2520 replies to this topic

[MastaG]

I think babsy98 meant to ask if the old dm800hd (non-se) also relies on: with-limited-mpeg4v2 ?

[christophecvr]

I think babsy98 meant to ask if the old dm800hd (non-se) also relies on: with-limited-mpeg4v2 ?

That I really don't . Since I do not have such box I can't test it. But You maybe wright.

[christophecvr]

Now it's clear for me that the dvbmediasink gst-1.0 corrections will never be done in openpli hosting server.

 

For that I now created a recipe in github

 

https://github.com/c...ox-dvbmediasink

 

It is first made with a base readme.

Then I added pli's mediasink master.

Then applied patch for gst-1.0 branch.

Then applied my latest 0001-0002-eos-vuplus-dreambox-3.patch

Then I added an example bb file as for how to build in image enviroment based on that from pli with correct uri then.

 

To use it on You're own gst-1 build

 

add the bb file to :

openpli40-gst1/openpli-oe-core/meta-openpli/recipes-multimedia/gstreamer

map

file name is gstreamer1.0-plugin-multibox-dvbmediasink.bb

content :

DESCRIPTION = "gstreamer dvbmediasink plugin"
SECTION = "multimedia"
PRIORITY = "optional"
LICENSE = "GPLv2"
LIC_FILES_CHKSUM = "file://COPYING;md5=7fbc338309ac38fefcd64b04bb903e34"

DEPENDS = "gstreamer1.0 gstreamer1.0-plugins-base libdca"
RDEPENDS_${PN} = "gstreamer1.0-libav"

GSTVERSION = "1.0"

SRC_URI = "git://github.com/christophecvr/gstreamer1.0-plugin-multibox-dvbmediasink;protocol=https"
SRCREV = "${AUTOREV}"

S = "${WORKDIR}/git"

inherit gitpkgv

PV = "${GSTVERSION}+git${SRCPV}"
PKGV = "${GSTVERSION}+git${GITPKGV}"
PR = "r0"

do_configure_prepend() {
        sed -i 's/AC_INIT.*$/AC_INIT(gst-plugin-dvbmediasink, 1.0.0, @pli4)/' ${S}/configure.ac
        sed -i 's/AM_INIT_AUTOMAKE.*$/AM_INIT_AUTOMAKE([foreign])/' ${S}/configure.ac
}

inherit autotools pkgconfig

FILES_${PN} = "${libdir}/gstreamer-${GSTVERSION}/*.so*"
FILES_${PN}-dev += "${libdir}/gstreamer-${GSTVERSION}/*.la"
FILES_${PN}-staticdev += "${libdir}/gstreamer-${GSTVERSION}/*.a"
FILES_${PN}-dbg += "${libdir}/gstreamer-${GSTVERSION}/.debug"

PACKAGE_ARCH = "${MACHINE_ARCH}"

EXTRA_OECONF = "${DVBMEDIASINK_CONFIG} --with-gstversion=${GSTVERSION}"

If You already build with gstreamer1.0-plugin-dvbmediasink just do in

openpli40-gst1/openpli-oe-core map

 

MACHINE=<you're machine> bitbake

 -b meta-openpli/recipes-multimedia/gstreamer/gstreamer1.0-plugin-dvbmediasink.bb -f -c cleanall

Then modifie the respectif machine config to have gstreamer1.0-plugin-multibox-dvbmediasink insteda off gstreamer1.0-plugin-dvbmediasink

 

and rebuild with

MACHINE=<you're machine> bitbake <You're image>

Once build ok :

 

telnet box.

 

issue commands :

opkg update

opkg remove gstreamer1.0-plugin-dvbmediasink --force-removal-of-dependent-packages

opkg install gstreamer1.0-plugin-multibox-dvbmediasink

and reinstall the other removed packages which You saw after uninstalling gstreamer1.0-plugin-dvbmediasink

 

Now You're box will follow the dynamic gst-1.0 mediasink updates.

[Andi316]

Now it's clear for me that the dvbmediasink gst-1.0 corrections will never be done in openpli hosting server.

 

For that I now created a recipe in github

 

https://github.com/c...ox-dvbmediasink

 

 

Why is this clear? Your work is perfect. Everybody here should be thankful for that!

[WanWizard]

I don't really see anyone saying that either.

[christophecvr]

 

Now it's clear for me that the dvbmediasink gst-1.0 corrections will never be done in openpli hosting server.

 

For that I now created a recipe in github

 

https://github.com/c...ox-dvbmediasink

 

 

Why is this clear? Your work is perfect. Everybody here should be thankful for that!

 

The one who are not are the pli4 Forum Moderators. The rest yes .

 

But the problem is solved. Just use the gstreamer1.0-plugin-multibox-dvbmediasink.bb and We can continu developpement to have the perfect gst-1 for all boxes and not only ET.

 

The Forum Moderators did not want to push the changes I did. At the end we end up with a to big patch. If someting is wrong it becomes almost impossible to trace then.

 

Now in that git I made we can work step by step. In further developpement and tuning in function off stb types.

Edited by christophecvr, 16 March 2015 - 15:39.

[MastaG]

I'm thankful, and as soon as I have some spare time, I'll rebuild the dm800se image with his patches applied and using the correct config parameters.

Then I'll report back of course.

I'll also borrow some recipes from oe-alliance and build one for the old dm800 just for fun.

I think Erik needs some confirmation from other users before he commits the patches.

So I hope more people chime in and keep on testing

Edited by MastaG, 16 March 2015 - 16:03.

[christophecvr]

I'm thankful, and as soon as I have some spare time, I'll rebuild the dm800se image with his patches applied and using the correct config parameters.

Then I'll report back of course.

I'll also borrow some recipes from oe-alliance and build one for the old dm800 just for fun.

I think Erik needs some confirmation from other users before he commits the patches.

So I hope more people chime in and keep on testing

It's not really Eric who does not wan't to push there are some Moderators which are just blocking it as for why .... Think the same reason as for why the bananas are curved.

 

Point is that I can not continu always with patch . The final patch will be far to big at the end to eventually trace the origin if a bug is discovered in a later stage (and you maybe shure that there will come some up).

 

Till now the gst-1.0 was only good for ET boxes . But even the base enigma2 and needed gstreamer1.0 packages where not right.

 

And that for more then a Year now. It's really time to have gstreamer1.0 . Even the vuplus does not work anymore for the majority of media with the latest master git off openpli. Looks a bit idiot of trying to solve this problem. Just take a look off the now already patches applied to ffmpeg. to just let work modern codecs with old gstreamer. And with the latest codecs it just does not work att all anymore.

 

Well just use my git . Extra dev's are welcome just ask me. Then we can make a dvbmediasink for all stb's not only for ET boxes

[WanWizard]

Just to be clear: forum moderators moderate the forum, they don't do code. I don't even have an account for the repository.

 

Who is blocking what, and why? Because I haven't seen anything to that effect...

[athoik]

This thread is almost one and a half year old and yet there is not a signle alpha/prebeta image with GStreamer 1.0 from OpenPLi.

The initial porting process was made by OpenPLi and then community is trying to solve pending issues!

But patches must be applied manually from community and provide feedback.

It's a long process and it is getting really annoy to solve issues and having patches left over this thread.

On the other hand OE-A is collecting patches from this thread and patches applied on a branch with GStreamer 1.0.


I guess OpenPLi is missing a patch apply process from this thread.

Eg apply patches every week or month that was reviewed by some people.

Edited by athoik, 16 March 2015 - 17:41.

[christophecvr]

Just to be clear: forum moderators moderate the forum, they don't do code. I don't even have an account for the repository.

 

Who is blocking what, and why? Because I haven't seen anything to that effect...

 

Well the patch is not pushed due to this comment off pieterg. (I gues)

 

 

So they are in the process of adapting their drivers to our sink, we should not patch the mediasink at the same time, might be better to avoid all manufacturer conditionals, and wait for what they come up with

 

But the box manufactors will never adapt the driver to a specific image (and then only the dvbmediasink) And so brake all other images. There is nothing wrong with vuplus driver concerning gst-1.0 or gst. And yes not the same as ET. But it's well same as all dreamboxes and .... So are the chipsets.

 

But no problem by just using my dvbmediasink git for all boxes except the ET series. Problem is solved and we can continu in a positif building dvbmediasink gst-1.0 which at the end should become ok for all mips stb in a first fase. The box type just given at configure time before compiling, so that it only does have the targets needed code. And for ET boxes it also will work perfect whitout a target stb.

[WanWizard]

It's dangerous to guess, especially if it creates a negative atmosphere.

 

Instead, let's find out what the stumbling blocks are, get them out of the way, and get this train rolling!

[christophecvr]

This thread is almost one and a half year old and yet there is not a signle alpha/prebeta image with GStreamer 1.0 from OpenPLi.

The initial porting process was made by OpenPLi and then community is trying to solve pending issues!

But patches must be applied manually from community and provide feedback.

It's a long process and it is getting really annoy to solve issues and having patches left over this thread.

On the other hand OE-A is collecting patches from this thread and patches applied on a branch with GStreamer 1.0.


I guess OpenPLi is missing a patch apply process from this thread.

Eg apply patches every week or month that was reviewed by some people.

 

Perhaps tell them to use my dvbmediasink git https://github.com/c...ox-dvbmediasink

 

I also can arange acces to more positif developpers. Then we can make a real multibox dvbmediasink for gst-1.0 since this does not exist yet. And just that is a problem in developpement for plugins.

[Andi316]

There are Gstreamer 1.4.5 Testimages in the OpenATV Forum for all VU+ Boxes, because of the work of christopheVcr. The reports are good, the playback is smooth! So let's use this new sink!

[christophecvr]

It's dangerous to guess, especially if it creates a negative atmosphere.

 

Instead, let's find out what the stumbling blocks are, get them out of the way, and get this train rolling!

 

That You're fully right.  Unfortunatelly already more then a year every dynamic helper tester and contributor in this toppic does not hit a stumbling block , but a armoured wall.

 

The only way to by pass this wall at the end is making an own dvbmediasink and then the dynamic users,testers and developpers can contribud this in a positif way. Which a the end will result in a real ok version for all boxes. and plugins can be adapted.

 

My two boxes dm8000 and vuduo2 are now just running fine with gst-1.0 the only missing things is plugins for gst-1.0 such as hbbtv. But once we have a stable git for gst-1.0 the plugins will be adapted.

[babsy98]

gst 1.45 i play mkv with dts and use jump keys and i get a  spinner not allways i play other file and ok, then play first file again i get no spinner all ok

[athoik]

@christophecvr, patience!

The process of adapting GStreamer 1.0 was never in a better state.

Currently you have at least VU+ developers trying to adapt OpenPLi dvbmediasink! And that was possible thanks to OpenPLi.

Also on this thread is @peteru is developer from Beyonwiz.

Having developers that know drivers internal looking at GStreamer 1.0 is much better (they are paid to do so...).

Of course Dreambox probably will never follow adapting OpenPLi dvbmediasink, so the --with-dreambox flag is ideal.


Your patches are almost ok. Give some time for people to test them It's hobby after all..

[babsy98]

hbbtv i have send xtrend and vu a mail with a request now we must wait, xtrend send mail back and say he testing gst 1.45 and check that is possible

[christophecvr]

@christophecvr, patience!

The process of adapting GStreamer 1.0 was never in a better state.

Currently you have at least VU+ developers trying to adapt OpenPLi dvbmediasink! And that was possible thanks to OpenPLi.

Also on this thread is @peteru is developer from Beyonwiz.

Having developers that know drivers internal looking at GStreamer 1.0 is much better (they are paid to do so...).

Of course Dreambox probably will never follow adapting OpenPLi dvbmediasink, so the --with-dreambox flag is ideal.


Your patches are almost ok. Give some time for people to test them It's hobby after all..

 

Yes indeed it's a hobby. Also for me

Hobby or not always bouncing into a armoured wall is really sad, especially the reason if ever one is mentionned makes no sence

 

That's why it takes so long to have a real ok image. On top of it the base build off gstreamer1.0 was actually not complete for what is needed in the stb's . This I actually I mentionned I gues 1 year ago ... ?? In those days i send even patches for it which never where applied well used by many users with positif results.  And then I just give it up for a while whit gst-1.0 the armoured wall was there.

 

That's now the reason I made a git for dvbmediasink multi box support . Ok that multi at this time is a bit limited    to ET (original pli4) Vuplus as far I tested 100 % dreambox pretty much all testing needs to show out and yes this will take some time i agree but dm8000 is almost 100 % ok except wmv3 on that I'll start works tommorow to check.

 

In this way we can make a more universal sink. Other image makers wo wan't t contribute in this git all welcome. At least so we will advance a bit.

Edited by christophecvr, 16 March 2015 - 18:48.

[Erik Slagter]

This is exactly why I am hesitant to apply your patches. You have only goal in mind, gstreamer-1 as soon as possible, on your settopboxes. We have other things to consider end-users not accepting it when media playback won't work as it used to. There are also other considerations you don't even know about.

 

And that is why I ask the other participants to test your patches first and if they're ok, I'll apply just the same.

 

If you have a problem with that, I suggest you fork the OpenPLi repository and use that as playground, with or without other participants, I don't have a problem with that approach at all.