16:21:20 GMT <WiuEmPe> hello, anyone can look on my slowlog? I dont know if prolem is in queries or in system performance
16:21:33 GMT <WiuEmPe> http://pastebin.com/tMkuEZTz
16:49:55 GMT <emr> Hello, what will be happen if i publish a message and no one receives?
16:50:19 GMT <minus> publish returns 0
16:50:50 GMT <emr> well zero is good
20:03:07 GMT <radens> I'm trying to model a series of package databases in redis. i.e. the gcc program can be found in the gcc package on ubuntu and debian but the gnu-compiler-collection package on arch (or whatever). This naturally lends itself to a series of tables (packages, distros, programs) with foreign keys all over the place, but I can't easily model this with redis. I have something with triply nested hash tables in mind,
20:03:08 GMT <radens> but that just seems wrong.
20:50:55 GMT <minus> i.e. apt-file + pkgfile + others in redis?
20:55:29 GMT <bonsaikitten> radens: it sounds a bit ... kinda ... relational
20:58:19 GMT <minus> yup, tree of hashes kinda
20:58:43 GMT <minus> how you can implement depends on which direction you want lookups to happen
20:59:55 GMT <bonsaikitten> if only we had databases that were good with relational structures
21:00:17 GMT <bonsaikitten> as a workaround one could use hierarchical databases like LDAP, they support some useful structures
21:01:29 GMT <minus> but yeah, SQL seems more fit to the problem
21:08:49 GMT <berken> im storing values as "tx:<someid>" "{some json}". each object has a "time" field. how can i query for values within a given time range (from, to)? i realize i'll have to store additional for this, not sure what/how to store to allow this kind of lookup. ideally id have one query to return ids that are within the range and then fetch all ids matching