B - Advanced 2 - Handling erroneous code
Table of contents
Purpose
Purpose of this section is to introduce the feedback Studio returns when OCL code is invalid. And to realise that it is always a good thing when there is a build error in the code, as this prevents invalid validation rules being visible in normal validation users are able to access.
Description
Following steps are done here:
Instructions
self.Id.matches("Id1") or self.Id.matches("Id2")
We can intentionally break this by breaking the syntax.
Lets change the code by removing the last bracket
self.Id.matches("Id1") or self.Id.matches("Id2"
When rebuilding the project, we should be receiving an error. To access the error message, we'll have to find the rule which has been highlighted with red and to open it. Technical error message is displayed there, which will indicate what has happened and why the compilation was not successful. In this case the actual error message is:
These may sound technical and confusing at start, but it usually helps to navigate to the position indicated in the message and paying attention to the characters stated to be erronoeus. Here character ( is stated, which should help us to realise that the ending bracket is missing.
We should also investigate other erroneous cases. At least following OCL and their corresponding error message should be checked:
self.Id.matches("Id1") or self.Id.matches("Id2) self.Id.matches("Id1") self.Id.matches("Id2") self.Id.matches("Id1") orself.Id.matches("Id2") self.Id.matches("Id1") or self.matches("Id2") self.Id.matches("Id1") nor self.Id.matches("Id2")
We can also change the context to be something else instead of TransactionType1. Example of another context in the schema is: PartyIdentification