[oe] [PATCH 29/31] xfce4-mixer: Remove, gstreamer-0.10 is obsolete

[Mark Hatle]

On 9/6/17 3:13 PM, Andreas Müller wrote:
> On Wed, Sep 6, 2017 at 10:02 PM, Mark Hatle <mark.hatle at windriver.com> wrote:
>> On 9/6/17 2:58 PM, Andreas Müller wrote:
>>> On Wed, Sep 6, 2017 at 9:23 PM, Mark Hatle <mark.hatle at windriver.com> wrote:
>>>> There is no upstream revision to the mixer to make it work w/o gstreamer-0.10
>>>>
>>>> also remove from packagegroup-xfce-extended
>>>>
>>>> Signed-off-by: Mark Hatle <mark.hatle at windriver.com>
>>>> ---
>>>>  meta-xfce/recipes-apps/xfce4-mixer/xfce4-mixer_4.11.0.bb   | 14 --------------
>>>>  .../packagegroups/packagegroup-xfce-extended.bb            |  1 -
>>>>  2 files changed, 15 deletions(-)
>>>>  delete mode 100644 meta-xfce/recipes-apps/xfce4-mixer/xfce4-mixer_4.11.0.bb
>>>>
>>>> diff --git a/meta-xfce/recipes-apps/xfce4-mixer/xfce4-mixer_4.11.0.bb b/meta-xfce/recipes-apps/xfce4-mixer/xfce4-mixer_4.11.0.bb
>>>> deleted file mode 100644
>>>> index bdbec45..0000000
>>>> --- a/meta-xfce/recipes-apps/xfce4-mixer/xfce4-mixer_4.11.0.bb
>>>> +++ /dev/null
>>>> @@ -1,14 +0,0 @@
>>>> -SUMMARY = "A volume control application based on GStreamer"
>>>> -LICENSE = "GPLv2"
>>>> -LIC_FILES_CHKSUM = "file://COPYING;md5=59530bdf33659b29e73d4adb9f9f6552"
>>>> -
>>>> -inherit xfce-panel-plugin
>>>> -
>>>> -DEPENDS += "glib-2.0 gst-plugins-base gtk+ xfconf libunique"
>>>> -
>>>> -SRC_URI = "http://archive.xfce.org/src/apps/${BPN}/${@xfce_verdir("${PV}")}/${BPN}-${PV}.tar.bz2"
>>>> -
>>>> -SRC_URI[md5sum] = "1b3753b91224867a3a2dfddda239c28d"
>>>> -SRC_URI[sha256sum] = "fb0c1df201ed1130f54f15b914cbe5a59286e994a137acda5609570c57112de2"
>>>> -
>>>> -RDEPENDS_${PN} = "gst-meta-audio"
>>>> diff --git a/meta-xfce/recipes-xfce/packagegroups/packagegroup-xfce-extended.bb b/meta-xfce/recipes-xfce/packagegroups/packagegroup-xfce-extended.bb
>>>> index c07188f..ff024b3 100644
>>>> --- a/meta-xfce/recipes-xfce/packagegroups/packagegroup-xfce-extended.bb
>>>> +++ b/meta-xfce/recipes-xfce/packagegroups/packagegroup-xfce-extended.bb
>>>> @@ -61,7 +61,6 @@ RRECOMMENDS_${PN} = " \
>>>>      xfce4-appfinder \
>>>>      xfce4-screenshooter \
>>>>      xfce4-power-manager \
>>>> -    xfce4-mixer \
>>>>      ristretto \
>>>>      xfce4-taskmanager \
>>>>      gigolo \
>>>> --
>>>> 1.8.3.1
>>>>
>>> NAK - I use it daily and there is no proper replacement (not using pulsaudio)
>>
>> That is exactly why I removed it.. there is no proper replacement for it.
>> Between gstreamer-0.10 not being supported and the existence of it in
>> meta-openembedded causing preferred provider problems.  It really needs to be
>> replaced with something else.  I have no idea 'what else', but I was looking and
>> most other distributions have dumped it.
>>
>> I saw some references to security issues in gstreamer-0.10 (not being resolved)
>> due to it no longer being supported as well, but no specifics.
>>
>> --Mark
>>
> You are about to remove a recipe I need very very very often. As you
> can see in meta-qt5-extra I use my build for making music. You end up
> in calling this other peoples problem (e.g mine) and don't really care
> of any replacement. Thank you - I will send out a removing me as
> maintainer for meta-xfce - fun times are definetely over.

You are missing the point.  We DO need a replacement.  However, I was unable to
find one.  With the few references I found when searching that there was no
replacement, and it was effectively obsolete (and removed in other
distributions), I followed their example.

As I said in the header, I don't believe this set is completely right.  But it's
a first stab and what it took for me to pass the testing script.

I am retesting with this patch and the removal of gstreamer removed from the
test set (it will take a while, I've got -30,000- tasks to run through for a
test).  It's possible that something else in the system was triggering a fault
that caused gstreamer-0.10 to fail and that was resolved elsewhere.

I tested the original work against master, but rebased to master-next after
talking w/ Martin... I then spot checked the modified recipes that they were
still working after the rebase.

> And please don't even think about running your magic script for
> meta-qt5-extra or send patches.

Again, you are missing the point here.  Someone has to start this work.  It's
the layer maintainers decision to accept or reject the patches, not mine..  but
I feel it's my obligation as a community member to submit the patches I create
to the community layers that I use.  I also feel it's my obligation as a patch
submitter to explain why (or why not) they are necessary for my specific
use-case.  (In this case, passing a specific compatibility test script.)

--Mark

> Over and out
> 
> Andreas
>