Conditional Writing in Alteryx

Do you know the situation? You create a table or a file with Alteryx, which is then used by other users. Master data maybe. Anyway, data that is needed desperately.?

So far, nothing special. But there is a small problem. In your workflow you process a number of files or tables you can' t always trust.?Sometimes one of the files is completely empty, sometimes the content of a field is incomplete or wrong. And when that happens, your result is also empty or wrong.

This does not happen to you? Lucky you.?

But if something like this happens to you sometimes, or if you can imagine that it could possibly happen - wouldn't it be better in this case if at least the last correct state of your table is preserved, i.e. you don't overwrite everything??

This is the point at which it would be helpful if you could put some kind of validation before the Output Data Tool, and only if this validation is passed (i.e. the data records are okay), it is actually written.?For simplicity, let's check here if there is any result data at all, but of course this can be extended as you like.

Maybe the Condition tool can help here. Not seen yet? We'll change that now.

However, you'll find this tool within the "Interface" palette - it seems we'll have to build a macro or an analytic app. For our use case, a macro seems to be the appropriate way to "wrap" the output data tool and to add the condition we want.

So - here we go. We open a new workflow and drag the Condition Tool onto the canvas.?The tool has a questions anchor, here we can "feed" in our case the number of result records that passed the validation.?

We can easily supply this number (or any other value to be checked) via a control parameter tool. We only have to give the tool a label (i.e. a name for the value that is to be delivered). Of course, it makes sense to name the parameter "Number".

The "input side" is done with it. On the "output side" there is a "T" and an "F" anchor, which are controlled depending on the condition we can set in the Condition tool.?In our use case, we check whether the delivered value (i.e. the number of records) is greater than 0. The parameter is passed as a variable "Connections from Questions". These variables are numbered - we have only one variable, i.e. "#1".?Passed is a string, so we need a conversion to a numeric value to perform the comparison "> 0".

Parameter transfer and parameter check are done with this, before we now take care of the actions that are linked to "T" and "F", let's have a look at another data transfer to our macro, because in addition to the value to be checked, the data that is to be written to the output file (if the workflow delivers data records) must of course also be transferred.

This handover can easily be done via a Macro Input tool. This tool is then connected to the Output Data tool, which is then used to write the data to the result file. The Macro Input tool gets the input name "Data" and the abbreviation (Anchor Abbreviation) "D". I'm perhaps a little bit ungenerous with the naming of Macro Input and Macro Output tools and the abbreviations for Input and Output anchors - let's say I learned from bad experiences.?

The Output Data tool must not always be active, but only when the condition is met, so it must be possible to switch it on and off. For this we use a little trick - we put the Output Data tool into a Tool Container.

In fact, the Tool Container has a switch by which we can enable or disable it (and all included tools). If there are no result data records, the tool container is simply deactivated, in the other case it should remain active. And this is exactly what we solve now via two Action tools, which we switch between the "T"/"F" outputs of the Condition tool and the Tool Container.

For both Action tools we select the Action Type "Enable/Disable Container from Condition". For the Action tool connected to the "T" output, the property is "Enabled", but for the "F" output it is "Disabled" - so if the condition is met, the Tool Container is enabled, but if the condition is not met, it is disabled and the Output Data tool does not write to the result file.?

In the overview our macro looks like this:

By the way, the Control Parameter Tool automatically sets the Workflow Type to "Batch Macro" (even though this isn't related to batch at all). We don't need any further configuration, but we could of course choose a custom icon for our macro using the Interface Designer.?Anyway we save our macro now, we can name it for example "ConditionalOutput", the file type is "yxmc" for a macro.?

All we have to do now is to add the macro to our workflow.?First, we add a Count Records tool to count the records. We can remove the existing Output Data Tool and add our macro instead.?The "D" Input Anchor we connect to the Summarize Tool (that's where the records come from that will be written if there are any).

The Question Anchor we connect to the Count Records Tool, which provides the condition to be checked. In the configuration we have to specify that Count (the output of the Count Records tool) returns the parameter Number, there is no "Group By" field here.

Time for a quick test. What happens if there is no result data?

There is no more message that a file is being written to - the existing file/table remains intact, the macro works as we expected.

So far, the macro is still very specific to our specific use case, but of course it can be easily extended - for example, we could pass the name and type of the file as additional parameters. Also a more flexible control of the condition can be easily implemented.??