| |
| |
| |
|
Page: 1 2 3
Comments:
<0> but I hate it so I just use green screen, and found that I could back out of one of the menus
<0> the main menu
<0> into the AS400's main menu
<1> hmmz?
<0> from there you can do basically anything on ANY user account be it a time sheet account, or an account for an entirely service that happens to be run on the AS400
<0> so anyone that has any kind of account (1000+ people) have access to HR, payroll, customer records, etc
<0> at least full read access
<0> probably full write
<0> but I'm not going to be testing that
<0> so since they wont let me have a real database on their servers (don't blame them) I linked my MS access db to their db to compare data and such
<0> handy as ****, saves me easily 8 hours a week
<0> to audit this PayCode thing I needed that query for, other depts pay an admin to sit and manually check each employee for several hours
<0> my favorite thing is how in out Food & Bev dept for example, they pay an Admin to do practically nothing but compare departmental hand-written time sheets against the company's time clocks to check for discrepancies
<0> 800 employees in that department, must take at least 16 hours
<0> makes me want to wear this shirt (if I could) http://www.thinkgeek.com/tshirts/frustrations/374d/
<0> thanks for that query btw, EXACTLY what I needed
<2> thanks for the info. that was fun to read
<2> :)
<0> um, plz don't h4x0r my work 8(
<3> ?
<0> If I ever know I'm either definately going to lose my job for another reason or I have 95% job security, I'd like to try to write to a terminated employees paid-time-off field just to see if I could
<0> or better yet, pay rate
<4> so you turn being fired into felony wage theft
<0> no, I wouldn't do it to my own record
<0> I'm just curious if the security is really THAT bad
<0> I'm an ethical person and all, I've never looked at anyone pay or anything other than what I need to make my job easier
<3> k i got this mdb file it has 97,000 rows in it I am guessing it was probably made with sql not access but i only have access 97 .. i installed openoffice to see if i could read the file and i can but I cant copy the table to the spreadsheet ap because there is a 65k row limit anyone have an idea how i can open this file and save it to csv i know some of the fields have commas in them too
<0> a ghetto solution I can think of, use VBA in excel along with an ODBC connection to read the data and manually write it to a text file
<4> cHiKNPad, You only need the access odbc driver. and then you can use the sql or query tool in excel to connect to it
<0> which can also take care of your comma problem
<0> since you can use replace()
<4> or you can download a trial of MS office with access
<4> for 30 days
<3> well my excel is old too its 97 and has the same limit 65k rows
<4> get the office 2003 trial
<3> yea maybe the trial
<0> cHiKNPad:
<0> I don't mean write to excel
<0> I mean use VBA to write to a text file
<0> you could VBA in outlook for all it matters
<3> maybe
<0> like I said, ghetto, but there if you run out of options
<3> hmm
<3> wonder if i could do it with php
<3> or perl
<4> yes, with odbc
<4> same as you can with excel directly
<4> theres a thousand ways to do this
<3> i have never used it for writing to a file just for doing stuff inside a file
<3> i will look it up
<0> I'll write some psuedo code for you real quick
<0> http://sql-servers.com/nopaste/?show=140
<0> there, that should do exactly what you need
<0> (if those functions/methods work with excel 97)
<3> o i will try thanks
<0> that's loaded with typos btw
<0> and possible logic errors
<0> but that's basically it
<0> http://sql-servers.com/nopaste/?show=141
<0> fixed
<0> err, I know you're pry just going to download the demo, but I fixed it for real this time
<0> http://sql-servers.com/nopaste/?show=142
<0> and I'm out, thanks for the help everyone
<3> nope knot werken but i find something
<3> tanks
<4> im finally done
<5> morning
<6> hi
<6> isn't "Using DatabaseName;" the correct syntax for selecting a database?
<6> in SQL server
<7> use databasename
<7> and loose the semicolon seperator
<6> thanks
<7> :-)
<6> you!!!
<7> hehe
<8> morning all
<9> hey d2d
<9> working on a saturday, d2d_w0rk ?
<9> hm...
<9> that was a real stupid question
<9> please forget i asked it
<10> hmm
<10> got a sql performance question
<10> in mssql i have 2 tables. one is table of syndications and the other is a table of syndication change logs. basically when they were updated, by whom and why the update was performed
<10> Now I'm going to pull a list of these syndications based upon which were updated last
<10> should i store the "lastUpdated" date also in the syndication id for performance sake so i dont have to do a subquery for each record?
<10> i know it kind of denormalizes the data a bit. I'm just wondering if it will be worth it sine the alternative is basically a subquery executed for each row
<5> Arrakis : it does really run counter to normalisation rules. plus you'll need to update two tables not one... ok, one insert, one update
<5> i would not add the extra field, and instead add a unique index on id and changedate
<5> that way, the join just needs to get max(changedate)
<5> and won't even touch the table
<10> hjmmm
<10> let me see if i can wrap my head around that
<10> can i order by that then?
<10> because basically ill be sorting by that "latest update" and ill also be pulling say the top 10 last updated syndications
<10> thats why i saw the subquery per row
<10> or perhaps im going about it wrong
<10> select syndications from in ( pull distinct syndication id from changelog order by changedate desc ) ?
<10> that would solve problem 1 i guess
<5> you can order by any field
<10> yeah but it gets hairy when you order by a many to one
<10> e.g. there may be 10 changelog entries per syndication
<10> and there may be 1000 syndications
<10> i would want to be able to sort by the latest update
<10> something like
<10> case when @orderby='latestupdate' and descending=0 then udftogetlatestsyndicationupdate(syndicationid) then
<5> that last sentence looked nasty - lets back up a little
<10> yeah it is nasty trust me
<10> but :-/ it works
<10> for the most part
<5> i would first create a view that does 'select id, max(date) as date from table group by id'
<5> then use that view in any joins
<5> that really will take the pressure off the nasty sql
<5> plus add the index i mentioned
<10> well the nasty sql is in there because i need dynamic sorting
<5> add the index first
<10> + paging
<5> k - but still will help
<10> + its web based
<10> can i use joins inside of cases then?
<10> ermm
<10> s/joins/views/
<10> or is the join / view juist to eliminate the need for the UDF ?
<5> yes - a view can be used pretty much anywhere you'd otherwise put a table
<10> okay
<5> udf? wtf?
<5> never mind that...
<10> hehehe
<10> kay
<10> ill play around with it a bit
<5> so, the view gets pre-compiled, and will use the index you create
<5> so it'll be really cheap to run
<5> then you just join to it on id to the the rows you want from the parent table
<10> okay
<10> thanks.
<5> you'll find that your main sql will become a damn site cleaner and clearer
<11> hi
<5> indeed
<11> i am using a database table for persisant storage in my application
<12> jLo: awesome ...
<12> I never thought of doing that before!
<5> outstanding - did you read the rules?
<12> that's an awesome idea
<11> i have around 90 variables though, is there anything potentially wrong with having a row with that many columns?
<11> object persistance*
<5> yes, there are potential problems with that many columns
<12> jLo: store the variables in rows, now columns ...
<11> my other option is an attributes table, attribute,attributeValue, objectID
<12> jLo: thats the better of the options
Return to
#sql
or
Go to some related
logs:
nzbget glibc detected *** double free or corruption
inukshuk cisco catalyst
#winxp
#beginner
zero.hastypastry.net/dojo/showthread.php
override password laptop
fabrikkinnstillinger acer laptop
ALCASE LORRAINE
#beginner
#stocks
|
|