| |
| |
| |
|
Page: 1 2 3 4
Comments:
<0> www.connectionstrings.com
<1> yes I'm checking that now
<2> lol
<3> Quant, You blundering bag of rotten bile
<3> Quant: heh i think friday is my last day
<1> i'm not sure which one to use. I have SQL Express server installed on the same machine in which I'm developing
<3> :)
<1> winxp Home PC
<2> haha
<2> monday is my first
<3> ya i know
<3> :)
<3> u drove me out
<3> well they are still fighting to keep me here till next friday
<2> what difference will that make?
<4> argh... if I have a many to many related set of tables, shouldn't I be able to use LEFT JOIN to see a record from each, mathcing or not?
<4> matching, that is
<2> umm what
<2> cross join
<2> not sure what you are asking :)
<3> one week?
<3> 300 an hour
<3> 40 hours
<3> 12000 ?
<3> EyePulp: outer join?
<3> a full outer join
<4> that's what I'm using now... not correctly apparently.
<3> u said ur using left jion
<4> I moved to FULL after quant said "cross join"
<3> well cross isnt hte same as full :)
<4> That's probably not explaining my problem well.
<3> as far as set notation goes... cross will give you |A| * |B| and full will give you |A U B|
<3> the number of rows that is
<4> two tables, a & b. they both have a field "state" for example. I'm trying to find all instances of table b that has no matching state to a. Which I thought a left outer would handle.
<3> yes
<3> select b.states from b left join a on b.stateid = a.stateid where a.stateid is null
<3> that will give everything in A that is not in B
<3> sorry
<3> other way
<4> right
<3> everything in B that is not in A
<3> btw.. i havent seen what ur query looksl ike.. but ive just seen like 3 people make this mistake in the last 2 days... try and move some of the where clauses into the join clause
<4> in the ON portion?
<3> yes
<4> I've seen that. It loks confusing somehow.
<4> looks
<3> so if u had... select ... from a left join b on a.id = b.id where b.blah = 10
<3> that will essenctially make it an inner join
<3> because it will ignore everything that has b.something is null
<3> because null != null
<3> so if u change that query to: select ... from a left join b on a.id = b.id and b.blah = 10
<3> that will work
<4> so drop the WHERE in some cases... that might be the problem.
<3> yes
<3> if you are using a where clause on anything on the table on the right (a left join b - b is on the right) then consider what you are doing with it
<4> uh oh... 18+ seconds and still running.
<3> lol
<4> hrm...
<3> did u manage to somehow do a cross join?
<3> heh: select * from sysobjects, sysobjects - that query takes over 2 mins on my mahcine :)
<3> heh sql server doesnt quite optimize that on
<3> one
<4> hah
<4> MS is teh sux0r, obviously
<3> aite im going home
<3> EyePulp: i should be there on the 22nd
<4> dammit. thx for your help.
<3> well ill be on IRC again in like 30 mins
<4> cool. keep me informed.
<3> so ill check up on u there :)
<4> heh
<5> Hi, im using SQL 2000. In a table i have a smallint column to indicate something. But only one record in this table can have the value set to 1. How can i create a trigger to avoid more than one record have value = 1 ?
<6> EyePulp: so howd that work out?
<7> Greetings all.
<6> asd21: msging people is the worst way to get help
<6> what did u like msg everyone?
<6> or was i just a lucky one that u picked?
<6> :)
<7> I'm working on a query in MS-SQL (2000) that I have built a procedure around, which gets pulled by a SharePoint page. Here it is: http://sql-servers.com/nopaste/?show=650
<5> hehe
<5> sorry, SicWork
<7> If someone can tell me how to *not show* the TrainAmt and SaleDate, but *do show* only the SUM amount at the bottom, I will be your friend. Officially.
<6> hmm got too many friends
<7> :-)
<8> queries are hard.
<7> Well, not if you know what you're doing I suppose. But I don't. So, yeah, they're tough.
<6> Ray_Haque: teh whole thing about compute is the fact that it generates multiple resultsets
<8> heh
<6> if you want hte sum... y odnt u just do a sum?
<6> i mean.. ur basically using compute and then defeating the purpose of compute
<5> SicWork, can you help me?
<7> Ah, so replace SELECT with SUM.
<6> asd21: do it in a stored proc or something
<6> Ray_Haque: how did u come about using compute but not know how to use sum?
<5> SicWork, but i want to generate an error if the user try to insert/update a record that have value = 1
<7> SicWork: I'm just pasting from examples. I don't know what I'nm doing. :-)
<6> asd21: you can do that through sp
<6> geez
<6> Ray_Haque: go read
<6> were not gonna give you sql 101
<7> Hah, I understand. Just fishing for clues.
<7> I think you gave me one.
<7> It's taken me a couple of hours to come up with that crappy kludge of a query. I'm quite proud.
<7> But SharePoint ****es upon it.
<7> Well thanks for the hint.
<6> np
<9> should multiple left joins be possible in Jet?
<9> I want to do SELECT master.key, sub1.c, sub2.c from master LEFT JOIN sub1 .. LEFT JOIN sub2
<9> (sub1 and sub2 are sub selects)
<9> http://sql-servers.com/nopaste/?show=652
<10> hi
<10> Anyone here certified with SQL Server 2005?
<11> hi all. looking for some help writing a query .. I have two tables. Table 1 has an ID column, table 2 has some information and a column that points to ID's in Table 1. Looking to get a recordset of the top entries in Table 1 that have references from Table 2. That is, the most popular referenced entries of Table 1. Anyone mind giving me a hand?
<9> onyx, more popular = more rows in table 2?
<11> right -- so, to put this in plainer terms.. Table 1 is called JournalEntries, table two is JournalComments
<9> select count(*), join tables on the ID, and then group by the ID
<11> the latter has a column that points to original entries in JournalEntries.. Looking to display the "most active" entries (those with the most comments, ordered highest to least). Really just looking for the top 10 or so
<12> yar
<9> order by count(*) desc, and limit 10..
<10> anyone know a book that will prepare me for SQL Server 2005 Certification?
<11> pardon my ignorance (god knows I'm pretty ignorant to SQL), but is LIMIT valid in SQL Server? more of a mysql thing, no?
<9> oh yeah, might be TOP in MS SQL maybe
<12> or maybe FIRST
<11> ok, let me give that a try. thanks much
<11> i think its TOP (at least thats what I usually use)
<9> in Access when I do SELECT a+b and either a or b is NULL it appears to return NULL, anyone know a function to convert NULL to zero?
<13> access is balls
<14> MSSQL 2000: Using Merge Replication, I'm keeping 2 databases synced, but it appears that when I do a batch of inserts (typically 12,000 - 30,000 at a time) I can see that only 100 inserts occur at a time (looking at the merge agent info in the replication manager). I'm also seeing that I'm averaging 65 rows/sec. Can either of these be controlled/managed?
<14> mmunky: ISNULL()
<14> or similar
<13> switch([whatever] is null, 0, [whatever] is not null, [whatever])
<13> only access could make nvl or coalesce more complex
<0> Access has Nz()
<9> thanks Halo_Four
<10> hi
<10> maybe someone can help me with something
<10> I am thinking of going for the SQL Server 2005 certification but I do not have any experience with SQL
<10> can I practice with mySQL or will it be different/wrong
<13> what makes you think you should go for "certification"?
<13> when obviously you do not know what you are doing
<10> ok, I will just ignore that
<10> thanks for your help
<13> people like you who memorize a book and try to get certifications are why certs are such big money makers
<13> eventually youll p***w ith enough retests
<13> and someone may be gullible enough to think it means something
<13> and youll trash their data
<10> ok
<10> I never said that I am going to take the certification now
<10> did I say that?
<10> I want to learn before I take the certification
Return to
#sql
or
Go to some related
logs:
nagafooki
#unixhelp
insm goldman
#hardware
imperialism and globalizatin
#windows
#gentoo
#politics
#heroin
#nhl
|
|