Mailing-List: contact dev-help@jmeter.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@jmeter.apache.org
MIME-Version: 1.0
Date: Sun, 5 Jul 2015 14:46:47 +0200
Message-ID: 
 <CAH9fUpacE4J24a7514XZuzhWdUkuYkEvEB7gz-DOc85V3C2Udw@mail.gmail.com>
Subject: Regression in 4.5 vs 4.2.6
From: Philippe Mouawad <philippe.mouawad@gmail.com>
To: HttpClient User Discussion <httpclient-users@hc.apache.org>
Cc: "dev@jmeter.apache.org" <dev@jmeter.apache.org>
Content-Type: multipart/alternative; boundary=001a1140f6e8ad8008051a202eb6

--001a1140f6e8ad8008051a202eb6
Content-Type: text/plain; charset=UTF-8

Hello,
We upgraded recently JMeter in a branch from HttpClient 4.2.6 to 4.5 and
are facing what seems to be a regression.


It seems that when we don't set a charset in StringBody , MultipartEntity,
FormBodyPart, no charset is written in ContentType generation while it was
the case at least in 4.2.6, I think issue was introduced in 4.3.

Note also that javadocs of these components say:
@param charset the character set to use, may be {@code null}, in which case
{@link MIME#DEFAULT_CHARSET} - i.e. US-ASCII - is used.

But it's not the case since 4.3

Reading the sepcs, I am not sure that when charset is US-ASCII , it should
be written but it was the case in 4.2.6.

I created https://issues.apache.org/jira/browse/HTTPCLIENT-1665

As a test case, you can have a look at JMeter test cases:

org.apache.jmeter.protocol.http.sampler.TestHTTPSamplersAgainstHttpMirrorServer#testPostRequest_FormMultipart3
org.apache.jmeter.protocol.http.sampler.TestHTTPSamplersAgainstHttpMirrorServer#testPostRequest_FileUpload3

The issue appears when no contentEncoding is set.
Thanks for help.
Regards
Philippe M.

--001a1140f6e8ad8008051a202eb6--