#mysql
07 October 2007
Total 21 pages. You are browsing page 2/21.
First :: Prev :: [1] [2] [3] [4] [5] [...] :: Next :: Last
01:23 <****> select css_options.name, sections.name, css.value from css_options inner join sections left outer join css on (sections.name = css.section and css_options.name = css.name) where sections.name = 'properties';
01:24 <****> probably dizzying
01:25 <****> I like FROM tableA AS a INNER JOIN tableB AS b ON a.name = b.section LEFT JOIN tableC AS c ON b.name = c.name WHERE ...
01:25 <****> I haven't used your syntax before... but if it works cool!
01:27 <****> huh.
01:27 <****> sounds fun
01:34 <****> rdm2: Bored of here yet?
01:41 <****> not just yet
01:42 <****> the odds are better here than efnet.. you and your bots n' all
01:42 <****> Wow, efnet that bad?
01:42 <****> man, it's like crickets there
01:43 <****> #sql is full of a-holes
01:43 <****> It's not like that here?
01:43 <****> Freenode's #sql?
01:43 <****> nah efnet
01:43 <****> oh
01:44 <****> sql used to be good
01:44 <****> they didnt like mysql though
01:44 <****> I need to calculate the number of times that each word of a title is found in a row of the database
01:45 <****> is there a faster way to do this than to do a binary search on each word individually
01:45 <****> or a way to speed up the binary searches?
01:45 <****> uh yah.... I mentioned mysql and the flipped
01:46 <****> mysql aint so bad.. sure, it's not oracle or postgres, but for some applications it's the best tool for the job
01:53 <****> rdm2: You'll find people are somewhat afraid of things they don't know about. Instead of commenting intelligently, they'll get mad or (on efnet) ban you because they don't want to be questioned about something they've said that they can't back up.
01:53 <****> Is there any way in mysql to dump the structure (but not the contents) of a database?
01:54 <****> Xgc: makes sense... seems like ego's prevail and knowledge has dwindled on efnet
01:54 <****> postgres isnt oracle... but a website doesnt need oracle - or postgres
01:54 <****> exactly
01:54 <****> some of mine run just fine on sqlite
01:55 <****> but i did do an entire application run from inside oracle
01:55 <****> the pl/sql toolkit
01:55 <****> it's probably built up on fear like Xgc said.. They fear that mysql is becoming more and more popular and getting more and more features that they can't debate that it's not a serious database .
01:55 <****> na
01:56 <****> couple years ago mysql didnt qualify as "proper sql"
01:56 <****> right
01:56 <****> or rather as a proper RDBMS
01:56 <****> still doesnt
01:56 <****> you only get transactions with innodb, isnt it?
01:56 <****> It's been a "proper" RDBMS as far as I'm concerned for years, but its featureset *is* different from postgresql's or oracle's.
01:57 <****> rdm2: While there are smart people on efnet, many of the chan-ops are not and instead of shutting up if they don't know, prefer to blather non-sense (or things they've heard from years ago) and then ban anyone that doesn't agree.
01:57 <****> that too :)
01:57 <****> i'm just glad to have found another option ...
01:58 <****> google is great.... documentation is great.... but sometimes after working on a problem for a few hours, I start to give up.
01:58 <****> efnet was just out of control. It's too bad.
01:58 <****> i remember people banning you for using bitchx because "its a bad script" - nobody bothered to look it up and see its a fullfeatured client, not a script
01:58 <****> rdm2 on mysql?
01:59 <****> ya
01:59 <****> I've used postgres mostly in the past... but mysql now
01:59 <****> rdm2, why?
01:59 <****> if you ask nicely, google will answer most questions :)
01:59 <****> why did I switch?
01:59 <****> heh.. google even answers questions when I'm in a really foul mood. rdm2, yes, why switch?
02:00 <****> i'm doing a website, either postgres or mysql would suffice ... nothing heavy, and mysql was already installed on the box i'm working on...
02:00 <****> so, went with mysql.
02:00 <****> Ah.
02:01 <****> but when I choose to install a database, I install postgres
02:01 <****> I thought maybe it'd have something to do with faster seek times.
02:01 <****> Can you say why you prefer postgres?
02:01 <****> postgres is slower with seek times for websites.
02:01 * sauvin isn't going to be doing any websites within any lifetime soon, but is trying to hassle out a pile of mp3s
02:02 <****> postgres is slower?
02:02 <****> you sure?
02:02 <****> slower for less data
02:02 <****> Trengo, for some things, this is what I usually hear, yes.
02:02 <****> postgres is faster if you have large amounts of data
02:02 <****> What's "large" versus "small"? Where's the boundary, roughly?
02:02 <****> an http response takes a couple seconds
02:03 <****> the difference is completely diluted
02:03 <****> you cant tell the difference
02:03 <****> or at least i cant
02:03 <****> Well, I was working on a team that had several hundreds of gigabytes of data..
02:03 <****> it was postgres
02:03 <****> Trengo, the difference might not be so diluted if you're talking about handling several thousand queries a minute.
02:03 <****> sauvin in a website??
02:03 <****> Yup!
02:04 <****> just throw more HW at it
02:04 <****> hah
02:04 <****> more cores.
02:04 <****> I don't think HW can solve all problems.
02:04 <****> both mysql and postgres implement clusters
02:04 <****> though i havent played with pg clusters yet
02:04 <****> THAT, I didn't know. Last I'd heard, posgresql didn't cluster.
02:04 <****> i think so
02:05 <****> (prokofiev - concerto No. 3 in C Major - GOTTA give it a listen!)
02:05 <****> is there anyway to do non-exact matching query .. something that would get me words that might be mis-spelled in database ?
02:05 <****> One of the major reasons I initially was drawn to postgresql was server-side perl.
02:06 <****> ramy, that might depend on how badly misspelled - could you search for fragments rather than approximations?
02:06 <****> sauvin: well, there is a hih probablity if a single letter missing or mis-spelled
02:06 <****> s/hih/high
02:07 <****> In that case, you might consider searching on "hi" or "probab".
02:07 * sauvin hides
02:07 <****> I don't remember pgcluster is a multi-master "cluster"
02:07 <****> I'm assuming mysql has a syntax similar to postgresql's "like '%something%'"
02:08 <****> sauvin: I want something generic that enhances a searching of user supplied phrases
02:10 <****> ramy, what you're asking, if implemented properly, would probably be in the realm of artificial intelligence.
02:10 <****> What kind of db are you working with?
02:10 <****> mysql
02:10 <****> Um, I meant, what kind of application are you creating?
02:10 <****> nope it doesnt, i remember seeing something about hapostgres, but was my mistake, sorry
02:14 <****> What is the most reliable way of getting a unique id, eg last_insert_id from a table etc, so I can use that in other tables etc as say an unique order number.
Total 21 pages. You are browsing page 2/21.
First :: Prev :: [1] [2] [3] [4] [5] [...] :: Next :: Last