting "store state locally" option for the "Store State" property UpdateAttribute will not only store the evaluated properties as attributes of the FlowFile but also as stateful variables to be referenced in a recursive fashion. This enables the processor to calculate things like the sum or count of incoming FlowFiles. A dynamic property can be referenced as a stateful variable like so:

This example will keep a count of the total number of FlowFiles that have passed through the processor. To use logic on top of State, simply use the "Advanced Usage" of UpdateAttribute. All Actions will be stored as stateful attributes as well as being added to FlowFiles. Using the "Advanced Usage" it is possible to keep track of things like a maximum value of the flow so far. This would be done by having a condition of "${getStateValue("maxValue"):lt(${value})}" and an action of attribute:"maxValue", value:"${value}". The "Stateful Variables Initial Value" property is used to initialize the stateful variables and is required to be set if running statefully. Some logic rules will require a very high initial value, like using the Advanced rules to determine the minimum value. If stateful properties reference other stateful properties then the value for the other stateful properties will be an iteration behind. For example, attempting to calculate the average of the incoming stream requires the sum and count. If all three properties are set in the same UpdateAttribute (like below) then the Average will always not include the most recent values of count and sum:

Instead, since average only relies on theCount and theSum attributes (which are added to the FlowFile as well) there should be a following Stateless UpdateAttribute which properly calculates the average. In the event that the processor is unable to get the state at the beginning of the onTrigger, the FlowFile will be pushed back to the originating relationship and the processor will yield. If the processor is able to get the state at the beginning of the onTrigger but unable to set the state after adding attributes to the FlowFile, the FlowFile will be transferred to "set state fail". This is normally due to the state not being the most up to date version (another thread has replaced the state with another version). In most use-cases this relationship should loop back to the processor since the only affected attributes will be overwritten. Note: Currently the only "stateful" option is to store state locally. This is done because the current implementation of clustered state relies on Zookeeper and Zookeeper isn't designed for the type of load/throughput UpdateAttribute with state would demand. In the future, if/when multiple different clustered state options are added, UpdateAttribute will be updated.

Combining the Advanced Usage with Stateful

The UpdateAttribute processor allows you to use both advanced usage changes (i.e., conditional) in addition to storing the values in state at the same time. This allows UpdateAttribute to act as a stateful rules engine to enable powerful concepts such as a Finite-State machine or keeping track of a min/max value. Working with both is relatively simple, when the processor would normally update an attribute on the processor (ie. it matches a conditional rule) the same update is stored to state. Referencing state via the advanced tab is done in the same way too, using "getStateValue". Note: In the event the "use clone" policy is set and the state is failed to set, no clones will be generated and only the original FlowFile will be transferred to "set state fail".

Notes about Concurrency and Stateful Usage

When using the stateful option, concurrent tasks should be used with caution. If every incoming FlowFile will update state then it will be much more efficient to have only one task. This is because the first thing the onTrigger does is get the state and the last thing it does is store the state if there are an updates. If it does not have the most up to date initial state when it goes to update it will fail and send the FlowFile to "set state fail". This is done so that the update is successful when it was done with the most recent information. If it didn't do it in this mock-atomic way, there'd be no guarantee that the state is accurate. When considering Concurrency, the use-cases generally fall into one of three categories:

The first and last cases are relatively clear-cut in their guidance. For the first, concurrency should not be used. Doing so will just waste CPU and any benefits of concurrency will be wiped due to misses in state. For the last case, it can easily be done using concurrency. Since updates are rare in the first place it will be even more rare that two updates are processed at the same time that cause problems. The second case is a bit of a grey area. If updates are rare then concurrency can probably be used. If updates are frequent then concurrency would probably cause more problems than benefits. Regardless, testing to determine the appropriate tuning is the only true answer.