IRC Networks
Irc Logs Stats
Start date: 2007-09-27 02:48:27
Last update: 2008-10-24 20:19:38
Channels: 41
Logged Lines: 6230436
Size: 1822.62 MB
Powered by
Channel Info
Network: freenodeChannel: #csharp |
Search in www.irclog.org
Log from #csharp at freenode 2006-05-25
[01:58]<kjr>tieTYT: I'd imagine it's sql 88 or 89 or whatever it was
[01:58]<kjr>that came before 92
[01:58]<kjr>89 I think
[01:58]<vynvyv>well at least one digit's right
[01:58]<kjr>Arild: wait, what are you supposed to use instead of outer joins?
[01:59]<azyfm>SELECT * FROM Bleh LEFT OUTER JOIN Blah ON Bleh.x = Blah.Y ?
[01:59]<kjr>as opposed to...?
[02:00]<kjr>that's still an outer join
[02:00]<azyfm>I'm talking about the *= and =* operators
[02:00]<sdnffdxdlld>http://codebetter.com/blogs/sahil.malik/archive/2005/11/16/134694.aspx
[02:00]<kjr>Arild: I'm not sure I follow
[02:00]<sdnffdxdlld>they have stopped the usage of the wrong style ones
[02:00]<azyfm>"In earlier versions of Microsoft® SQL Server™ 2000, left and right outer join conditions were specified in the WHERE clause using the *= and =* operators. In some cases, this syntax results in an ambiguous query that can be interpreted in more than one way. SQL-92 compliant outer joins are specified in the FROM clause and do not result in this ambiguity. Because the SQL-92 syntax is more precise, detailed information about using the old Transa
[02:00]<azyfm>"
[02:01]<kjr>oh, I'm not familiar with MSSQL's proprietary syntacies
[02:01]<azyfm>*= and =* is common in other databases too
[02:01]<vynvyv>Arild: oracle uses (+)
[02:02]<kjr>Arild: weird... I don't recognize it from PG or MySQL... can't say I've worked with too many others
[02:02]<kjr>of course, I'm not a crackhead... so I try and write things that are proper
[02:02]<kjr>I think that MySQL.com should document on their very large manual in large letters sections per syntax about it deviating from SQL92
[02:03]<kjr>like the SELECT modifiers they use
[02:03]<kjr>Arild: if it used *= and =*... what'd inner joins use?
[02:03]<azyfm>=
[02:03]<kjr>ah
[02:04]<azyfm>that's still supported
[02:04]<azyfm>without the compat switch
[02:04]<kjr>well, that's pretty logical
[02:04]<kjr>the idea of = being an inner join is pretty sane... *= is pretty abstract
[02:04]<azyfm>"The SQL-92 standard does support the specification of inner joins in either the FROM or WHERE clause. Inner joins specified in the WHERE clause do not have the same problems with ambiguity as the Transact-SQL outer join syntax."
[02:04]<kjr>yeah
[02:05]<kjr>Arild: I was never sure if filtering via WHERE clauses produced a cartesian product, or optomized exactly the same as an explicit INNER JOIN
[02:05]<kjr>Arild: I once did a test, but I think I accdentally used a LEFT JOIN
[02:05]<azyfm>ah
[02:05]<kjr>because I was smoking SQL crack, and it was late at night
[02:05]<azyfm>(crack, that is)
[02:06]<kjr>and for some reason, I came to the conclusion that filtering via WHERE does a cartesian product and reduces and that explicit INNER JOINs work differently ... they probably work the exact same way though
[02:06]<kjr>I can't imagine why they wouldn't - one is implicit of the other
[02:06]<vynvyv>you thought filtering via WHERE always does a cartesian product?
[02:06]<kjr>it must have been some high quality SQL92 crack
[02:06]<txnjzd>Kog: I've found cartesian and explicit inner to be different at least on mysql
[02:06]<vynvyv>Kog: i was once told they can be optimized better in the from
[02:07]<kjr>Theory: yes, but a cartesian product may not be produced by filtering in WHERE
[02:07]<kjr>Theory: as opposed to a full outer join
[02:07]<txnjzd>(orders of magnitude different in speed)
[02:07]<kjr>tieTYT: I think it optomizes the same if you explicitly inner join or do a where
[02:07]<vynvyv>Kog: I do too, but someone told me otherwise in #sql
[02:07]<kjr>oh schnap, you read it on IRC... must be true
[02:07]<vynvyv>the optimizer would be a dumbass if it didn't
[02:08]<kjr>tieTYT: most of the people in #sql are tards anyway
[02:08]<vynvyv>yeah dick's too
[02:08]<kjr>I just join whatever RDBMS I am curious about and speak to people that know what they're talking about
[02:08]<kjr>like #mysql and #postgresql
[02:08]<kjr>tieTYT: man, I found a lovely issue w/ MySQL last night
[02:08]<vynvyv>what's that?
[02:08]<kjr>tieTYT: I was doing a BETWEEN on a DATE_FORMAT'd datetime
[02:08]<kjr>tieTYT: but it would only compare month/day... it didn't take into account the year
[02:09]<vynvyv>were you formatting it right?
[02:09]<kjr>so if I searched a date range of 05/17/1902 -> 05/22/2012 it would find it
[02:09]<vynvyv>including years in the format?
[02:09]<kjr>but 12/12/1902 -> 12/12/2012 wouldn't
[02:09]<kjr>tieTYT: %m/%d/%Y
[02:10]<kjr>tieTYT: so...I ended up doing DATEDIFF
[02:10]<vynvyv>well that doesn't sound very good at all
[02:10]<kjr>no, it's fucking insane, and it took me hours to figure out what it was doing
[02:11]<vynvyv>haha
[02:11]<vynvyv>that sucks ass
[02:11]<kjr>now... they document that %X%A is invalid because it won't check a portion of the date, but not the format I was using
[02:11]<kjr>tieTYT: oh yeah, I was REAL fucking awake when the headhunter woke me up at 10:30AM after going to bed at 0600
[02:11]<vynvyv>i donno what %X%A means, oracle doesn't use that stuff
[02:11]<vynvyv>ah so that's what you were talking about this morning
[02:12]<kjr>%X is the week of the year or something
[02:12]<vynvyv>oh
[02:12]<kjr>I never use it, because it's dumb
[02:12]<kjr>who reckons time by what week of the year it is
[02:12]<kjr>"oh shit, it's week 37! we're late to class!"
[02:13]<kjr>tieTYT: the good news is... I talked to the recruiter at Intel, the sponsor for the project (a fellow I guess) thinks I'm awesome
[02:13]<kjr>tieTYT: so he forwarded it to "the guy" who's doing the hiring
[02:13]<vynvyv>damn nice
[02:13]<kjr>I mean, I'm overqualified as all hell for the job....
[02:13]<kjr>tieTYT: I'm also trying to get on the core dev team for osCommerce if HPDL ever figures his shit out
[02:13]<vynvyv>hopefully they'll notice while you're working
[02:14]<kjr>tieTYT: well, hopefully Ill get out of my contract around december and end up working w/ WalMart's IT dept in SF
[02:14]<kjr>WalMart is offering me 10-20k a year more, and it's just about as big in terms of computing as any given country
[02:15]<kjr>but Intel on my resume... worth its weight in gold
[02:15]<vynvyv>how often do you switch jobs?
[02:16]<kjr>at this point? just about every 6 months
[02:16]<kjr>but the offers get progressively better
[02:16]<kjr>tieTYT: also, my employers KNOW I'm underpaid and choose to do nothing about it, and I've got a concise case to present to head hunters et al
[02:17]<kjr>$10.50/hr is fucking sad
[02:17]<rxrcr_>intel weighs alot too
[02:17]<vynvyv>wow
[02:17]<vynvyv>yeah that is sad
[02:17]<vynvyv>and what type of work are you doing at that pay?
[02:18]<kjr>tieTYT: all kinds of shit...
[02:18]<kjr>tieTYT: I've written a lot of awesome software while here
[02:18]<vynvyv>dude my friend at the library makes that much
[02:18]<vynvyv>and he doesn't do shit
[02:18]<vynvyv>he's not even a librarian
[02:18]<wyfaj>might as well go work for mcdonalds and flip some burgers for that money ;)
[02:18]<kjr>tieTYT: I'm retooling my RedStamp tool







