Opened 20 years ago
Closed 20 years ago
#1768 closed patch (fixed)
Fix off by one error in assembly of PSIP
| Reported by: | anonymous | Owned by: | danielk |
|---|---|---|---|
| Priority: | minor | Milestone: | unknown |
| Component: | dvb | Version: | |
| Severity: | medium | Keywords: | |
| Cc: | Ticket locked: | no |
Description
attached patch fixes an of by one error in assembly of PSIP from ts packets. It is only triggered if the PSIP ends at the end of one tspacket.
I see that bug only after the removal of the sectionreader in DVBSiparser in [9816].
Is there a reason why the real data starts at PSIOffset()+1? I think I remember it makes sense if you look at the mpeg standard. But it confuses me everytime I look at it. So we should at least add a big red blinking comment stating the reason.
Attachments (1)
Change History (2)
by , 20 years ago
| Attachment: | assemble_psip_off_by_one_error.diff added |
|---|
comment:1 by , 20 years ago
| Resolution: | → fixed |
|---|---|
| Status: | new → closed |
Note:
See TracTickets
for help on using tickets.
(In [9857]) Closes #1768. Applies Janne's patch for off by one error in PSIP assembly against mythtv-eit branch.
We believe the +1 is needed due to the one byte allocated for to the pointer field in payload start packets.