Closed GoogleCodeExporter closed 9 years ago
Original comment by se...@google.com
on 23 Mar 2011 at 9:46
I presume Google had the same desire internally. The best choice externally
may be different, but could you provide insight into which of pure
java/jni/something else was used inside Google?
Original comment by chris.bu...@gmail.com
on 25 Mar 2011 at 2:02
I'm sorry, I can't say much about what we do internally.
Original comment by se...@google.com
on 25 Mar 2011 at 2:38
Same here — I'd have a good use for snappy in a Clojure application, but for
that I need it to exist in the Java world. Doesn't have to be pure Java,
JNI/JNA is fine.
A maven-distributed library would be ideal.
Original comment by jrych...@gmail.com
on 26 Mar 2011 at 11:21
[deleted comment]
Hi all,
I am not sure how snappy is used with Java inside the Google, but
I started to develop a java-port of snappy, snappy-java, which uses JNI-based
implementation to access natively compiled snappy.
http://code.google.com/p/snappy-java
I confirmed that snappy can be used fine from Java.
I would like to hear your advice to improve the snappy-java API to
compress/decompress the data.
Original comment by taroleo
on 30 Mar 2011 at 9:48
Hi,
As there now is a third-party JNI wrapper around Snappy, and people seem happy
about it, the need for something in the Snappy distribution itself seems rather
low. Thus, I'm closing this. Thanks for making this available!
Original comment by se...@google.com
on 2 Jun 2011 at 12:24
Just wanted to add that I've implemented a pure Java version of Snappy. The
compressor is not yet the fastest, but achieves mostly better result than the
original implementation. The decompressor is reasonable fast.
You can find the implementation here on Google Code in the jsnappy project:
http://code.google.com/p/jsnappy/
Original comment by tejarn...@googlemail.com
on 26 Aug 2011 at 10:52
Nice. Would you like me to add that to the web page along with the other ports?
Do you have any benchmarks comparing this to the JNI version?
Original comment by se...@google.com
on 30 Aug 2011 at 8:49
Yes, please link it everywhere you see it fit.
I haven't compared it to the JNI version myself, but Chris Burroughs is already
working on adding it to his JVM compressor benchmark suite. I hope that at
least the decompressor is not far behind.
Original comment by tejarn...@googlemail.com
on 30 Aug 2011 at 9:20
Thanks; added a link.
Original comment by se...@google.com
on 30 Aug 2011 at 9:26
Original issue reported on code.google.com by
bryan.du...@gmail.com
on 23 Mar 2011 at 3:04