On Mon, 26 Jan 2026 16:11:26 GMT, Maurizio Cimadamore <[email protected]> wrote:
>> `JavacParser` is still only recording end positions for the trees where it >> calls `storeEnd`, so the same kinds of trees have end positions recorded in >> the new world. I think that potentially `JavacParser` could start recording >> end positions for all trees, and then this switch could be removed. I will >> investigate that more. If it works out, I wonder if it might be better as a >> separate follow-up change, to keep the initial refactoring to remove the end >> pos table as behaviour preserving as possible. What do you think? > > I think it's ok to start simple and refine as we go. As part of this change > it would be perhaps helpful to at least simplify the method so that > "unreachable cases" are removed? (longer term, (as discussed offline with @lahodaj ) I think it would be nice to have end positions set up correctly by parser for all trees, since now it doesn't cost extra to do so) ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/28610#discussion_r2728321043
