I am using ZLibCompressedStream (Deflate64 option) to perform in memory compression and write the results to a database. I have found a few cases (20 in 30000) where the decompressed data does not match. The failure is repeatable – ie I have source data that always fails in a consistent way. The failure does not occur if the Deflate option is used on the same data. My data is characterised by large numbers of zeroes which result in a very high compression ratio which may be a factor. The fault occurs in v3.1 and v3.4 although the failure is slightly different. Has anyone experienced this problem? Is this a known bug in these versions that is fixed in later versions.
Other users may want to be cautious in using the deflate64 option until Xceed have had a chance to comment on this and confirm the bug or otherwise
Attached are two files compressed from the same data using ZlibCompressedStream but with different deflate options. The Deflate64 does not decompress correctly
Imported from legacy forums. Posted by Richard (had 1055 views)
This has been confirmed as a bug on v3.4 that was fixed in v3.5 and later
Also confirmed that even in the most recent versions ZLibCompressedStream does not use the header to determine the compression method so it is necessary to know in advance the type of compression and set this on instantiation
Imported from legacy forums. Posted by Richard (had 1280 views)