How to utilize existing javadoc parsing in a custom language plugin using grammar kit

[budjb]

Following the documentation and tutorial around creating a language plugin for Jetbrains IDE's, it's unclear how one might utilize the existing JavaDoc parser as a sub-parser while parsing a custom grammar. I've defined the grammar elements to parse a JavaDoc-style comment block into a token, and I've tried mixing in or extending the element for it with a custom ASTWrapperPsiElement class where I can invoke the JavaDoc parser. However, the getChildren function on that class is never called, and I'm not sure how else to wire that parser in based on the generated output of the lexer and parser.

Is this possible, and are there docs that cover this?

[maxmedvedev]

I don't think there is any doc on the matter at the moment.

What you need is:

1. mark your doc comment rule as extraRoot = true in your bnf file.

2. make a custom element type for the doc-comment with ILazyParseableElementType.
   See groovy doc as an example:
   https://github.com/JetBrains/intellij-community/blob/master/plugins/groovy/groovy-psi/src/org/jetbrains/plugins/groovy/lang/groovydoc/parser/GroovyDocElementTypes.java
   You can use elementTypeFactory rule attribute to let know GK about custom element type.

3. The corresponding psi implementation must extend LazyParseablePsiElement.
   See https://github.com/JetBrains/intellij-community/blob/master/plugins/groovy/groovy-psi/src/org/jetbrains/plugins/groovy/lang/groovydoc/psi/impl/GrDocCommentImpl.java

Let me know if it does not work