<html dir="ltr"><head></head><body style="text-align:left; direction:ltr;" bgcolor="#ffffff" text="#4c4c4c" link="#007aa6" vlink="#007aa6"><div>Rémi,</div><div><br></div><div>The bitstream dependencies can be brought into the library's private header, they are trivial and it will be under 20 lines of code. The GUI part is not needed and can be left out of the backport as well.</div><div><br></div><div>With these changes, the modules are 100% self contained and do not change the GUI, core or the build system. I will add additional patches to my tree to remove the bitstream dependency.</div><div><br></div><div>Regards,</div><div><br></div><div>Sergio</div><div><br></div><div>-----Original Message-----</div><div><b>From</b>: Rémi Denis-Courmont <<a href="mailto:%3d%3fISO-8859-1%3fQ%3fR%3dE9mi%3f%3d%20Denis-Courmont%20%3cremi@remlab.net%3e">remi@remlab.net</a>></div><div><b>Reply-To</b>: Mailing list for VLC media player developers <<a href="mailto:Mailing%20list%20for%20VLC%20media%20player%20developers%20%3cvlc-devel@videolan.org%3e">vlc-devel@videolan.org</a>></div><div><b>To</b>: Mailing list for VLC media player developers <<a href="mailto:Mailing%20list%20for%20VLC%20media%20player%20developers%20%3cvlc-devel@videolan.org%3e">vlc-devel@videolan.org</a>></div><div><b>Subject</b>: Re: [vlc-devel] [PATCH 3.0.x 0/4] Backport the RIST module</div><div><b>Date</b>: Mon, 02 Nov 2020 17:25:14 +0200</div><div><br></div><pre>Le maanantaina 2. marraskuuta 2020, 17.05.17 EET Thomas Guillem a écrit :</pre><pre></pre><pre>Hello Rémi</pre><pre><br></pre><pre>On Mon, Nov 2, 2020, at 14:19, Rémi Denis-Courmont wrote:</pre><pre></pre><pre>Le maanantaina 2. marraskuuta 2020, 15.03.35 EET Thomas Guillem a écrit :</pre><pre></pre><pre>Quoting Sergio:</pre><pre><br></pre><pre>"I would like to open the conversation for the backporting of the</pre><pre>simple-profile rist modules into the next 3.x release (3.0.12).</pre><pre><br></pre><pre>These modules have been tried and tested for over a year on the main</pre><pre>branch (since November 2018) and are of negligible or no risk to the</pre><pre>stability of the build ...</pre><pre></pre><pre><br></pre><pre>I'll have to disagree when the module is modifying common and GUI code.</pre><pre>There are modules that actually have negligible risk because they are</pre><pre>fully self- contained, and RIST is *not* one of them.</pre><pre></pre><pre><br></pre><pre>I share your concern. Maybe we should not merge the GUI part.</pre><pre>But there is still a problem about requiring a new dependency:</pre><pre>bitstream.Maybe we can move the bitstream header into the rist module for</pre><pre>3.0, I don't know.</pre><pre></pre><pre><br></pre><pre>I don't really care about adding a dependency - on technical level. It can be </pre><pre>a problem on a licensing level though, but probably not this specific case.</pre><pre><br></pre><pre>I do care about making changes to the build system, the core, the GUI, </pre><pre>anything, that affects other modules. There are simply no ways to convince </pre><pre>oneself that they won't add a subtle or not-so-subtle breakage in the stable </pre><pre>release due to accumulated differences from the master branch - no matter how </pre><pre>much the module has been allegedly testing in 4.0.</pre><pre><br></pre><pre>This changes both the GUI and the build system.</pre><pre><br></pre><pre>I understand that people want this in 3.0 rather than wait however long still </pre><pre>for 4.0 to come out. But IMO, this is a case for out-of-tree module in the </pre><pre>mean time.</pre><pre><br></pre><pre><br></pre></body></html>