RFR: JDK-8190452: javac should not add MethodParameters attributes to v51 and earlier class files

[Liam Miller-Cushon]

I reworked the test to use class reading, and added the -Xlint:option
warning.

Here's the updated patch:
http://cr.openjdk.java.net/~cushon/8190452/webrev.02/

On Sat, Feb 3, 2018 at 10:39 AM, Vicente Romero <vicente.romero at oracle.com>
wrote:

>
>
> On 02/03/2018 12:37 AM, Liam Miller-Cushon wrote:
>
> Hi Vicente, thanks for the review!
>
> I inlined the golden files into the test class. I don't think inlining
> Lib.java works with the current approach of using `@compile --release 7/8`
> and reflection to access the parameter names, because the reflective APIs
> were added in 8. I could rewrite the test to use class reading instead of
> reflection if you prefer.
>
>
> yes I was thinking about class reading but I'm OK with the current version
> of the patch
>
>
> Here's the updated webrev: http://cr.openjdk.java.net/~cushon/8190452/
> webrev.01/
>
> > Regarding your questions about what to do with the -Xlint:X options. I
> don't have any opinion on one way or the other, is there any reason to
> change them?
>
> I thought it was potentially surprising that -parameters will now be
> silently ignored for Java 7 and earlier language levels. Warning about that
> flag combination would make the new behaviour more discoverable.
>
> Since the bug shipped in 9 and 10 there are some <v52 class files with
> MethodParameters in the wild, and I've seen cases where it broke builds
> using -Xlint:classfile and -Werror.
>
> I don't think either of those are common problems. If you think we should
> leave the -Xlint handling as-is that sounds good to me.
>
>
> as discussed off-line with Jon, yes please add the warning, but feel free
> to do it as part of this bug or in a different one.
>
> Vicente
>
>
> On Fri, Feb 2, 2018 at 6:17 AM, Vicente Romero <vicente.romero at oracle.com>
> wrote:
>
>> Hi Liam,
>>
>> The fix looks OK, regarding the test, I don't see the need for the two
>> golden files as they can be constants in the test per se. In addition, the
>> whole test could be self contained in only one class that compiles the
>> Lib.java source. Regarding your questions about what to do with the
>> -Xlint:X options. I don't have any opinion on one way or the other, is
>> there any reason to change them?
>>
>> Thanks,
>> Vicente
>>
>>
>> On 02/01/2018 01:56 PM, Liam Miller-Cushon wrote:
>>
>> Bump. I'm happy to implement either of the alternatives I mentioned, but
>> was hoping to get feedback on the approach first.
>>
>> On Wed, Jan 17, 2018 at 11:51 AM, Liam Miller-Cushon <cushon at google.com>
>> wrote:
>>
>>> Please review a fix for JDK-8190452. The change causes javac to not
>>> emit MethodParameters attributes when targeting v51 class files.
>>>
>>> The change implements the suggestion from this thread:
>>> http://mail.openjdk.java.net/pipermail/compiler-dev/2018-Jan
>>> uary/011579.html
>>>
>>> There are two related changes that may be worth considering:
>>> * now that -parameters will be ignored when compiling with --release <
>>> 8, should this combination of flags result in a warning if -Xlint:options
>>> is enabled?
>>> * since this wasn't fixed in JDK 9, there are v51 class files in the
>>> wild that contain unexpected MethodParameters attributes. Should
>>> -Xlint:classfile be relaxed to avoid warning on those?
>>>
>>> bug: https://bugs.openjdk.java.net/browse/JDK-8190452
>>> webrev: http://cr.openjdk.java.net/~cushon/8190452/webrev.00/
>>>
>>
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/compiler-dev/attachments/20180203/2314a913/attachment-0001.html>