Transformations

Transformations can be used to modify any of the extracted values according to the desired output. All these transformations are done through natural language.

To initiate a transformation, hover over a data field from the data mapper and select the double arrow symbol . Then, you can use the text field to add the instruction for the transformation to be performed.

After you add the text instruction, to apply the transformation either press the Enter key or click outside the text box and you can see the transformation taking shape immediately.

However, if the text instruction is not clear enough, instead of a new value, the text Empty is displayed to indicate that the transformation did not happen and the data field is not going to be populated with a value in the destination document.

There is a wide range of transformations you can perform, as follows:

• converting one date format to another by specifying both the original date format and the desired date format; e.g. MM-DD-YY to DD/MM/YYYY
• converting a unit of measurement to another by specifying both the original unit of measurement and the desired unit of measurement; e.g. meters to miles
• converting a currency to another by specifying both the original currency and the desired currency; e.g. USD to EUR
• asking questions to obtain a certain output; e.g. Is this address is Ohio?
• translating text from one language to another
• extracting only the subset of a certain value
• asking for a summary
• any other text changes, like spelling a word differently (from lower case to upper case) or abbreviating a word (for instance, developer to dev) etc.

As you would expect, only the new value, the one derived from the transformation, is going to be pasted in the destination document. For example, when asking a question, only the answer of the question is pasted further, so if you have a question like 'Is this address is Ohio?' and the answer is Yes, only the value Yes is populated in the destination document.

After the transformation is applied, you can:

• edit the transformation by hovering over the data field and selecting the pencil symbol ; this opens again the text field and you add another instruction for the transformation to be performed
• delete the transformation by hovering over the data field and selecting the trash bin symbol ; this deletes the transformation and the original extracted value is restored