<AWT Dev> [9] Review request for: 8136763 [macosx] java always returns only one value for "text/uri-list" dataflavor even if several files were copied
mikhail cherkasov
mikhail.cherkasov at oracle.com
Wed Oct 21 15:38:31 UTC 2015
Thanks a lot!
On 10/21/2015 18:32, Alexander Scherbatiy wrote:
>
> The fix looks good to me.
>
> Thanks,
> Alexandr.
>
> On 10/21/2015 6:08 PM, mikhail cherkasov wrote:
>> Hi again,
>>
>> And one more check was added:
>> http://cr.openjdk.java.net/~mcherkas/8136763/webrev.04/src/java.desktop/macosx/classes/sun/lwawt/macosx/CDataTransferer.java.udiff.html
>>
>> Thanks,
>> Mikhail.
>>
>> On 10/21/2015 17:09, mikhail cherkasov wrote:
>>> Hi Alexander,
>>>
>>> You are right, NPE can be thrown there, I added check for null.
>>> A new webrev is being uploaded right now:
>>> http://cr.openjdk.java.net/~mcherkas/8136763/webrev.03
>>>
>>> Thanks,
>>> Mikhail.
>>>
>>> On 10/21/2015 16:42, Alexander Scherbatiy wrote:
>>>> On 10/21/2015 4:36 PM, mikhail cherkasov wrote:
>>>>> Hi all,
>>>>>
>>>>> Please review the fix. it requires to have two reviewer at least,
>>>>> so I need one more reviewer for
>>>>> this fix.
>>>>
>>>> It looks like the dragQueryFile(bytes) method can return null
>>>> and we can get NPE in the if(strings.length == 0) check in the
>>>> CDataTransferer.
>>>>
>>>> Thanks,
>>>> Alexandr.
>>>>
>>>>>
>>>>> Thanks,
>>>>> Mikhail.
>>>>>
>>>>> On 10/19/2015 14:01, mikhail cherkasov wrote:
>>>>>> Sergey, thank you.
>>>>>>
>>>>>> Alexander , could you please review the fix too?
>>>>>>
>>>>>> Thanks,
>>>>>> Mikhail.
>>>>>>
>>>>>>
>>>>>> On 10/19/2015 13:51, Sergey Bylokhov wrote:
>>>>>>> The fix looks fine. please fix the typo "returens" in
>>>>>>> CDataTransferer before the push.
>>>>>>>
>>>>>>> On 19.10.15 10:39, mikhail cherkasov wrote:
>>>>>>>> Hi there,
>>>>>>>>
>>>>>>>> http://cr.openjdk.java.net/~mcherkas/8136763/webrev.02/
>>>>>>>>
>>>>>>>> During adding the test, I've mistakenly returned flavormap
>>>>>>>> version with
>>>>>>>> removed "text/uri-list".
>>>>>>>> Now "text/uri-list" is in place again.
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>> Mikhail.
>>>>>>>>
>>>>>>>>
>>>>>>>> On 10/15/2015 11:42, mikhail cherkasov wrote:
>>>>>>>>> Hi Sergey, Alexander,
>>>>>>>>>
>>>>>>>>> Please review the new version of web rev with updated test.
>>>>>>>>>
>>>>>>>>> Thanks,
>>>>>>>>> Mikhail.
>>>>>>>>>
>>>>>>>>> On 10/7/2015 15:40, mikhail cherkasov wrote:
>>>>>>>>>> Hi Sergey,
>>>>>>>>>>
>>>>>>>>>> Sorry, I've added wrong test, there's a correct version:
>>>>>>>>>> http://cr.openjdk.java.net/~mcherkas/8136763/webrev.01/
>>>>>>>>>>
>>>>>>>>>> -Mikhail.
>>>>>>>>>>
>>>>>>>>>> On 10/6/2015 18:29, Sergey Bylokhov wrote:
>>>>>>>>>>> Looks like the test is not in the final state.
>>>>>>>>>>>
>>>>>>>>>>> On 06.10.15 11:01, mikhail cherkasov wrote:
>>>>>>>>>>>> Hi there,
>>>>>>>>>>>>
>>>>>>>>>>>> Please review a fix:
>>>>>>>>>>>> bug: https://bugs.openjdk.java.net/browse/JDK-8136763
>>>>>>>>>>>> webrev:
>>>>>>>>>>>> http://cr.openjdk.java.net/~mcherkas/8136763/webrev.00/
>>>>>>>>>>>>
>>>>>>>>>>>> The problem is similar to
>>>>>>>>>>>> https://bugs.openjdk.java.net/browse/JDK-8081787
>>>>>>>>>>>> When user requests data for uri-list flavor, java returns
>>>>>>>>>>>> xml without
>>>>>>>>>>>> any interpretation.
>>>>>>>>>>>> It is fixed by processing xml by dragQueryFile() method.
>>>>>>>>>>>>
>>>>>>>>>>>> The other problem: if several files has been copied, there're
>>>>>>>>>>>> available
>>>>>>>>>>>> two types in pasteboard for reading:
>>>>>>>>>>>> NSURLPboardType
>>>>>>>>>>>> NSFilenamesPboardType
>>>>>>>>>>>> but text/uri-list is associated only with NSURLPboardType -
>>>>>>>>>>>> which
>>>>>>>>>>>> returns only first value from list of files that was copied.
>>>>>>>>>>>> To resolve this I added:
>>>>>>>>>>>> "FILE_NAME=text/uri-list;eoln="\r\n";terminators=1" line above
>>>>>>>>>>>> "URL=text/uri-list;eoln="\r\n";terminators=1".
>>>>>>>>>>>> So now, java will select FILE_NAME and return a full list
>>>>>>>>>>>> of copied
>>>>>>>>>>>> files.
>>>>>>>>>>>>
>>>>>>>>>>>> Thanks,
>>>>>>>>>>>> Mikhail.
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>
More information about the awt-dev
mailing list