[mythtv] DVB Development

Taylor Jacob rtjacob at earthlink.net
Sun May 23 01:50:41 EDT 2004


I JUST sent you an email about the exact same subject, before I checked the dev list.  I have a
bunch of routines I have been working on to do NIT/SDT/EIT/PAT/PMT scanning for integrating into
Myth that are 99% complete.  I am actually a few days away from placing it into
dvbsections/dvbchannel.  I already have written some code for setup to create virtual inputs for
diseqc switches that I had planned on rolling out to make satellite setup a bit more logical.

I appoligize for not keeping people up to date on what I am working on, but its been hard to find
time to work on it.  I also am going to be out of town for work, and then vacation and possibly
have limited Internet access for the next 3 weeks starting Monday.

I have written most of functions just from the ETSI specs, and have the values populated into
QValueList objects that could be re-usable not only for scanning, but automaticlly tuning to a
service, and removing the need for the dvb_pids table.  I attempted to take some code from VDR,
but the adaptation time seemed longer than it was worth.

My code is mostly still in a hacked up version of dvbdate that I have been building everything
from as a base.  If you are interested in taking a look at what I have email me, and I will be
glad to share.  Most of the time has been spent learning about QT, and all the nice objects you
can create which is what I dump the tables into.  I was thinking that some of the objects would be
dual purpose.  For example the PMT QValueList will be used for tuning a service, but also will be
useful as part of the service scanning feature that needs to be added into setup very badly.

Please look in your personal inbox for what I already sent you.

I am going to make the assumption that you and I are the only people really interested in taking
this large task on since I have asked numerous times on the list for help testing code, and I have
only received 1 email from someone who would help me.

I do have the EIT scanning code already done (outside of myth), and with the limited availablity
of EPG in North America all of my tests look positive.  I can dump 2 days of guide data for around
300 channels in about 25 seconds.

> Giddiup!
>
> Pretty general subject, I know. Basically what I want to bring forth is who is
> working on what feature, because there has been some people that have claimed
> to be working on features like autopid, scan and epg. However I have not seen
> any further mails about this subject, and I'm wondering about wether these
> features is in progress or not?
>
> I've been working a little on a parsing library for dvb tables and
> descriptors, so that scan/autopid/epg features can be implemented with very
> simple code on top of this parsing library/code. This code is essentially
> developed for another project, but it has no external dependencies, is
> released under the gpl, and so is easily portable into for example mythtv.
> Now, if nobody is working on the forementioned features, I would like to
> import this library into mythtv, and base all other dvb features upon it.
> This would however require a final restructuring of the dvb database
> parameters, as the library handles parsing of dvb tables from the dvb network
> it would be cleaner to make the internal storage in mythtv like what is in
> the network. This adds another dependency if you want to call it that, in
> that all tuning information will be stored in a binary format, and will not
> allow easy editing without thoroug knowledge about the inner workings about
> dvb descriptors. When this is done it will contain features such as autoscan
> with empty tables, which will try to do a step by step tuning (if no direct
> source is configured). Once the dvb code gets a lock on a frequency it will
> start scanning for other channels available from the network information
> table. When channel tables are filled the code will enter an update state, so
> that every time a channel is tuned into, it will, in the background, reload
> and check the information in the database for consistency. Any new channels
> will be added and channels that is no longer existing in the dvb information
> will be removed, and only tv channels will be treated in this process, in
> other words any channel that does not have at least on audio pid and one
> video pid will be treated as bogus. What basically the autopid feature will
> do is pick the highest pids from a prioritylist of languages, so that tv
> channels will always be output with your default languange if it exist or any
> other available language if not (english beeing a preffered language if no
> other is set or found, this could also be made a configurable). On the
> subject of epg, I have not been able to extract anything other than now/next
> information about the current program/other program on the same transponder.
> I know for a fact that vdr does this without a hitch (but also has a
> dependency xmltv just in case), and I have not looked at how vdr gathers this
> information. If anybody knows how to gather epg data from a satellite or
> other dvb source, I would greatly appreaciate the contribution, and use more
> time on this project than I have done in the last months.
>
> If there is anything in here you belive that I have come short to think of, or
> have anything to add/clearify to this statement, please reply to this post to
> the mythtv-dev at mythtv.org list and not to my personal e-mail account,
> plainly because my e-mail used on this list has been so cluttered with so much
> spam that I rearly catch any personal e-mails at all (why don't someone fix
> this bug?).
>
> --
> Kenneth *
>
> * Looking forward to the day I can contribute to the Linux Community!
> _______________________________________________
> mythtv-dev mailing list
> mythtv-dev at mythtv.org
> http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
>



More information about the mythtv-dev mailing list