Simplified usage of class AssociableToAST

[4everTheOne]

According to the documentation in AssociableToAST#toAst() all the Resolved declaration most be asociable to a AST.

Currently this doesn't happen, making it hard do access a node from a ResolvedDeclaration. To work around this we need to make some check that requires a lot of boilerplate like in the example below.

javaparser/javaparser-symbol-solver-core/src/main/java/com/github/javaparser/symbolsolver/javaparsermodel/contexts/CompilationUnitContext.java

Lines 353 to 366 in 1ce7d85

	private boolean isAncestorOf(ResolvedTypeDeclaration descendant) {
	if (descendant instanceof AssociableToAST) {
	Optional<Node> astOpt = ((AssociableToAST<Node>) descendant).toAst();
	if (astOpt.isPresent()) {
	return wrappedNode.isAncestorOf(astOpt.get());
	} else {
	return false;
	}
	} else if (descendant instanceof JavaParserEnumDeclaration) {
	return wrappedNode.isAncestorOf(((JavaParserEnumDeclaration) descendant).getWrappedNode());
	} else {
	throw new UnsupportedOperationException();
	}
	}

This PR suggest the remove of the type parameter of the class AssociableToAST and use a helper method to convert to the desired type.
To make it more clear lets take a look at the example mentioned above but this time in the new format:

private boolean isAncestorOf(ResolvedTypeDeclaration descendant) {
    return descendant.toAst()
            .filter(node -> wrappedNode.isAncestorOf(node))
            .isPresent();
}

Other example, previously we would have to do something like this:

ResolvedDeclaration resolvedDeclaration = ...;
if (resolvedDeclaration instanceof ResolvedFieldDeclaration) {
    ResolvedFieldDeclaration resolvedField = (ResolvedFieldDeclaration) resolvedDeclaration;
    Optional<FieldDeclaration> fieldDeclaration = resolvedField.toAst();
}

Now we can do it by simple doing:

ResolvedDeclaration resolvedDeclaration = ...;
Optional<FieldDeclaration> fieldDeclaration = resolvedDeclaration.toNode(FieldDeclaration.class);

This also allows add room for more complex examples that were not possible with the previous format, like looking for specific declarations. We can do it now by simply:

Optional<VariableDeclarator> variableDeclarator = resolvedDeclaration.toNode(VariableDeclarator.class);

[codecov]

Codecov Report

Merging #3063 (5eaa434) into master (24925df) will decrease coverage by 0.027%.
The diff coverage is 42.307%.

❗ Current head 5eaa434 differs from pull request most recent head bbcf0cd. Consider uploading reports for the commit bbcf0cd to get more accurate results

@@               Coverage Diff               @@
##              master     #3063       +/-   ##
===============================================
- Coverage     57.443%   57.415%   -0.028%     
- Complexity      2541      2555       +14     
===============================================
  Files            636       637        +1     
  Lines          33809     33853       +44     
  Branches        5841      5839        -2     
===============================================
+ Hits           19421     19437       +16     
- Misses         12307     12336       +29     
+ Partials        2081      2080        -1     

Flag	Coverage Δ
AlsoSlowTests	57.415% <42.307%> (-0.028%)	⬇️
javaparser-core	51.944% <0.000%> (-0.009%)	⬇️
javaparser-symbol-solver	36.646% <42.307%> (-0.001%)	⬇️
jdk-10	57.411% <42.307%> (-0.019%)	⬇️
jdk-11	57.411% <42.307%> (-0.028%)	⬇️
jdk-12	57.411% <42.307%> (-0.028%)	⬇️
jdk-13	57.411% <42.307%> (-0.028%)	⬇️
jdk-14	57.411% <42.307%> (-0.028%)	⬇️
jdk-15	57.405% <42.307%> (-0.031%)	⬇️
jdk-16	57.411% <42.307%> (-0.028%)	⬇️
jdk-8	57.413% <42.307%> (-0.028%)	⬇️
jdk-9	57.405% <42.307%> (-0.028%)	⬇️
macos-latest	57.404% <42.307%> (-0.024%)	⬇️
ubuntu-latest	57.401% <42.307%> (-0.028%)	⬇️
windows-latest	57.395% <42.307%> (-0.028%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...arser/resolution/declarations/AssociableToAST.java	0.000% <ø> (ø)
...r/resolution/declarations/ResolvedDeclaration.java	100.000% <ø> (ø)
...declarations/ResolvedTypeParameterDeclaration.java	25.000% <0.000%> (-0.397%)	⬇️
...thub/javaparser/symbolsolver/JavaSymbolSolver.java	58.088% <0.000%> (-0.431%)	⬇️
...el/declarations/DefaultConstructorDeclaration.java	46.153% <0.000%> (-3.847%)	⬇️
.../declarations/JavaParserAnnotationDeclaration.java	73.529% <0.000%> (-2.229%)	⬇️
...rations/JavaParserAnnotationMemberDeclaration.java	80.000% <0.000%> (-8.889%)	⬇️
...larations/JavaParserAnonymousClassDeclaration.java	75.824% <0.000%> (-0.843%)	⬇️
...declarations/JavaParserConstructorDeclaration.java	50.000% <0.000%> (-2.381%)	⬇️
...eclarations/JavaParserEnumConstantDeclaration.java	87.500% <0.000%> (-12.500%)	⬇️
... and 40 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 24925df...bbcf0cd. Read the comment docs.

[4everTheOne]

After the conversation with jlerbsc at #1940 (comment) that demonstrated the impact of this changes, decided to update this PR to reduce this impact. To reduce this impact a new AssociableToAst was created with a new name AssociatedWithAST and behavior exactly as described in the first comment.

Also AssociableToAst has been marked as Deprecated for not complying with the expected documentation.

[jlerbsc]

It would be possible to have a clearer PR by deleting all unnecessary commits.

[jlerbsc]

After the conversation with jlerbsc at #1940 (comment) that demonstrated the impact of this changes, decided to update this PR to reduce this impact. To reduce this impact a new AssociableToAst was created with a new name AssociatedWithAST and behavior exactly as described in the first comment.

Also AssociableToAst has been marked as Deprecated for not complying with the expected documentation.

Given that the next version of JP will introduce a break in the API (this will be a major version certainly 4.0.0.), I send back my opinion, so that you can integrate this change directly into the existing interface.

[jlerbsc]

Did you complete your PR?

[4everTheOne]

Yes I did. Sorry I forgot to leave a comment.

[jlerbsc]

Thank you for this contribution.