Nuo Mi

Results 110 comments of Nuo Mi

if we can use hash, we can use a more generic way to do the hash verification. uint8_t hash[AV_HASH_MAX_SIZE]; uint8_t *target = h->md5 const char* name[] = {"MD5", "CRC24", "adler32"}...

> If we want to add this test to ffmpeg fate, we may need to generate a new one without license. I can do this, thank you.

> > > > @QSXW can we use av_hash_alloc instead of hand writingcheck_md5 and check_checksum? for crc, maybe we need to add 16bits crc to hash.h. But it should be...

> > > > > > @QSXW can we use av_hash_alloc instead of hand writingcheck_md5 and check_checksum? for crc, maybe we need to add 16bits crc to hash.h. But it...

try CRC-CCITT (0x1D0F) at https://www.lammertbies.nl/comm/info/crc-calculation or CRC-AUG-CCITT at https://www.sunshine2k.de/coding/javascript/crc/crc_js.html CCITT is a part of ITU :) all crc will append zeros see https://www.zlib.net/crc_v3.txt

> Hi @nuomi2021 . > > Based on the poly and the initial value matching the CRC16-CCITT-False, I test the data with it but the result also doesn't match the...

@rdoeffinger, thank you for the guidance. However, that's not quite it. We only need to use 0x0F1D as the initialization value. This value is the result of XOR, shift, and...

@QSXW CRC fixed. please help review and send the entire pr to upstream for review.

Hi Reimar Thank you for your code > Also by your description, 0x0F1D is the crc16 of two bytes of 0. It's 0xFF 0xFF, not two byes of 0 We...

> > It's 0xFF 0xFF, not two byes of 0 > > It's equivalent, initialize the crc to 0 and push 2 times 0xff or initialize to 0xfffff and push...