I don't pretent to understand all the crypto-locking stuff in pycurl.c.
The pycurl_ssl_init() calls etc. for OpenSSL, nss and GnuTLS. I've
built my libcurl (MSVC) with WinSSL and it all seems to work just
fine w/o any locking. Is this the case?

A small test (essentially example/retriever.py) retrieves a .html from
https://www.fortify.net/sslcheck.html just fine AFAICS.

And here is the version output from
  print >>sys.stderr, 'Testing', pycurl.version

Testing PycURL/7.19.0.3 libcurl/7.34.1-DEV WinSSL zlib/1.2.8 c-ares/1.10.1-DEV WinIDN
 libssh2/1.4.4_DEV nghttp2/0.3.0

(allthough the above libssh2 uses OpenSSL, my libcurl doesn't)

--gv
_______________________________________________
http://cool.haxx.se/cgi-bin/mailman/listinfo/curl-and-python
Received on 2014-01-15