There are 24 failing tests if the `isValid()` check is omitted in HtmlTree.add(Content). I started going after „invalid“ HtmlTree instances using stack traces to locate their origin, but gave up after 3 or 4 due to the reasons given earlier: there’s quite a few of them, and even if we find them all there’s no way to prevent them from popping up in the future.
My conclusion would be to close this PR/issue as „won’t fix“ and start working on finding a better solution. IMO that solution would still be implemented in Content/HtmlTree, but should have a name that better reflects its purpose, and would possibly be implemented using different means at a different level. Hannes > Am 02.06.2021 um 16:32 schrieb Jonathan Gibbons <jonathan.gibb...@oracle.com>: > > > On 6/2/21 7:13 AM, Hannes Wallnöfer wrote: >> >> I agree that our current `isValid` implementation does not feel right. >> >> But there are many (dozens?) of places in the code that potentially generate >> empty elements. If we want to avoid these, shifting the responsability to >> the content producer will blow up the code a lot. I'm trying out different >> things to see if I can come up with a better solution. > > Do we have any sense of how often the "isValid" check prevents an empty node > to be added to the tree? > > > >> >> ------------- >> >> PR: https://git.openjdk.java.net/jdk/pull/4130
