@romani I have some custom checks for JavaDocs and, I'm afraid, it's not quite clear what exactly is supposed to be used instead of getFileContents(). Could you (or someone else) explain what is supposed to be used?

My goal is it check that JavaDoc is presented on every element that matches specific condition, has some tag (@author) and the tag's value is equal to some constant.

@remal instead of getContents() checks should use only AST and do not rely on text in file.
About the case you described - it seems to me that you dont need a custom check for it, MatchXPathCheck can provide same functionality.

@remal , please look at

With such inheritance, you will get whole ast of javadoc content.
The only downside is performance degradation, but you can measure how critical it for you in your projects.

Another approach is to override

I am on JavadocStyleCheck

@aayushRedHat , please be aware that refactoring of javadoc check needs changing extension of abstract class to get AST of javadoc. It might be bigger update than for non javadoc Checks. Do not hesitate to set draft PR to share with us your direction and double check that you care right path.

I am on ImportControlCheck

@Kevin222004 , please do not deal with Javadoc* Checks, it will require big redesign we do not have plan for this, please deal with regular Checks for now.

@romani I want to work on this. I can take EmptyLineSeparatorCheck. Is there anything I should keep in mind before starting? and
is it ok to split work for this Check in separate PRs to ease work and review because there are two usages of getFileContents() in this check or full migration should be done in one PR?

@mahfouz72, please help us to refactor this Check, yes, few PRs are good to ease review.