[ 
https://issues.apache.org/jira/browse/GUACAMOLE-2298?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18095064#comment-18095064
 ] 

Nick Couchman commented on GUACAMOLE-2298:
------------------------------------------

Thanks for the additional info [~vlouvet3] - this all makes sense, and I 
definitely understand/appreciate the desire for the royalty-free, web-based 
formats. I think #687 will get merged, here, shortly, so you can build on top 
of that.

> guacenc: add -c option to encode recordings as VP9/VP8 (WebM)
> -------------------------------------------------------------
>
>                 Key: GUACAMOLE-2298
>                 URL: https://issues.apache.org/jira/browse/GUACAMOLE-2298
>             Project: Guacamole
>          Issue Type: Improvement
>          Components: guacenc
>            Reporter: Vicente Louvet III
>            Priority: Minor
>         Attachments: guacenc-vp9-webm.patch
>
>
> Currently guacenc only produces MPEG-4 video (FILE.m4v). It would be useful to
> also support a royalty-free, browser-native output format, so that recordings
> can be played back directly in a browser and shared without codec-licensing
> concerns. Encoding to VP9 (or VP8) in a WebM container covers this, and libvpx
> is already available wherever guacenc's libavcodec is.
> The attached patch adds an optional -c flag which selects the libavcodec
> encoder by name. For example:
>     {color:#172b4d}guacenc -c libvpx-vp9 recording.guac -->   
> recording.guac.webm{color}
> Details:
>  - With no -c option, guacenc encodes mpeg4 to FILE.m4v exactly as before. The
>   default behavior is unchanged.
>  - The VP8 (libvpx) and VP9 (libvpx-vp9) encoders are written to a WebM 
> (.webm)
>   container. The output container is derived from the codec so the encoded
>   stream always lands in a compatible container.
>  - A codec name that isn't available in the local libavcodec build logs a
>   warning and falls back to mpeg4, rather than aborting the run.
> guacenc already passes the codec name straight through to libavcodec, so no
> encoding logic changed beyond the container selection and the new option. The
> diff is small and self-contained – guacenc only, no libguac / protocol /
> client changes. A CUnit test suite under src/guacenc/tests covers the
> codec-to-container mapping, and the man page documents the option.
> I'm happy to open a pull request against guacamole-server if this looks
> reasonable.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to