Re: CALL for DISCUSSION: eXIf 20170115
Glenn Randers-Pehrson <[email protected]>
| Newsgroups | gmane.comp.graphics.png.general |
|---|---|
| Message-ID | <CA+PdXct_wuDifayhf-F=U0zM_eaieiAyZF3HtFjVOtmZFX9F6g@mail.gmail.com> |
We probably ought to send both the original crc32 followed by the comp-chunk crc32. This provides assurance that nothing went wrong with the comp-decomp operation, that we wont get if we reconstitute the original crc32 while decoding. So, length of the comp chunk is compressed-chunk length + 14, and the comp chunk ends with <original crc32> <comp-chunk crc32> Glenn On Sat, Jan 28, 2017 at 8:29 PM, Glenn Randers-Pehrson <[email protected]> wrote: > > <chunk-length>: compressed length + 9 > > On Sat, Jan 28, 2017 at 8:23 PM, Glenn Randers-Pehrson <[email protected]> > wrote: > >> >> In effect, the decoder copies the original prefix, then decompresses >> the rest of the data. For chunks that don't have a prefix, the only >> difference from John's proposal is the presence of a prefix-length byte >> containing a zero. >> > > Also, > <chunk-length>: compressed length + 10 > instead of > <chunk-length>: compressed length + 9 > > Glenn > ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot _______________________________________________ png-mng-misc mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/png-mng-misc