| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Comments:
<will[werk]> !tell us about joins
<SQL> will[werk] asked me to tell you this: Very basic join information: http://hashmysql.org/index.php?title=Introduction_to_Joins - For more indepth info: http://dev.mysql.com/tech-resources/articles/mysql-db-design-ch5.pdf
<Narada> Hehe very quick with that will. :)
<Narada> Oh wow a mysql wiki. Nice.
<will[werk]> I have a blackbelt in bot-fu
<Narada> Lol.
<Narada> Fear will.
<will[werk]> hi ya!
<MetaMorfoziS> hi all
<Jax> http://bugs.mysql.com/bug.php?id=16553 we need
<Jax> ;)~
<MetaMorfoziS> anybody on windows?
<Jax> yeah
<MetaMorfoziS> please anybody check me this with ie: http://syschat.hu/proba.html
<will[werk]> Window users will be banned.
<MetaMorfoziS> please check the smile's background
<MetaMorfoziS> is it transparent?
<eleftherios> windows? are there people on freenode who actually use windows??
<MetaMorfoziS> will[werk]: good idea
<Jax> MetaMorfoziS eh which smiley?
<tld> Could anyone give me a quick hand to figure out if I understand MySQLs index useage correctly?
<tld> I have an ID that's almost unique, but not entirely. I need to store it, because I'll be running queries against it. Large number of rows, so I need an index to make retrival fast. I do however also want an entirely unique ID. Would it make sense to add one column for the first ID, then augment that with more randomness in a second column, and create a single index for both of them? (usuallyUnique, alwaysUnique)?
<MetaMorfoziS> jax, http://metamorfozis.hu/p/biztosvoltam.png
<Jax> MetaMorfoziS give me the link to the png
<MetaMorfoziS> but in linux, i think the wine based explorer's png trasparency isn't supported
<MetaMorfoziS> http://e107.org/e107_images/emotes/MSN_Crystal/dunno.png
<Jax> well IE opens it in quicktime :D
<MetaMorfoziS> don't care:)
<Narada> Xgc: Wow that query is so much more elegant. Thanks for that. I love it.
<MetaMorfoziS> quicktime?:DD
<MetaMorfoziS> lol.
<Jax> works in FF, but not in IE
<Jax> IE->black
<MetaMorfoziS> lol
<MetaMorfoziS> ie's alphalayer support is a big big ****
<Nathalas> mmkay
<MetaMorfoziS> thank you
<Nathalas> http://pastebin.com/654354
<MetaMorfoziS> :)
<Jax> this Design by Contract stuff is killing me >;/
<MetaMorfoziS> fcuk fcukfcuk
<craigtmackenzie> can anyone help me please? i need to "uninstall" mysql from my mac, i don't know how
<MetaMorfoziS> Jax it's a only a probe
<Nathalas> It seems as though, if I do run the %va_% search, then every following query that involves a wildcard in the beginning runs considerably faster
<will[werk]> Nathalas, Could be because MySQL cached all the data. Bad thing to use % in front though.
<MetaMorfoziS> ie fcuk.
<will[werk]> MetaMorfoziS, Calm down.
<Nathalas> I know, but there must be a way
<Nathalas> cant you for example reverse the indices somehow?
<will[werk]> Nathalas, What are you doing?
<Nathalas> so they work the other way? :o
<will[werk]> heh
<Nathalas> Letting people search a list of albums, they need to be able to search like %busta%rhymes% without it having to take 5 minutes
<will[werk]> Nathalas, Use a full text index.
<enygma> fulltext searching
<will[werk]> Either MySQL's or look up lucerne
<Nathalas> now I've let it run through with %VA_%, %VA-% took 1.4s
<Nathalas> so maybe an alternative solution is just to let it run through that once when I boot up
<Narada> Weird though. The old query was faster it seems marginally :(
<sjrussel> uh no
<sjrussel> you can't count on the caching like that
<sjrussel> a bad query is still a bad query
<Nathalas> Ack.
<sjrussel> all you've done is gotten the OS to cache the entire table in memory
<sjrussel> if you want that, put in in a ENGINE=MEMORY table
<sjrussel> then, wait to run out of memory :)
<will[werk]> How about just index it properly? :)
<craigtmackenzie> can anyone help me please? i need to "uninstall" mysql from my mac, i don't know how
<sjrussel> LIKE % no usey no indexy
<Nathalas> if I use fulltext indices I need to restructure my queryes ;oP
<will[werk]> sjrussel, Fulltext :)
<Nathalas> queries*
<will[werk]> Nathalas, Start rewriting.
<Nathalas> You're killing me here.
<sjrussel> gentlement, start your editors
<will[werk]> Should have done it right the first time.
<Nathalas> Cant win every time
<will[werk]> sjrussel, .....
<sjrussel> ... unless you're Chuck Norris ?
<will[werk]> Who?
<MetaMorfoziS> anybody know a normal smile pack?
<MetaMorfoziS> that alphalayer works on ie?:)
<will[werk]> MetaMorfoziS, Ask in #web
<sjrussel> use GIF
<MetaMorfoziS> :))
<sjrussel> and get some treadmill time, you've got a double-chin growing there
<will[werk]> Are you calling me fat?
<sjrussel> pretty soon will won't find you attractive anymore
<sjrussel> no. Just shapely
<will[werk]> When did you?
<will[werk]> You mean, I'm attractive now?
<sjrussel> ???
<sjrussel> are you channelling MetaMorfoziS?
<will[werk]> I'll channel him out of here!
<MetaMorfoziS> what?
<Xgc> Nathalas: Your reference to queries beginning with '*' or not was why your original comments didn't add up.
<MetaMorfoziS> if you think to /amsg no:)
<will[werk]> amsg?
<Nathalas> Used to * being wildcard ;oP
<Nathalas> didnt realize my mistake til you brought it up.
<will[werk]> % looks like a better wildcard.
<Nathalas> I dont really care about how it looks :P
<will[werk]> oh
<sjrussel> that's too bad. Things that work well usually look better
<will[werk]> Things that look better usually work well.
<will[werk]> Reading up on HEAP/Memory tables. Manual says "Deleted rows are put in a linked list...". What does that mean?
<will[werk]> Deleted rows are tagged deleted?
<sjrussel> it means the memory isn't freed
<sjrussel> and it's reused on the next insert
<sjrussel> MyISAM does the same thing
<will[werk]> Is there a way for it to free it? Since it doesn't seem to want to OPTIMIZE.
<sjrussel> is it wasting alot of space?
<Narada> Is there a quick way of dropping all stored procedures for a database?
<will[werk]> Well, I'm doing testing, so putting in several MB of data in it.
<will[werk]> Ah, I can only truncate.
<will[werk]> Interesting. "MEMORY tables use a fixed-length row storage format." From manual. "session_header HEAP 9 Dynamic" from show table status....
<sjrussel> I think it's not true anymore
<will[werk]> Must not be.. but I'm running 4.1.16
<jjb_lyceum> is there a way to specify a blank p***word at the command line
<jjb_lyceum> mysql -uroot -p''
<jjb_lyceum> will ask me for a p***word.
<Xgc> jjb_lyceum: Blank represents no p***word.
<Xgc> jjb_lyceum: Don't specify the -p option.
<Paleo> don't put -p
<Paleo> eheh
<will[werk]> Don't specify the -p option.
<jjb_lyceum> presto!
<jjb_lyceum> thanks...
<Xgc> jjb_lyceum: Really. Just say no to p***words.
<jjb_lyceum> overall?
<will[werk]> heh
<adamc> hello
<adamc> is the date field in mysql m/d/y?
<will[werk]> Y-M-D
<will[werk]> !m adamc date type
<SQL> adamc: (Column Type Storage Requirements) : http://dev.mysql.com/doc/mysql/en/Storage_requirements.html
<adamc> how can I change this?
<will[werk]> You don't.
<adamc> hmm
<inviso> !m adamc date time functions
<SQL> adamc: (Date and Time Functions) : http://dev.mysql.com/doc/mysql/en/Date_and_time_functions.html
<inviso> adamc: see to_date, iirc
<MetaMorfoziS> adamc
<MetaMorfoziS> the easiest way if you use uunix timestamp
<MetaMorfoziS> time()
Return to
#mysql
or
Go to some related
logs:
stopping apache instances
845GM Fedora
#python
ubuntu what is build essential
wiki scrot 1024x768
#oe
k3b burn daa
Error fsck.ntfs
intel brookdal
#oe
|
|
