IRC log for #wowroster on 20080213

02:45.46*** join/#wowroster poetter_ (n=poetter@e177160195.adsl.alicedsl.de)
07:24.05*** join/#wowroster PleegWat|Work (n=ple@82-148-206-57.fiber.unet.nl)
07:24.05*** mode/#wowroster [+o PleegWat|Work] by ChanServ
07:44.22*** join/#wowroster PleegWat|Work (n=ple@82-148-206-57.fiber.unet.nl)
07:44.22*** mode/#wowroster [+o PleegWat|Work] by ChanServ
07:58.14*** join/#wowroster Tupsi (n=Miranda@217.7.101.202)
09:41.30*** join/#wowroster luciver[oFF] (n=luciver@h1082.serverkompetenz.net)
14:35.34*** join/#wowroster Zanix (n=Zanix@69.92.238.219)
14:35.34*** mode/#wowroster [+o Zanix] by ChanServ
14:58.39*** join/#wowroster Zanix|Work (n=Zanix@71.216.154.51)
14:58.39*** mode/#wowroster [+o Zanix|Work] by ChanServ
16:14.07*** join/#wowroster Nefuh (n=nefuh@dslb-084-062-243-067.pools.arcor-ip.net)
16:19.07*** join/#wowroster Gaxme (n=chatzill@64.247.200.42)
17:01.11*** join/#wowroster fewyn (i=fewyn@rrcs-24-73-143-79.se.biz.rr.com)
18:43.28*** join/#wowroster Zanix|Work (n=Zanix@71.216.154.51)
18:43.28*** mode/#wowroster [+o Zanix|Work] by ChanServ
18:46.44*** join/#wowroster PleegWat (n=PleegWat@212-123-189-114.ip.telfort.nl)
18:46.44*** mode/#wowroster [+o PleegWat] by ChanServ
20:26.05GaxmeI'm rusty. The difference between == and === for comparisons is what again?
20:27.09PleegWat=== checks type as well
20:27.25PleegWatEG (false == '') but not (false === '')
20:27.35GaxmeAhh, that's right. Thank you.
20:34.20Zanix|Workgood explanation
20:44.06GaxmeSooooo close to done with this darn backend.
20:44.21GaxmeJust need to finish all the logical checking for not duplicating entries in the db if they exist.
20:46.13Zanix|Workwoot
20:48.33GaxmeObviously I'm also redundant about my redundancy. Heh.
20:48.44Gaxme"duplicating entries...if they exist".
20:48.52GaxmePower to the people!
20:49.47Zanix|Workwelcome the the Department of Redundancy Department
20:50.07GaxmeThanks, glad to be here!
20:50.28GaxmeMan, some of this code is ugly, though.
20:50.32GaxmeBut it works.
20:50.40Zanix|Workhehe, get it to work first, then clean it up
20:50.44GaxmeAnd it profiles fairly well, so I guess I won't complain.
20:50.47Zanix|Workthats how I work
20:51.26GaxmeNot a bad plan. =)
20:51.51GaxmeBy the way, does the roster flush its mysql result records on normal queries, or do we need to call that?
20:52.18GaxmeI saw that one of them does, like query_first() or the like.
20:52.25Zanix|Workthats the only one
20:52.39GaxmeJust trying to do whatever I can to drop the potential leaks.
20:52.42Zanix|Workotherwise you should flush them
20:52.45GaxmeOk.
20:52.56Zanix|WorkI love query_first though
20:53.06GaxmeYeah, I've used it where I can.
20:53.15Zanix|WorkI think fetch all might flush too
20:53.16Gaxmefetch_all makes me happy also. =)
20:53.35GaxmeNow I just need transactional writes and I'm perfect.
20:53.45GaxmeI almost wrote them in, but I didn't want to do it from a code layer.
20:54.08GaxmeSince it really needs to make sure that on table creation the server's mysqld can handle transactional tables.
20:54.32Zanix|Workyou are starting to talk above my understanding...
20:54.33Zanix|Work:(
20:54.38GaxmeDoh, sorry. =)
20:54.41Zanix|Workits ok
20:54.48GaxmeI know enough to be dangerous, I think.
20:54.57Zanix|Workall I really know about mysql is select * from `crap`
20:55.11Zanix|Work...to get all the crap
20:55.12GaxmeUsually that's enough.
20:55.13Zanix|Worklol
20:55.32Zanix|WorkI have to copy another sql statement for more completity
20:55.40Zanix|Workand try to tailor it to my needs
20:55.42Gaxmetransactions mean I can start a transaction, enter a ton of sql commands in sequence, and if any one of them fails, I can roll back everything I did previously.
20:55.51Zanix|Workah
20:56.07GaxmeIt's a failsafe without having to do per-insert checking, and risking leaving orphaned data.
20:56.18Zanix|Workwell, if you ever write a patch for lib/dbal/mysql.php that doesnt break other functionality, then we can probably include it
20:56.30GaxmeI may look into that.
20:56.40GaxmeAs it stands, I've done it the old one-line way for this.
20:57.13GaxmeWhat I don't know how to do is check mysql versions, or whether it's even feasible to change to the transactional table type from the standard type.
20:57.21GaxmeI have NO idea if that completely zorks data.
20:58.10GaxmeThe only transactional databases I've worked with were built that way from the ground up.
20:58.28GaxmeThinking out loud, here.
20:58.43GaxmeThe db model would have to be able to check for the functionality on install.
20:59.14GaxmeAnd then create/change tables to that model. It really only sits at a db layer level until someone uses it.
20:59.25GaxmeSo the rest of the code could be changed in whatever time frame.
20:59.48GaxmeI wonder if doing temporary copies of the tables, destroying the old ones, and replacing them with transactional versions works.
20:59.56GaxmeI've never tried stuff like that. =)
21:00.32GaxmeAhh, still. One thing at a time. Finished addon, then base code changes. LOL
21:02.05PleegWatI'm the complicated sql guy for roster
21:02.09PleegWatwhat's the addon?
21:02.20Gaxmeraidtracker
21:02.27PleegWatah
21:02.30GaxmeThe extended version. =)
21:03.07GaxmeWith any luck I'll have a beta ready to go by the end of the weekend.
21:03.16GaxmeBackend is about done, then I have to build the front.
21:03.32PleegWatThe transaction thing reminds me. Something is breaking the addon install rollbacks.
21:03.39PleegWatBeen planning to look at it for weeks :(
21:03.45GaxmeNow's yer chance!
21:03.50GaxmeHeh
21:04.41Zanix|Workyay
21:04.55Zanix|WorkIve noticed that too PleegWat
21:04.56PleegWatah, damn.
21:05.00PleegWatdebug mode
21:05.04Zanix|Workjust forgot to say something about it
21:05.06PleegWatit always dies on queries
21:05.14PleegWatnow why did we put that in...
21:05.35PleegWatAny way to juggle it down into the php error box?
21:06.19PleegWatGet out of there and answer the question please?
21:06.26PleegWatYou wrote the error box part
21:09.41Zanix|Workmaybe
21:09.46Zanix|WorkId have to look at it
21:09.51PleegWatah, never mind
21:09.55Zanix|Worklol
21:09.57PleegWatI think I can just call the handler directly
21:10.08PleegWator trigger_error, but that's got a 1024 char limit and that's too low
21:10.25Zanix|Workyeah, you should be able to just call the handler
21:11.00Zanix|Workwell, die_quietly was ment to do just that
21:11.08Zanix|Workdie, but quietly
21:11.15PleegWatdie_quietly dies
21:11.22PleegWatCan't rollback after a die
21:11.29Zanix|Workindeed
21:11.56Zanix|Workroster_die is die quietly, without the debug
21:12.38Zanix|Workon the installer, are we just die'ing on errors?
21:13.03*** join/#wowroster fewyn (n=fewyn@46.210.33.65.cfl.res.rr.com)
21:13.21PleegWatmysql.php dies on db errors if debug mode is on
21:13.28PleegWat> dev
21:20.12GaxmeDid either of you use the old raidtracker?
21:21.37PleegWatno
21:22.12Gaxme*nod
21:24.35Zanix|Worknope
21:25.05Zanix|Workbut I noticed alot of frFR people do since thats the most active topic in the raidtracker sub-forum
21:26.28GaxmeYeah. Just trying to test my logic.
21:26.49GaxmeThe old version didn't collate data at all
21:26.56GaxmeIf the raid existed, it skipped it on upload.
21:27.15GaxmeSo two people running the tracker and uploading data meant that the first one to log off stored the data for the raid.
21:27.30GaxmeMeaning if more bosses were killed after that first person left, they were not recorded.
21:27.47GaxmeMore loots, etc.
21:28.23GaxmeSince bosses are affiliated with the raids, which get a unique raid ID, I'm pretty sure that I can simply check to see if the boss name has been uploaded in relation to that raid. If the name exists, it can be skipped.
21:28.44GaxmeBut I'm trying to think of examples where you might kill the same named boss mob on the same raid.
21:28.51GaxmeAt different times.
21:29.09GaxmeIE: Is it important to ID for collation by more than just the boss name.
21:29.18GaxmeSorry, just thinking outloud again.
21:31.29Zanix|Workits ok
21:31.34Zanix|Workwe do that same
21:32.02Zanix|Workit hink im gonna go home
21:32.09Zanix|Workits boring here
21:33.46GaxmeEnjoy home!
21:37.00Zanix|Workoh i will
21:37.02Zanix|Worklater
21:37.18Zanix|Workbrb
21:37.21Zanix|Work...at home
22:17.07GaxmeGonna head out myself.
22:17.11Gaxmebbl
22:21.06Zanixwow
22:21.27ZanixI really wish someone would have said "oh by the way, it snowed 4 freaking inches today"
22:21.38Zanixclear and sunny when I went to work
22:21.43Zanix40 degrees
22:22.10Zanixcold, cloudy, slightly windy and snowing with 3-4 inces accumulated when I go home
22:30.32PleegWatWe've had fog all day
22:32.12Zanixoh dang, tomorrow is valentines day
22:49.29*** join/#wowroster Nefuh (n=nefuh@dslb-084-062-243-067.pools.arcor-ip.net) [NETSPLIT VICTIM]
22:53.58*** join/#wowroster Zanix (n=Zanix@69.92.238.219) [NETSPLIT VICTIM]
22:53.58*** join/#wowroster luciver[oFF] (n=luciver@h1082.serverkompetenz.net)
22:53.58*** mode/#wowroster [+o Zanix] by irc.freenode.net
22:56.57*** join/#wowroster Nefuh (n=nefuh@dslb-084-062-243-067.pools.arcor-ip.net) [NETSPLIT VICTIM]

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.