diff options
| author | agc <agc@pkgsrc.org> | 2007-05-05 00:03:54 +0000 |
|---|---|---|
| committer | agc <agc@pkgsrc.org> | 2007-05-05 00:03:54 +0000 |
| commit | 72f70f2fc6a70ba615b9fd1d55530a0f9c434e9d (patch) | |
| tree | ec1bca5ffe8c47af90c0b120a6a0ec41100940af /security/py-SSLCrypto/DESCR | |
| parent | 9001703891cc02b263665420a905ba37d5c0baf3 (diff) | |
| download | pkgsrc-72f70f2fc6a70ba615b9fd1d55530a0f9c434e9d.tar.gz | |
Initial import of py-SSLCrypto-0.1.1 into the Packages Collection.
SSLCrypto is a package for Python that dramatically eases the task of
adding encryption to Python programs.
It provides a unified API that is almost totally compatible with that
of ezPyCrypto, except that it takes advantage of the OpenSSL Crypto
Library to deliver massive improvements in speed and security.
After using ezPyCrypto myself, I found that while it performed ok with
smaller public key sizes, it proved impossibly slow with larger keys.
This slowness, resulting from non-optimal code in its backend (the
Python Cryptography Toolkit) meant that on a 1.5 GHz Athlon XP, it was
taking several minutes to generate 4096-bit keys. Completely
unacceptable if you need real security.
Performance is absolutely critical for an encryption API. If slowness
deters people from using adequate-sized keys, security will be
severely compromised, almost to the extent that there's little point
in using encryption in the first place.
Diffstat (limited to 'security/py-SSLCrypto/DESCR')
| -rw-r--r-- | security/py-SSLCrypto/DESCR | 18 |
1 files changed, 18 insertions, 0 deletions
diff --git a/security/py-SSLCrypto/DESCR b/security/py-SSLCrypto/DESCR new file mode 100644 index 00000000000..f15834802f0 --- /dev/null +++ b/security/py-SSLCrypto/DESCR @@ -0,0 +1,18 @@ +SSLCrypto is a package for Python that dramatically eases the task of +adding encryption to Python programs. + +It provides a unified API that is almost totally compatible with that +of ezPyCrypto, except that it takes advantage of the OpenSSL Crypto +Library to deliver massive improvements in speed and security. + +After using ezPyCrypto myself, I found that while it performed ok with +smaller public key sizes, it proved impossibly slow with larger keys. +This slowness, resulting from non-optimal code in its backend (the +Python Cryptography Toolkit) meant that on a 1.5 GHz Athlon XP, it was +taking several minutes to generate 4096-bit keys. Completely +unacceptable if you need real security. + +Performance is absolutely critical for an encryption API. If slowness +deters people from using adequate-sized keys, security will be +severely compromised, almost to the extent that there's little point +in using encryption in the first place. |