Thank you Riccardo, On Tue, Mar 29, 2016 at 5:45 PM, Riccardo Magliocchetti < [email protected]> wrote:
> One possible solution would be to write the JSON parsing in C, but I can't > write > >> proper C code anymore... :( >> > > Have you tried ujson? maybe it's fast enough and does not consume too much > memory > The problem is not the JSON decoder itself I guess, but keeping the whole structure in memory. It's a few millions dicts, and dicts are heavy... But I may be wrong, I've no numbers Another solution would be to call a LUA RPC, but we would have memory >> problems >> because uWSGI RPC only supports strings as an exchange format, a single >> string I >> think. Are there tricks to easily dump to file/unix pipes or something? >> > > If you want to go down this route maybe you can send to lua just the query > url, process the file in lua and just return the processed output position > instead of its full content? > Any experience spawning something like a thread in LUA, returning before it finishes and consuming the generated file from python? In this way I would be able to speed up response writing. Thank you again. -- Davide Setti code: http://github.com/vad
_______________________________________________ uWSGI mailing list [email protected] http://lists.unbit.it/cgi-bin/mailman/listinfo/uwsgi
