On Mon, 24 Oct 2022 20:44:08 GMT, Magnus Ihse Bursie <i...@openjdk.org> wrote:
>> It would be possible to mark classes with `@sealedGraph` to indicate that >> the sealed hierarchy should be rendered graphically in the JavaDocs. Classes >> not marked would be unaffected, creating an opt-in approach. >> >> [Here](https://bugs.openjdk.org/secure/attachment/101126/Screen%20Shot%202022-10-19%20at%2012.47.47%20PM.png) >> is an example of MemoryLayout class rendering. >> >> This feature has been developed in cooperation with Per Minborg. > > Magnus Ihse Bursie has updated the pull request incrementally with one > additional commit since the last revision: > > Rename sealedGraphDotPath to sealedDotOutputDir Some minor comments, but generally OK make/Main.gmk line 1114: > 1112: docs-reference-api: docs-reference-api-graphs > 1113: > 1114: # We must generated javadoc first so we know what graphs are needed grammar: `generated` -> `generate` make/jdk/src/classes/build/tools/taglet/SealedGraph.java line 40: > 38: import java.util.*; > 39: import java.util.stream.Collectors; > 40: Weirdly-sorted imports. It is common practice to put `java.*` and `java.*` imports first. make/jdk/src/classes/build/tools/taglet/SealedGraph.java line 106: > 104: } catch (FileNotFoundException e) { > 105: throw new RuntimeException(e); > 106: } It's certainly unusual for a taglet `.toString` call to have this sort of side effect. It would not the a showstopper, but have you checked whether this code is just executed once or multiple times? ------------- Marked as reviewed by jjg (Reviewer). PR: https://git.openjdk.org/jdk/pull/10761
