RFR: JDK-8075778: Add javadoc tag to avoid duplication of return information in simple situations. [v6]

Pavel Rappo prappo at openjdk.java.net
Tue Dec 8 22:42:36 UTC 2020

On Tue, 8 Dec 2020 19:27:32 GMT, Jonathan Gibbons <jjg at openjdk.org> wrote:

>> src/jdk.javadoc/share/classes/jdk/javadoc/internal/doclets/toolkit/taglets/ReturnTaglet.java line 76:
>>> 74:         } else {
>>> 75:             List<? extends DocTree> firstSentence = utils.getFirstSentenceTrees(input.element);
>>> 76:             if (firstSentence.size() == 1 && firstSentence.get(0).getKind() == DocTree.Kind.RETURN) {
>> There's one more place where `firstSentence.size() == 1` is checked. I wonder if it could be a simpler `!firstSentence.isEmpty()` check?
>> Are there any non-trivial cases which might bite us later, should we change that check?
> I think the code is better as is.  It would be wrong to have a firstSentence containing multiple items, beginning with `{@return}` and then to only return the first one in the following line of code. Arguably, a better form may be to use `isEmpty()` and then subsequently assert (somehow) that the size is exactly 1, but using assertions like that is a different discussion (as you know!)

I guess I was thinking about a corner case like `{@summary {@return ...} ...}` which this patch and the complementary specification consider illegal anyway.


PR: https://git.openjdk.java.net/jdk/pull/1355

More information about the compiler-dev mailing list