RFR: JDK-8180480 Use "requires transitive" relationship when determining modules for javadoc

Erik Joelsson erik.joelsson at oracle.com
Fri May 19 08:16:52 UTC 2017


Looks good.

/Erik


On 2017-05-19 10:08, Magnus Ihse Bursie wrote:
>
>
> On 2017-05-18 17:30, Mandy Chung wrote:
>>> On May 18, 2017, at 12:54 AM, Magnus Ihse Bursie 
>>> <magnus.ihse.bursie at oracle.com> wrote:
>>>
>>> When the build system tries to figure out which modules that should 
>>> be included by the javadoc build, it locates the set of modules 
>>> "required" by already included modules, starting from an initial set 
>>> and repeating recursively (a method which we unfortunately called 
>>> "transitive”).
>> Which module has this name clash?
>>
>>> However, for javadoc, we should only look at those re-exported 
>>> modules that are classified by a "requires transitive" relationship 
>>> (which I therefore have called "re-exported" to not clash with the 
>>> established term "transitive”).
>> Javadoc uses the “Indirect Exports” term for the exported API 
>> packages are from the “requires transitive” modules.
>
> "Indirect exports" was a good term. I dropped the "Deps" part of the 
> name and it all became much clearer.
>
> Here's an updated webrev:
> http://cr.openjdk.java.net/~ihse/JDK-8180480-docs-should-use-requires-transitive/webrev.02 
>
>
> /Magnus
>
>>
>> I suggest to rename “ReExported” to “IndirectRequires” or “Indirect”.
>>
>> Mandy
>>
>>> The effect of getting this wrong is e.g. that we copy module graph 
>>> png files that we should not, and that we look at too broad a set of 
>>> files for looking at changes that should trigger a re-build of the 
>>> javadoc.
>>>
>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8180480
>>> WebRev: 
>>> http://cr.openjdk.java.net/~ihse/JDK-8180480-docs-should-use-requires-transitive/webrev.01
>>>
>>> /Magnus
>




More information about the build-dev mailing list