Hello Vincent, Thank you very much for the effort. I've created a JIRA issue to follow this bug [1]. Which version of mime4j are you using? Please submit the patch via JIRA and accept the ASF license agreement so we can include it. Also, please add a test that demonstrates the bug. We are all volunteers and a bit low on manpower so all help is greatly appreciated. You can take chekout the code from SVN or git, whichever you prefer [2], [3]. I think the best place to add the test is core/src/test/java/org/apache/james/mime4j/codec/QuotedPrintableInputStreamTest.java .
[1] https://issues.apache.org/jira/browse/MIME4J-227 [2] http://svn.apache.org/repos/asf/james/mime4j/trunk/ [3] https://github.com/apache/james-mime4j Thank you, On Mon, Apr 15, 2013 at 5:41 AM, Vincent Untz <[email protected]> wrote: > Hi, > > I noticed that when using text using LF for line breaks as input, > QuotedPrintableOutputStream has a mix of LF and CRLF in the output for > line breaks: current code just keeps LF where they are, but uses CRLF > when doing a soft break. As quoted-printable is specified as using CRLF > for line breaks, I guess it makes sense to just always use CRLF. > > This has the nice side-effect of fixing two other bugs when input is > using LF: > - soft break counter was not reset after the LF, so soft breaks are > happening with no reason on some short lines > - trailing space/tab is not encoded > > Cheers, > > Vincent > > PS: I'm not subscribed to the list, so please cc me on replies. > > -- > Les gens heureux ne sont pas pressés. -- Ioan Eugen Stan 0720 898 747
