[FFmpeg-devel] [PATCH] MLP/TrueHD Decoder - 2nd try

[Ramiro Polla]

Michael Niedermayer wrote:
> On Tue, Jul 01, 2008 at 02:09:10AM +0100, Ramiro Polla wrote:
>> Michael Niedermayer wrote:
>>> On Sun, Jun 29, 2008 at 12:41:17PM +0100, Ramiro Polla wrote:
>>>> Hello,
>>>>
>>>> 01_mlp_parser_gb.diff makes a function in mlp_parser receive a 
>>>> GetBitContext instead of uint8_t buffers. This helps out in the decoder 
>>>> to simplify the code.
>>> ok, after fixing issue below
>> [...]
>>
>>>> -int ff_mlp_read_major_sync(void *log, MLPHeaderInfo *mh, const uint8_t 
>>>> *buf,
>>>> -                           unsigned int buf_size)
>>>> +int ff_mlp_read_major_sync(void *log, MLPHeaderInfo *mh, GetBitContext 
>>>> *gb)
>>>>  {
>>>> -    GetBitContext gb;
>>>>      int ratebits;
>>>>      uint16_t checksum;
>>>>  -    if (buf_size < 28) {
>>>> +    assert(get_bits_count(gb) == 0);
>>>> +
>>>> +    if (gb->size_in_bits < 28 << 3) {
>>>>          av_log(log, AV_LOG_ERROR, "Packet too short, unable to read 
>>>> major sync\n");
>>>>          return -1;
>>>>      }
>>>>  -    checksum = mlp_checksum16(buf, 26);
>>>> -    if (checksum != AV_RL16(buf+26)) {
>>>> +    checksum = mlp_checksum16(gb->buffer, 26);
>>> You should document the "no stuff read yet" requirement in a doxy comment
>>> (or add get_bits_count()>>3 iam not sure which is better ...)
>> Is this what you meant? I just added this line to ff_mlp_read_major_sync's 
>> doxy comment:
>>  *  gb must be a freshly initialized GetBitContext with no bits read.
> 
> patch ok

Applied.