RFR: 8210047 : api/overview-summary.html contains content outside of landmark region
Priya Lakshmi Muthuswamy
priya.lakshmi.muthuswamy at oracle.com
Wed Sep 12 11:57:42 UTC 2018
Hi Jon,
updated the webrev with the changes for frames and the refactoring done
for handling the tags.
webrev : http://cr.openjdk.java.net/~pmuthuswamy/8210047/webrev.00/
[accidentally uploaded the changes in webrev.00 itself]
Thanks,
Priya
On 8/31/2018 5:54 AM, Jonathan Gibbons wrote:
>
>
> On 08/28/2018 01:25 AM, Priya Lakshmi Muthuswamy wrote:
>> Hi,
>>
>> Kindly review the fix for
>> https://bugs.openjdk.java.net/browse/JDK-8210047
>> webrev : http://cr.openjdk.java.net/~pmuthuswamy/8210047/webrev.00/
>>
>> Thanks,
>> Priya
>
>
> Just because we don't use frames in the JDK API docs doesn't mean they're
> not a supported feature, for a while at least.
>
> So, shouldn't the fix also cover docs where frames -are- in use.
>
>
>
> Separately, the fix is "ugly" because it makes a (different) bad
> situation worse,
> and I'm not sure at this stage what the best way forward is.
>
> The general bad situation, that needs cleaning up, is the overall
> handling
> of "htmlTree" and HtmlTag.MAIN. The existing code is pretty ugly in the
> way that htmlTree is set up (too early) and then later handled with code
> like
>
> 165 if (configuration.allowTag(HtmlTag.MAIN)) {
> 166 htmlTree.addContent(div);
> 167 } else {
> 168 body.addContent(div);
> 169 }
>
> It would be better to be building stuff in a more bottom up approach so
> that you build the content, and then at a single place, decide whether
> it needs to be wrapped in a MAIN tag.
>
> I need to think whether we should go with your fix, and make more places
> that need to be cleaned up later, or whether we should just get it
> right, now.
>
> -- Jon
More information about the javadoc-dev
mailing list