| |
| |
| |
|
Page: 1 2 3 4 5 6
Comments:
<0> Halo_Four: i dont see why not
<1> Halo: If you live in scotland.
<2> Scotland has a lot of sheep
<2> 12m, i think
<2> Beautiful landscapes, tho. Glaciers ripped a very interesting landscape when they receded.
<3> Darken, sure you can spare 12 sheep? what with the brothel and all
<4> Anyone know if there are any tools to extract all of the information from SQL Server 2005 to generate documentation?
<5> dfworking i'm only the manager for your father
<5> it isn't my concern where the sheep go
<3> no, Darken, i am your father
<5> :o
<3> Halo_Four never told you what happened to your father
<6> strategy-, i wrote a tool once
<6> its not hard
<0> just look at syscomments
<6> not even
<6> but thats a help
<7> SicLuDe: how would the join statement look?
<6> u need to analyze a few tables if u wanna do it poperly
<0> peano: ur taking too long to answer and i have the memory of an elephant so i even forgot what you wnated to do
<0> i think they call it add or something
<7> SicWork: sorry
<7> SicLuDe: let's say i have a column of the following integer data: 1,2,3,4,5,6,...
<0> that represnt times?
<7> SicLuDe: i want to simultaniously select 1,2,3,4,5... and 2,3,4,5,6,... in that order
<7> SicLuDe: sure
<0> how do u wnat the output?
<7> its just an offset
<7> SicLuDe: two columns are ok
<6> as a table peano?
<7> yes
<6> lol
<6> i still dont know what u want to do
<0> select (select data from datatable order by data) as data1, (select data from datatable where data != 1 order by data)
<0> something like that?
<0> :)
<7> hmm
<7> that could work
<7> thnx
<0> are you seirous?
<6> wtf
<7> sure
<0> lol ok
<0> each to his own
<5> no
<5> "to each his own"
<3> no
<0> i was throwing out a trivial solution in more of a mocking fashion but ok
<3> "one for you, two for me"
<5> niglet don't no me
<5> :/
<7> SELECT own WHERE each = his
<0> dfworking: i like "one for me, one for me"
<6> peano
<6> u in no position to talk
<6> :P
<6> wtf was that anyways
<7> what are you talking about
<7> this is mathematicly correct
<7> select data from table where data != first record, select data from table where data != last_record
<5> wtf?
<5> no man
<0> Darken: he is happy with it
<0> let it be :)
<5> w/e
<5> his happiness is my pain
<8> hehe
<8> darken sup
<7> Darken: what's wrong with it
<7> here is my data: 1,2,3,4. The query gives: 1,2,3 as column1 and 2,3,4 as column 2
<7> day 1, day 2
<7> day 2, day 3
<7> day 3, day 4
<7> then you divide column2 / column1
<7> that gives the return
<7> daily return
<7> plus 1
<3> i don't think a time series fluctuation can be done quite so simply
<6> i would rather do it the safer way and create a cursor and iterate the data set
<3> i recall seeing examples of it in a book
<6> dfworking correct
<6> ^^
<7> hm
<7> dfworking: any idea which book?
<3> peano, hold and i'll try to find an example
<0> well u dont need a cursor per say
<7> thanks
<3> yes, guru's guide to tsql, let me sift the cd code
<3> couple mins
<0> sql server 2005 has recursive quries
<6> SicLuDe, its safer
<6> SicLuDe, not backwards compatable enough for my liking
<0> voidzilla: then you would never use any new features
<6> correct, cos i have customer still running sql server 2000
<7> this will fail if there are duplicate dates in the table
<0> peano: what will fail?
<6> sadly we cant *force* them to spend a few thousand dollars upgrading. oh how i'd love it tho.
<0> voidzilla: thats your special circumstance
<0> he said he is using 2005 and didnt mention any backwards compatibility
<6> true true
<0> although i dont know too much about 2005 :(.. sadly i switched jobs right when 2005 came out and now i deal with oracle
<7> SicLuDe: select data from table where data != last_record, select data from table where data != first_record
<0> o i miss mssql server
<6> ye
<0> peano: dude u never gave any requirements
<6> oracle can be an unecessary pain in the balls
<0> i still dont know what u are really tryign to do
<7> SicLuDe: do you know what a return series is in finance?
<0> u asked a question and i answered it
<0> voidzilla: the main thing i miss are the client tools.. i love tracer
<7> SicLuDe: i have stock price data. i have to turn the daily price data into daily returns data
<7> SicLuDe: the daily return is simplyt he percent change today from yesterday
<7> it is the percent profit / loss you would have made had you bought yesterday and sold today
<0> so the formula u gave is wrong
<7> P(t)/P(t-1) - 1
<7> thats the formula
<7> if P(t) is the price at time t
<3> peano: http://www.coldfugitive.com/paste/?post=1881
<0> u didnt give the -1 before
<7> SicLuDe: heh im lazy
<7> dfworking: thanks a lot
<0> dfworking: that seems to ***ume that each date will be in the system
<3> i don't think i've ever really thought about it or considered using those types of queries, i just recalled seeing them
<0> atleast the first one
<0> didnt look at all the other ones
<0> this is very similar to doing running totals
<0> i normally just use a temp table to do that
<0> because we cant garuntee all the possible dates being there
<0> (yes i cant spell)
<3> here, for anyone thats interested, here is the 'cd' from the book with various code samples, good reference: http://www.websolete.com/misc/files/GG_TSQL.iso.zip
<3> its an iso and can be burned or opened in winrar for example
<3> peano, fwiw, the code i pasted is from ch09
<0> dfworking: thats only one part of a multi part zip
<7> thnaks
<0> or atleast thats what winzip claims
<3> don't use winzip
<3> the file is an .iso
<3> i had to rename for the server to allow downloads
<0> ah ok
<3> rename to .iso, and use a decent decompressor, like winrar to simply extract the files
<4> is there an index explaining what the various samples contain?
<3> need to buy the book for that, methinks :)
<8> mount ISO!
<3> guru's guide to tsql i think is one of the best sql books out there, the author doesn't **** around and shows very worthwhile things. he's like, you need to do this? here's how you do it. and here's another way. and here's the best way.
<3> he doesn't waste anyone's time with jokes and stupid ****ing stories and ****
<3> for some reason, writers of technical books simply don't realize that not only are they not funny, they make themselves look even stupider for trying to be
<9> Any of you guys ever heard of or used Neverfail for Windows/SQL Server?
<1> df: I hate it when tech writers feel compelled to put a giant photo of themselves on the cover.
<4> someone doesn't like Wrox books...
<3> heh
Return to
#sql
or
Go to some related
logs:
dizee mikel22 efnet
bout-disk.
cache:_8gpPI3ixKgJ:www.quotesdb.info/efnet/firebird/01Mar2006/1.html www.newsfil
#sex
#stocks
#computers
#flash
opengl rotated math
.figlet mirc
#networking
|
|