Re: Is this a bug?(About multi-sections)

Hi all,

from ISO/IEC 13818-1, Annex C, C.2 iii):

"The section_number field allows the sections of a particular table
to be reassembled in their original order

by the decoder. There is no obligation within this Recommendation |
International Standard that sections

must be transmitted in numerical order, but this is recommended,
unless it is desired to transmit some

sections of the table more frequently than others, e.g. due to
random access considerations."

Regards

Josep

Al 05/12/2012 09:26, En/na Jean-Paul Saman ha escrit:

HI,

On Wed, Dec 5, 2012 at 8:45 AM,
jiangpengfei <[email protected] >
wrote:

Hi everyone:

I tested the libdvbpsi v1.0.0 pre2 in
my project today, I found there is a problem when
libdvbpsi handles a table which split into
multi-sections.

My network is a DVB network, and there
is a NIT split into two sections(the last section
number is 1). In the test, I found only the
information in the last section of NIT had been
handled in the most conditions, the sec0’s data
couldn’t be handled correctly. After read
libdvbpsi’s source code, I found this function:
dvbpsi_decoder_psi_sections_completed verifies if a
table’s sections have been gathered completely. But
I think this
condition(p_decoder->i_last_section_number ==
p->i_number) is not perfect, because if the first
section I received is last section of table, this
condition will be true and will active the callback
I registered. So I change the function implement to
the following:

It does not sound logical to transmit the last part of a
multi section psi before sending other parts. I'll check
with the official standards if this behaviour is allowed.

Does your own equipment generate this transport stream with
multi-section psi tables?

Could you provide me with a sample file?

Kind regards,

Jean-Paul Saman.

bool
dvbpsi_decoder_psi_sections_completed(dvbpsi_decoder_t*
p_decoder)

{

assert(p_decoder);

assert(p_decoder->i_last_section_number <=
255);

bool b_complete = true;

/* dvbpsi_psi_section_t *p =
p_decoder->p_sections;

* while (p)

* {

* if
(p_decoder->i_last_section_number ==
p->i_number)

* b_complete = true;

* p = p->p_next;

* }

*/

int i;

for(i = 0; i <=
p_decoder->i_last_section_number; i ++)

{

dvbpsi_psi_section_t * p =
p_decoder->p_sections;

while(p)

{

if(i == p->i_number)

break;

p = p->p_next;

}

if(p == NULL)

b_complete = false;

}

return b_complete;

}

After recompile, the libdvbpsi can
handle my NIT correctly.

I’m a newer of libdvbpsi, I don’t want
change any code of it, so can anyone tell me this is
a bug or I misunderstand something?

Fred from China.

_______________________________________________

libdvbpsi-devel mailing list

[email protected]

http://mailman.videolan.org/listinfo/libdvbpsi-devel

_______________________________________________
libdvbpsi-devel mailing list
[email protected]
http://mailman.videolan.org/listinfo/libdvbpsi-devel

_______________________________________________
libdvbpsi-devel mailing list
[email protected]
http://mailman.videolan.org/listinfo/libdvbpsi-devel