Hi Pedro,
The patch looks good!
I am going to feed it to our internal formal review tool
and commit to the main trunk afterward.
Some more (one bigger and another minor) things bellow:
1. I noticed that (and it also applies to FileDataBodyPart class)
the tests only cover "constructor works fine" stuff.
Would you be willing to also contribute some more tests to get
at least the "and also works fine with multipart reader/writer"
stuff covered?
2. a minor comment: i suggest to fix the tests where an exception is
expected from:
try {
expectedToThrow();
} catch (ExpectedException e) {
}
to
try {
expectedToThrow();
fail(); // should never get here
} catch (ExpectedException e) {
}
but that i will fix by myself, no need to send updated diff for that.
And again: many thanks for your contribution!
~Jakub
On 5.10.2011 9:35, Pedro Kowalski wrote:
> Sure thing Jakub,
>
> Will wait for your review.
> Have a nice time at J1 - I envy you - at least I can read some news
> from Twitter :-)
>
> Cheers!
>
>
> 2011/10/4 Jakub Podlesak <jakub.podlesak_at_oracle.com
> <mailto:jakub.podlesak_at_oracle.com>>
>
> Hi Pedro,
>
> Thanks for the patch!
>
> I apologize for delayed response. Attending J1 this week.
> Will try to have a look and get back to you.
>
> Thanks,
>
> ~Jakub
>
>
> On 29.9.2011 15:34, Palli wrote:
>> Hi,
>>
>> I've attached the patch to the JIRA:
>> http://java.net/jira/browse/JERSEY-777
>> It adds the support for sending InputStream as a
>> multipart/form-data body part entity, which I think is more
>> adequate and still can be used for byte[] with ByteArrayInputStream.
>>
>> Can anyone take a look if it's acceptable?
>>
>> Cheers!
>> Pedro
>>
>> On 09/21/2011 07:41 PM, Jakub Podlesak wrote:
>>> Thanks. Looking forward to your contribution.
>>> Please keep the list posted if you have any question.
>>>
>>> ~Jakub
>>>
>>> On 21.9.2011 18:09, Palli wrote:
>>>> Hi Jakub,
>>>>
>>>> Great, there is a JIRA for this enhancement:
>>>> http://java.net/jira/browse/JERSEY-777
>>>>
>>>> As soon as I'll be able I'll take a look at it, as I'm quite
>>>> fancy getting known Jersey :-)
>>>>
>>>> Cheers!
>>>>
>>>> Pedro
>>>>
>>>> On 09/21/2011 02:46 PM, Jakub Podlesak wrote:
>>>>> Hi Pedro,
>>>>>
>>>>> +1 for the 2. option.
>>>>>
>>>>> And yes, contributions are warmly welcome.
>>>>> Please file a RFE, and attach a patch there.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> ~Jakub
>>>>>
>>>>> On 21.9.2011 12:05, pallipp_at_gmail.com
>>>>> <mailto:pallipp_at_gmail.com> wrote:
>>>>>> Howdy!
>>>>>>
>>>>>> I think it would be nice if we could have an additional
>>>>>> support for
>>>>>> byte[] instead of just flat File objects in FormDataBodyPart.
>>>>>> I think
>>>>>> there are situations in which we don't need to hit the disk
>>>>>> before
>>>>>> sending the data through the wire (or do we?)
>>>>>>
>>>>>> According to the javadoc [1] there is only a File object
>>>>>> support.
>>>>>>
>>>>>> If you agree that such feature is acceptable, than I think
>>>>>> that either:
>>>>>> 1. The FileDataBodyPart should be adjusted to support byte[] or,
>>>>>> 2. Another subclass of FormDataBodyPart should be created
>>>>>> which will
>>>>>> support the byte[].
>>>>>>
>>>>>> Personally, I'd prefer the 2. option.
>>>>>>
>>>>>> What do you think about that?
>>>>>> Are you open for contributions?
>>>>>>
>>>>>> The idea for this feature was born in Arquillian project [2].
>>>>>>
>>>>>> Cheers!
>>>>>>
>>>>>> Pedro
>>>>>>
>>>>>> [1] -
>>>>>> http://jersey.java.net/nonav/apidocs/1.9.1/contribs/jersey-multipart/co
>>>>>>
>>>>>> m/sun/jersey/multipart/file/FileDataBodyPart.html
>>>>>> [2] - https://issues.jboss.org/browse/ARQ-569
>>>>>>
>>>>
>>>>
>>
>