-
-
Notifications
You must be signed in to change notification settings - Fork 1.4k
-
Hi,
To improve the consistency of the codebase, I propose standardizing the parsing of identifiers in JSqlParserCC.jjt. Currently, there are multiple methods used to parse identifiers, including direct usage of S_IDENTIFIER, using RelObjectName(), and allowing various options with OR conditions. This inconsistency can lead to maintenance challenges and potential parsing errors.
By unifying the parsing method to consistently use a single approach, whether it is RelObjectName() or another method, we can achieve the following benefits:
- Improved Code Readability: A single, consistent method for parsing identifiers will make the code easier to read and understand.
- Reduced Maintenance Effort: With a unified approach, future changes to identifier parsing will only need to be made in one place.
I recommend refactoring all instances where identifiers are parsed to use a single, consistent method. This change will streamline the code and improve overall code quality.
I would appreciate your feedback on this proposal and any additional suggestions you might have.
Beta Was this translation helpful? Give feedback.
All reactions
-
❤️ 1
Replies: 1 comment
-
I totally agree with you though this will be a lot of work.
In my opinion, we should aim for one single production ObjectIdentifier(int flag)
where the flag indicates certain constraints, e.g.
- ObjectIdentifier is a Label (because the previous token was an identifier or an
AS
token) - if
VALUES
is allowed or not (when its an INSERT)
So the flag
should be a combination of things and then be used in a Semantic Lookahead.
Beta Was this translation helpful? Give feedback.
All reactions
-
👍 1