You are viewing a plain text version of this content. The canonical link for it is here.
Posted to httpclient-users@hc.apache.org by Andrey Pokhilko <ap...@ya.ru> on 2015/05/27 12:55:45 UTC
FIleBody constructor regression
Hi,
I have a situation when upgrading from httpmime-4.2.6 to httpmime-4.3.1
(same for httpmime-4.4.1) causes regression. I'm using this constructor:
public FileBody(File file, String mimeType)
In version 4.2.6 there was a logic to set "filename" to file.getName().
In version 4.3.1 this logic was lost. This leads to situation when I
have to upgrade to 4.3+ because one of my libraries requires it, but it
breaks another library in my app, which relies on this auto-setting
filename. I have no control over both libraries source code.
I'm aware that the constructor above is deprecated since 4.3, but as I
understand deprecation, it does not imply breaking old behavior.
Hope I was clear with my explanations.
What would you suggest? Is this a regression of httpmime?
--
Andrey Pokhilko
Re: FIleBody constructor regression
Posted by Andrey Pokhilko <ap...@ya.ru>.
Done! Pull request is here: https://github.com/apache/httpclient/pull/30
Andrey Pokhilko
On 05/27/2015 03:09 PM, Andrey Pokhilko wrote:
> Great!
>
> I'll try to compose a pull request ASAP. Wait for me!
>
> Andrey Pokhilko
>
> On 05/27/2015 02:29 PM, Oleg Kalnichevski wrote:
>> On Wed, 2015-05-27 at 13:55 +0300, Andrey Pokhilko wrote:
>>> Hi,
>>>
>>> I have a situation when upgrading from httpmime-4.2.6 to httpmime-4.3.1
>>> (same for httpmime-4.4.1) causes regression. I'm using this constructor:
>>>
>>> public FileBody(File file, String mimeType)
>>>
>>> In version 4.2.6 there was a logic to set "filename" to file.getName().
>>> In version 4.3.1 this logic was lost. This leads to situation when I
>>> have to upgrade to 4.3+ because one of my libraries requires it, but it
>>> breaks another library in my app, which relies on this auto-setting
>>> filename. I have no control over both libraries source code.
>>>
>>> I'm aware that the constructor above is deprecated since 4.3, but as I
>>> understand deprecation, it does not imply breaking old behavior.
>>>
>>> Hope I was clear with my explanations.
>>>
>>> What would you suggest? Is this a regression of httpmime?
>>>
>> Most likely it is. Feel free to submit a patch or raise a PR at Github.
>> If you hurry the fix can still make it into HC 4.5 GA release.
>>
>> Oleg
>>
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache.org
>> For additional commands, e-mail: httpclient-users-help@hc.apache.org
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache.org
> For additional commands, e-mail: httpclient-users-help@hc.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache.org
For additional commands, e-mail: httpclient-users-help@hc.apache.org
Re: FIleBody constructor regression
Posted by Andrey Pokhilko <ap...@ya.ru>.
Great!
I'll try to compose a pull request ASAP. Wait for me!
Andrey Pokhilko
On 05/27/2015 02:29 PM, Oleg Kalnichevski wrote:
> On Wed, 2015-05-27 at 13:55 +0300, Andrey Pokhilko wrote:
>> Hi,
>>
>> I have a situation when upgrading from httpmime-4.2.6 to httpmime-4.3.1
>> (same for httpmime-4.4.1) causes regression. I'm using this constructor:
>>
>> public FileBody(File file, String mimeType)
>>
>> In version 4.2.6 there was a logic to set "filename" to file.getName().
>> In version 4.3.1 this logic was lost. This leads to situation when I
>> have to upgrade to 4.3+ because one of my libraries requires it, but it
>> breaks another library in my app, which relies on this auto-setting
>> filename. I have no control over both libraries source code.
>>
>> I'm aware that the constructor above is deprecated since 4.3, but as I
>> understand deprecation, it does not imply breaking old behavior.
>>
>> Hope I was clear with my explanations.
>>
>> What would you suggest? Is this a regression of httpmime?
>>
> Most likely it is. Feel free to submit a patch or raise a PR at Github.
> If you hurry the fix can still make it into HC 4.5 GA release.
>
> Oleg
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache.org
> For additional commands, e-mail: httpclient-users-help@hc.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache.org
For additional commands, e-mail: httpclient-users-help@hc.apache.org
Re: FIleBody constructor regression
Posted by Oleg Kalnichevski <ol...@apache.org>.
On Wed, 2015-05-27 at 13:55 +0300, Andrey Pokhilko wrote:
> Hi,
>
> I have a situation when upgrading from httpmime-4.2.6 to httpmime-4.3.1
> (same for httpmime-4.4.1) causes regression. I'm using this constructor:
>
> public FileBody(File file, String mimeType)
>
> In version 4.2.6 there was a logic to set "filename" to file.getName().
> In version 4.3.1 this logic was lost. This leads to situation when I
> have to upgrade to 4.3+ because one of my libraries requires it, but it
> breaks another library in my app, which relies on this auto-setting
> filename. I have no control over both libraries source code.
>
> I'm aware that the constructor above is deprecated since 4.3, but as I
> understand deprecation, it does not imply breaking old behavior.
>
> Hope I was clear with my explanations.
>
> What would you suggest? Is this a regression of httpmime?
>
Most likely it is. Feel free to submit a patch or raise a PR at Github.
If you hurry the fix can still make it into HC 4.5 GA release.
Oleg
---------------------------------------------------------------------
To unsubscribe, e-mail: httpclient-users-unsubscribe@hc.apache.org
For additional commands, e-mail: httpclient-users-help@hc.apache.org