SSIS help in deriving negative values

  • Hi,

    I have a requirement where from a string we need to derive the amount and also based on the alphabet we need to make it negative or positive amount. Here is the string

    '1047383658000012345678990000002014083J354310020140850000005335R0000'

    In this string '0000005335R' is the amount with 2 decimals 000000533.5R

    We need to decode A = 1, B= 2 till I = 9 and J = 1 with negative values till S = 9(negative)

    so the final output in this string should be:

    if the string is '1047383658000012345678990000002014083J354310020140850000005335A0000'

    output should be 533.51

    if the string is '1047383658000012345678990000002014083J354310020140850000005335S0000'

    output should be -533.59

    I need expert help on this asap.

    Thanks,

    RK

  • rkota2008 (7/16/2014)


    Hi,

    '1047383658000012345678990000002014083J354310020140850000005335R0000'

    In this string '0000005335R' is the amount with 2 decimals 000000533.5R

    We need to decode A = 1, B= 2 till I = 9 and J = 1 with negative values till S = 9(negative)

    What psychopath has you locked up in the basement trying to drive you insane? Call 911! STAT!

    ... Assuming that's not the case and there's some really strange business precedent for this... somewhere... How do you determine the position in the string that is used for values? Is it the character position (ie: fixed width flat file)?

    Assuming that it's consistently identifiable, the process is relatively straightforward. You'll need to pass it into a synchronous script transformation, translate the character directly to the value needed, concatenate it onto the rest of the string, and include a multiplier value in a second variable. If A-I, then make your multiplier 1, J-S -1. Throw this all together, convert to a numeric, divide by 100 (for your decimals), and then multiply by the multiplier. Finally, kick the row back out into the stream.

    You could, in theory, do all this via some insanely convoluted expression in a derived column component, but I'd avoid that. The Expression would be a few dozen lines long, at least, and almost incomprehensible.

    And no, I don't have any code that does anything like that currently... I'd never foresaw the use of something like that.


    - Craig Farrell

    Never stop learning, even if it hurts. Ego bruises are practically mandatory as you learn unless you've never risked enough to make a mistake.

    For better assistance in answering your questions[/url] | Forum Netiquette
    For index/tuning help, follow these directions.[/url] |Tally Tables[/url]

    Twitter: @AnyWayDBA

  • Thanks Craig! It's a fixed width easily identifiable string. I am working on creating variables able to derive the values but not able to get the negative values, code is really becoming messy, i need some help if there is any script task or any other solution that can make my life easy. Maybe there is a better way to do it, but so far this is what I have done

    P.S. aaaa contains the string..

    (DT_CY)((SUBSTRING(SUBSTRING(aaaa,53,10),2,LEN(SUBSTRING(aaaa,53,10)) - 2) + "." + RIGHT(SUBSTRING(aaaa,53,10),1)) + REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(SUBSTRING(aaaa,63,1),"A","1"),"B","2"),"C","3"),"D","4"),"E","5"),"F","6"),"G","7"),"H","8"),"I","9"),"J","1"),"K","2"),"L","3"),"M","4"),"N","5"),"O","6"),"P","7"),"Q","8"),"R","9"),"{","0"),"}","0"))

    I am able to get the values but not able to derive the negative values...

  • Well, as I mentioned, it gets messy...

    The problem here is that you have to consider this a multi step process.

    1) You have to get the letter by itself.

    2) You have to do two things with the letter. You have to translate it into its numerical equivalent AND create a another column (variable, whatever) as 1 or -1, depending on the A-I vs. J-S location.

    3) You have to append the numerical equivalent back onto the tail of the string for the rest of the number. You then convert that string to numeric (x,2) and then divide it by 100, to get your decimals.

    4) Finally, multiply the result from 3 by your 1/-1 variable/column you found in 2.

    Don't try to do this all in a single step, you'll drive yourself absolutely batty if you ever have to maintain it. If you're not looking to use the .NET transformation script component, then I'd recommend using multiple derived columns in the stream, chaining off each other. First one would split out your two pieces of the amount string, next one would derive the # and the multiplier, third one would concatenate the values and convert to numeric, the final one would apply your division and use your multiplier.


    - Craig Farrell

    Never stop learning, even if it hurts. Ego bruises are practically mandatory as you learn unless you've never risked enough to make a mistake.

    For better assistance in answering your questions[/url] | Forum Netiquette
    For index/tuning help, follow these directions.[/url] |Tally Tables[/url]

    Twitter: @AnyWayDBA

  • Thanks Craig for your reply. I am able to resolve the issue. I am posting the link for whoever wants help on this

    http://www.ssistalk.com/2007/03/14/ssis-working-with-cobol-zoned-signed-decimals/

  • Excellent! I'm glad you found a thorough post to help walk you through the process.


    - Craig Farrell

    Never stop learning, even if it hurts. Ego bruises are practically mandatory as you learn unless you've never risked enough to make a mistake.

    For better assistance in answering your questions[/url] | Forum Netiquette
    For index/tuning help, follow these directions.[/url] |Tally Tables[/url]

    Twitter: @AnyWayDBA

Viewing 6 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic. Login to reply