gots.InsertPTS improvements #121
Conversation
| @@ -129,7 +129,7 @@ func InsertPTS(b []byte, pts uint64) { | |||
| b[4] = byte(pts&0xff) << 1 // PTS[7..0] | |||
|
|
|||
| // Set the marker bits as appropriate | |||
| b[0] |= 0x21 | |||
| b[0] |= 0x01 | |||
There was a problem hiding this comment.
I'm not really sure about this change. The tests pass with and without it.
There was a problem hiding this comment.
Oh, I'm glad you commented on this, as we don't want to change from 0x21 to 0x01. 0x21 sets the PTS_DTS_flags field to "has PTS". However, I will agree that's a bit sneaky and we should probably include a clarifying comment.
There was a problem hiding this comment.
I see, so that means this function is not suitable for writing DTS. Don't the number of marker bits required vary depending on whether DTS is being written also?
There was a problem hiding this comment.
You could use this function to write DTS, but you would have to set the mask on b[0] to 0x11 for the DTS portion of the PTS_DTS field and then append the two generated byte buffers together. Realistically I don't think that's the best plan if you need to write both, but it would be your code, not mine :) The actual marker_bits are the same between the two fields, and they don't change position or anything like that if you have both PTS and DTS.
If you need to write a PTS_DTS field, rather than just PTS, I would probably recommend implementing a InsertPTSDTS(b []byte, pts uint64, dts uint64) where you could call InsertPTS(b[:5], pts) and then insert the DTS and modify the PTS_DTS_flag (b[0]) accordingly.
| @@ -129,7 +129,7 @@ func InsertPTS(b []byte, pts uint64) { | |||
| b[4] = byte(pts&0xff) << 1 // PTS[7..0] | |||
|
|
|||
| // Set the marker bits as appropriate | |||
| b[0] |= 0x21 | |||
| b[0] |= 0x01 | |||
There was a problem hiding this comment.
Oh, I'm glad you commented on this, as we don't want to change from 0x21 to 0x01. 0x21 sets the PTS_DTS_flags field to "has PTS". However, I will agree that's a bit sneaky and we should probably include a clarifying comment.
cleaned up the tests a bit, and fixed what appears to be a bug in the marker bit insertion.