some nashorn testing
Marcus Lagergren
marcus.lagergren at oracle.com
Thu Jan 10 08:19:04 PST 2013
So code generation wise, I've verified that the type system is internally consistent. This is a rhino/nashorn different when it comes to JSR223.
/M
On Jan 10, 2013, at 4:18 PM, Attila Szegedi <attila.szegedi at oracle.com> wrote:
> Hi Andreas,
>
> I'm sorry - maybe then I did not understand the motivation for you sending that last code snippet with 2D arrays. Did you just want to illustrate another difference in array handling, or does it illustrate a further problem that you'd need solved?
>
> Attila.
>
> On Jan 10, 2013, at 4:05 PM, Andreas Rieber <rieberandreas at gmail.com> wrote:
>
>> Hi Attila,
>>
>> yes i read all mails and there are already four different workaround for the array topic. No problem, just different from Rhino.
>>
>> Andreas
>>
>> On 10.01.13 14:34, Attila Szegedi wrote:
>>> Andreas, did you read Sundar's response where he explained to you the need to use Java.toJavaArray() for explicit conversion of JavaScript arrays to Java arrays? It seems that your code still doesn't use Java.toJavaArray().
>>>
>>> Attila.
>>>
>>> On Jan 10, 2013, at 2:07 PM, Andreas Rieber <rieberandreas at gmail.com> wrote:
>>>
>>>> Hi Marcus,
>>>>
>>>> i improved the test and found another case (sorry for that ;-)
>>>>
>>>> Whenever you have a fix, i could provide the patch for jrunscript (workaround working):
>>>>
>>>> ./usr/local/src/jdk8/build/linux-i586/j2sdk-image/bin/jrunscript -cp /usr/local/src/jdk8/nashorn/dist/nashorn.jar:. -l nashorn
>>>> nashorn> printf("%03d %2.2f%n",1,2);
>>>> 001 2.00
>>>> nashorn>
>>>>
>>>> - Andreas
>>>>
>>>> On 09.01.13 16:50, Marcus Lagergren wrote:
>>>>> My bad. This actually reproduces through the JSR223 bridge. Must be an eval related problem. I am on it!
>>>>>
>>>>> /M
>>>>>
>>>>>
>>>>> On Jan 9, 2013, at 4:26 PM, Marcus Lagergren <marcus.lagergren at oracle.com> wrote:
>>>>>
>>>>>> Where did you get/build your nashorn.jar? We'll get proper versioning in there as soon as we've integrated fully with the OpenJDK build system. I still get nothing when I manually build the latest nashorn.jar from the openjdk and run the JS snippet from the command line with java -jar nashorn.jar <program.js> (Haven't gone through the script engine bridge, but as it seems to be a JavaScript problem it should be sufficient to run it as a standalone script). This very much looks like a bug that I fixed before christmas and pushed into the closed repo - I've verified that it made it into the openjdk source.
>>>>>> /M
>>>> <ObjectArray.java>
>
More information about the nashorn-dev
mailing list