Why not to use nsv_incr to generate sequential session id? Or sequential ids is a bad idea?
Ian Harding wrote: > Stephen Deasey wrote: >> On Sun, Feb 15, 2009 at 6:55 PM, Ian Harding <harding....@gmail.com> wrote: >> >>> Hello, >>> >>> I am migrating to Naviserver from AOLServer mostly because of the >>> excellent nsdbi module. >>> >>> My existing code is depended on the nssession module >>> http://bas.scheffers.net/aolserver/ . It doesn't compile and run >>> unmodified under Naviserver, but before I start hacking on it, is >>> there an existing session module or a tcl solution that someone else >>> has already written? >>> >>> >> >> I'm not sure that it's worth porting that module. The cache locking is >> broken, the file handling looks dodgy, and in general it has some odd >> ideas about configuration and string handling. >> >> If all you need to do is replicate some ns_session Tcl commands so >> your existing code works then the easiest way would be to create a >> work-alike using the existing naviserver Tcl commands: >> >> * cookie handling: naviserver has this built in: >> http://naviserver.sourceforge.net/n/naviserver/files/ns_cookie.html >> >> * caching: also now built in: >> http://naviserver.sourceforge.net/n/naviserver/files/ns_cache.html >> >> * persisting the session: just use nsdbi, if you're already using >> it. If you really want >> to use the file system, use ns_hashpath in nsd/pathname.c to >> prevent enormous >> directories from bogging down the server. >> >> * ns_rand: already exposed. >> >> >> It will be more robust than than the C nssession module, and almost >> certainly faster too. >> >> >> If you want to create something fancier as a C module, I'd be >> interested in helping. >> > Here is my lame hack at a tcl module. It has some problems and is not > done, but it does seem to work for certain small values of "work". > > http://code.google.com/p/grop/source/browse/trunk/nssession.tcl > > I want to "hide" the internal methods (id, load, etc) in a separate > namespace, and will use a proper session id generator. I'd like to get > rid of the upvar session_id. I wanted to use the connid as a key in a > cache to lookup the session_id for those cases where there is not a > cookie set yet. I couldn't figure out how to get it. I was hoping > there was a "ns_conn id" I could use. > > Anyway, I would appreciate all input on things I am doing wrong or could > do better. > > - Ian > > ------------------------------------------------------------------------------ > Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA > -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise > -Strategies to boost innovation and cut costs with open source participation > -Receive a $600 discount off the registration fee with the source code: SFAD > http://p.sf.net/sfu/XcvMzF8H > _______________________________________________ > naviserver-devel mailing list > naviserver-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/naviserver-devel > ------------------------------------------------------------------------------ Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise -Strategies to boost innovation and cut costs with open source participation -Receive a $600 discount off the registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H _______________________________________________ naviserver-devel mailing list naviserver-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/naviserver-devel
