| |
| |
| |
|
Page: 1 2 3 4 5 6
Comments:
<0> sure
<1> Gm0nEy : which version?
<2> SQL 2000
<1> Gm0nEy : yeah - select * from <table> will report all the rows
<1> what more could you want?
<2> lol
<3> What kind of "reporting features"?
<4> you can use MS reporting services on top of sql 2000
<4> as lng as you can stomach that whole .net arena in designing the reports
<5> CoJones: is there an equivalent command to minus in access sql?
<4> - ?
<5> I have two views, identical fields.. how do I get View1 minus View2's records in View1? using access 2003
<6> can you do a left join in access?
<5> i believe so..
<5> ive tried all sorts of joins and i always get a cartesian product of records
<6> heh
<5> View1 is all the records.. View2 contains all the records I don't want in View2
<4> then youre not trying the joins properly
<5> obviously..
<6> a LEFT JOIN b ON a.id = b.a_id WHERE b.a_id IS NULL
<4> so did you try reading about joins ?
<5> ah, thank you upb
<5> you'd have no idea ive taken to db courses in my life..
<5> *2
<4> nop, we wouldnt
<5> i guess it all depends on what you choose to specialize in
<6> why didi you take the courses then?:P
<6> bad excuse
<5> i didnt have a choice
<5> part of comp eng and comp sci curriculum's
<4> a better excuse woulda been 'im a bit crap and didnt pay attention to any of it'
<7> just because you don't specialize in something doesnt mean you avoid it like the plague
<5> i agree, but ive gotten by on just basic sql knowledge
<5> till today
<4> thats arguable ;-)
<7> hehe
<4> left join isnt really that hardcore
<6> hehe like some guy told he had 10+ tables in his schema and didnt use joins ever
<4> lemme guess, one for each of his 10 customers ?
<6> what he did was iterate over the resultset in php and do another query for each row :P
<6> it was on this chan btw ;P
<4> :~(
<5> only db's ive used out of the cl*** room is for simple web apps.. i never really did grap the whole right/left/inner/outer joins.. just simple primary/foreign key matching
<4> well, do yourself a favour
<4> spend an hour learning them cos thats all it takes
<4> and then a db opens itself up
<5> yeah.. i will
<7> there are probably some good join tutorials on the net
<5> this ones pretty good but it just tells you the bare essentials of what each type of join does: http://www.devx.com/dbzone/Article/17403/0/page/1
<5> didnt help me with a Set A minus SubSet B
<4> depends if you want the answer on a plate, or if youre willing to apply what youve learnt
<8> heloo
<5> yes.. i was at fault
<8> it seems
<5> hey, you can't flog me too
<9> dynamic ORDER BY without a case statement in the lastest version of MSSQL, has anyone done it?
<4> oh not you again
<9> inside a user defined function heh
<9> yeah I have to look at it again
<4> yeah, and remember last time ?
<9> I remember somewhat, been working a lot lately
<4> we discussed it for wayyy too long without you ever saying why CASE is not acceptable
<4> its not possible. use CASE.
<4> case closed.
<9> oh because I don't know what I'll be ordering by
<4> then use CASE
<9> if I can't enumerate what I'll be ordering by, how could I use case?
<4> you dont understand what CASE does, do you?
<9> yes, I understand what a case statement is
<4> tell me
<9> CASE @this
<9> WHEN 'A' THEN blah end
<4> ok great
<4> but you say
<9> however, I don't know what @this might be
<4> ?
<9> so how can I ***ign whens for it
<4> you dont know its range of values ?
<9> no, because @this needs to be dynamic
<4> why?
<4> you have infinite fields in your table ?
<9> no, but an extremely large set that would be next to impossible to maintain in a CASE
<9> I use a CASE now, but it's just getting way too big
<4> get used to it, thats sql im afraid
<4> a big query doesnt mean its a bad query
<9> I know, I was hoping some SQL hacker had figured out a way around it
<9> without using SQL 2005
<4> ok, well without 2005 im fairly confident in saying the only way is a big case
<4> or call diff procs depending on the order by field, but again youre just using a case to decide which to call
<9> 2005 isn't even service packed yet, we can't even pretend we're going to use it or we'd lose customers
<4> however you slant it, you need an IF THEN ELSE
<4> i dont use 2005 yet either
<9> I just can't do that for the amount of fields, sigh, this ****s
<4> well maybe go back to the drawing board and rethink your design
<4> why cant you anyway, once its written its written
<4> its not like you gotta write it 1000 times
<9> yeah, we were just doing that even though it's ****load of work when you have 1200 sps and udfs
<4> maybe youd be better sorting it client side then?
<9> not a possibility for us =(
<9> oh well, thanks for the hel Carpathia
<4> :(
<9> err help
<4> theres no other way, what can i say :)
<9> yeah heh
<4> id say sorry, but i didnt design sql ;P
<9> be back soon I'm sure, adios
<4> heh
<6> what is he trying to do?:P
<4> some bs
<6> heh
<4> order by an arbitrary field without using a CASE
<6> ah
<4> time for him to call Gandalf i think
<6> and he has 1000 fields?:P
<4> well, id take that with a pinch of salt
<4> more like 10
<6> lol
<9> I'm back already hah
<4> hey thats no joke
<9> What about creating tables with duplicate columns and then using the columns to order?
<9> duplicate sorted columns
<4> ?
<9> #temp tables
<4> what?
<9> this is all inside a user definied function
<4> UDFs cant create a temp table
<7> VRik what is your scenario here
<9> attempting dynamic order by
<7> broader
<9> can't use case statement
<9> ok
<4> no, be honest, wont, not cant
<9> we have a udf that gets ids depending on what they specify in search parameters. Right now we use CASE statements in our ORDER BY to accomplish the sorting, as in CASE @this WHEN 'A' THEN blah.... END
<4> ok, and what problem is that causing ?
<9> The problem is that we are starting to need more and more possibilities for @this
<7> VRik so this is a search engine?
<4> how about p*** @this2 ?
<7> where the user may want to order by different columns
<4> or make @this an int instead of a char
<9> yeah, call it a search engine
<9> right and that number of columns is growing much larger
<4> ok so lets say you find a way without case
<7> what's wrong with adding more cases?
<4> exactly how do you envision it working ?
<4> sooner or later you gotta make some decision base don a param
<4> ie, an IF
<9> tenfour : too many column possibilites makes it crazy to maintain
<4> ie, a CASE
<7> somewhere in your logic you are going to have to branch on @this
<7> so logically there's no problem with putting it in a case - that's exactly what it's built for
<4> i think youre not really telling us the real problem
Return to
#sql
or
Go to some related
logs:
1680x1050 wallpaper
#beginner
unable to initialize Ntbackup
#mirc
#politics
#windows
#freebsd
#freebsd
#debian
#fedora
|
|