RFR JDK-8174245: Javadoc is not working for some methods

[Robert Field]

Looks good.

Style nit (fix or not):  When conditionals are multi-line, JShell uses

    conditional
        ? trueCase
        : falseCase

Which is consistent with this and other style guides:

http://www.oracle.com/technetwork/java/javase/documentation/codeconventions-136091.html

The JavadocHelper.java code is in the compiler module, which by 
non-random sampling appears to use several styles, including the above, 
but more:

    conditional?
           trueCase
        : falseCase

I'm only pointing this out because I found the two line form hard to read.

-Robert


On 02/09/17 08:02, Jan Lahoda wrote:
> Hi,
>
> Bug:
> https://bugs.openjdk.java.net/browse/JDK-8174245
>
> Webrev:
> http://cr.openjdk.java.net/~jlahoda/8174245/webrev.00/
>
> The primary problem here is that showing javadoc for e.g.:
> Collections.min(
> or:
> Collections.max(
>
> fails. The reason is that when JavadocHelper computes the javadoc for 
> the method, it needs to parse the source for the given class, and find 
> the method in the parsed source. But, when parsing/entering 
> java.util.Collections outside of the java.base module, the 
> MethodSymbols for the min/max (and other) methods are too much broken 
> to be recognized as the methods we are searching for (because the 
> java.util package belongs to the java.base module). The proposed 
> solution is to use the new --patch-module semantics to parse/enter the 
> class as-if it was in the java.base module.
>
> Note that patch depends on this change:
> http://cr.openjdk.java.net/~jlahoda/8173777/langtools.02/
>
> A secondary problem is that when printing multiple javadocs, there is 
> a note printed:
> <press space for next javadoc, Q to quit>
> but it may not be fully overwritten when writing the next javadoc (or 
> page). The proposed solution is to simply overwrite it with spaces 
> before writing the further text.
>
> Feedback is welcome!
>
> Thanks,
>     Jan