Huh?: Gstreamer Basic Tutorial / API (3: Dynamic pipelines)

Thu May 16 13:29:34 UTC 2019

Le jeu. 16 mai 2019 09 h 25, R C <cjvijf at gmail.com> a écrit :

> So basically one would never have  two elements that are the same, and
> only differ in the name you made/created them with?
>

Element names must unique per GstBin only. You'd get an error from
gst_bin_add() if you had a duplicate. Always check your return values,
there is no exception in C to save the day.

> On Thu, May 16, 2019 at 7:19 AM Nicolas Dufresne <nicolas at ndufresne.ca>
> wrote:
>
>>
>>
>> Le mer. 15 mai 2019 23 h 25, R C <cjvijf at gmail.com> a écrit :
>>
>>> so while messing with it,  I forgot to change teh second part, and even
>>> though it seems incosistent, can someone explain why this works?  (I am
>>> just putting the declarations here,  rewritting to ugly got's into
>>> something nicr is not the problem)
>>>
>>>
>>> in the main part the declarations I had are:
>>>
>>> // Create the elements
>>> data.source = gst_element_factory_make("uridecodebin", "source");
>>> data.audioconvert = gst_element_factory_make("audioconvert",
>>> "audioconvert");
>>> data.audiosink = gst_element_factory_make("autoaudiosink", "audiosink");
>>> data.videoconvert = gst_element_factory_make("videoconvert",
>>> "videoconvert");
>>> data.videosink = gst_element_factory_make("autovideosink", "videosink");
>>>
>>>
>>> and I forgot to make the changes in the second part (pad_added_handler),
>>> surprised it worked, and seems counter-intuitive?
>>>
>>> GstPad *audiosink_pad = gst_element_get_static_pad(data->audioconvert,
>>> "sink");
>>> GstPad *videosink_pad = gst_element_get_static_pad(data->videoconvert,
>>> "sink");
>>>
>> The names in the first part are the element names, while in this
>> callback, you are handling pad names. Pad names are chosen by the elements
>> and follow the pattern in the pad template (see gst-inspect-1.0 element to
>> see the template).
>>
>>
>>
>>> but  ..  why??  How can it even find the right GstElement, since none of
>>> the are named "sink"?
>>>
>>>
>>> docs:
>>>
>>> ******************************************************************
>>> gst_element_get_static_pad ()
>>>
>>> GstPad <https://gstreamer.freedesktop.org/data/doc/gstreamer/head/gstreamer/html/GstPad.html> *
>>> gst_element_get_static_pad (*GstElement <https://gstreamer.freedesktop.org/data/doc/gstreamer/head/gstreamer/html/GstElement.html> *element*,
>>>                             *const gchar <https://gstreamer.freedesktop.org/usr/share/gtk-doc/html/glib/glib-Basic-Types.html#gchar> *name*);
>>>
>>> Retrieves a pad from *element* by name. This version only retrieves
>>> already-existing (i.e. 'static') pads.
>>>
>>> *Parameters*
>>>
>>> element a GstElement
>>> <https://gstreamer.freedesktop.org/data/doc/gstreamer/head/gstreamer/html/GstElement.html>
>>> to find a static pad of.
>>>
>>> name the name of the static GstPad
>>> <https://gstreamer.freedesktop.org/data/doc/gstreamer/head/gstreamer/html/GstPad.html>
>>> to retrieve.
>>> Returns
>>>
>>> the requested GstPad
>>> <https://gstreamer.freedesktop.org/data/doc/gstreamer/head/gstreamer/html/GstPad.html>
>>> if found, otherwise NULL
>>> <https://gstreamer.freedesktop.org/usr/share/gtk-doc/html/glib/glib-Standard-Macros.html#NULL:CAPS>.
>>> unref after usage.
>>>
>>> MT safe.
>>>
>>> ************************************************************************
>>>
>>>
>>>
>>> thanks,
>>>
>>>
>>> Ron
>>>
>>>
>>>
>>> On 5/15/19 8:33 PM, R C wrote:
>>>
>>> Hello,
>>>
>>> I am trying to get a little more familiar with gstreamer, and thought
>>> I'd do some
>>>
>>> of the "basic" programming tutorial examples.
>>>
>>>
>>> I did play with the 3rd example:
>>>
>>>
>>> https://gstreamer.freedesktop.org/documentation/tutorials/basic/dynamic-pipelines.html?gi-language=c
>>>
>>>
>>> If I copy the example, and compile it, it works, no surprise there. But
>>> when I change it and do the "exercise", I get some weirdness.
>>>
>>>
>>> What I noticed is (well I probably missed something)...  but the
>>> function call:
>>>
>>> data.sink = gst_element_factory_make ("autoaudiosink", "sink");
>>> According to the docs, the 2nd parameter can be 'whatever' you want to name it.
>>> Since I wanted to do the video part too, I figure I should change the name to
>>> "audio-something", and changed the second function call to:
>>>   data.convert = gst_element_factory_make ("audioconvert", "convert");   (original)
>>>   data.sink = gst_element_factory_make ("autoaudiosink", "audiosink");   (changed name)
>>>
>>> of course in the even handler (pad_added_handler), the name needs to be changed in to:
>>> GstPad *sink_pad = gst_element_get_static_pad (data->convert, "audiosink");
>>>
>>> Browsing through the code, that should be it, BUT it seems "sink_pad" always is NULL
>>>  and the pipeline elements don't/can't be linked. (see errors below)
>>>
>>> since this is supposed to be a simple example,  what am I missing?
>>>
>>> thanks,
>>>
>>> Ron
>>>
>>>
>>> errors:
>>> # ./basic-tutorial-3
>>> Pipeline state changed from NULL to READY:
>>> Received new pad 'src_0' from 'source':
>>>
>>> (basic-tutorial-3:14679): GStreamer-CRITICAL **: 20:22:54.991: gst_pad_is_linked: assertion 'GST_IS_PAD (pad)' failed
>>> It has type 'video/x-raw' which is not raw audio. Ignoring.
>>>
>>> (basic-tutorial-3:14679): GStreamer-CRITICAL **: 20:22:54.992: gst_object_unref: assertion 'object != NULL' failed
>>> Received new pad 'src_1' from 'source':
>>>
>>> (basic-tutorial-3:14679): GStreamer-CRITICAL **: 20:22:54.992: gst_pad_is_linked: assertion 'GST_IS_PAD (pad)' failed
>>>
>>> (basic-tutorial-3:14679): GStreamer-CRITICAL **: 20:22:54.992: gst_pad_link_full: assertion 'GST_IS_PAD (sinkpad)' failed
>>> Type is 'audio/x-raw' but link failed.
>>>
>>> (basic-tutorial-3:14679): GStreamer-CRITICAL **: 20:22:54.992: gst_object_unref: assertion 'object != NULL' failed
>>> Error received from element source: Internal data stream error.
>>> Debugging information: gstbasesrc.c(2950): gst_base_src_loop (): /GstPipeline:test-pipeline/GstURIDecodeBin:source/GstSoupHTTPSrc:source:
>>> streaming stopped, reason not-linked (-1)
>>>
>>>
>>> _______________________________________________
>>> gstreamer-devel mailing list
>>> gstreamer-devel at lists.freedesktop.org
>>> https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel
>>
>> _______________________________________________
>> gstreamer-devel mailing list
>> gstreamer-devel at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel
>
> _______________________________________________
> gstreamer-devel mailing list
> gstreamer-devel at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/gstreamer-devel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/gstreamer-devel/attachments/20190516/c5c282af/attachment.html>