Search found 26 matches

by paultechm
Mon May 23, 2011 10:31 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Logic to find no. of saturdays and sundays betwen two dates.
Replies: 17
Views: 6746

The above logic doesn't work out. Basically, I want logic to subtract N weekdays from a given date.While subtracting,we will not consider Saturdays and Sundays. This will give you the no.of weekends in that 23 days (YearweekFromDate(DateFromDaysSince(23, StringToDate('2011-05-09')))-YearweekFromDat...
by paultechm
Mon May 23, 2011 10:03 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: partial output from sequential file
Replies: 11
Views: 5805

Re: partial output from sequential file

Hi IF the input column contains two words i have to take first word in to one variable and second in to another and again in the first word i have to check for the hyphen(-) and if it is there i have to take the part before the (-) in one variable and part after the (-) in another variable. thats w...
by paultechm
Mon May 23, 2011 7:58 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: partial output from sequential file
Replies: 11
Views: 5805

Re: partial output from sequential file

word1 : if StageVar1=0 Then to_processing.pri_emp_add1 Else to_processing.pri_emp_add1[1, StageVar1-1]

firstpart : if StageVar=0 then Word1 else Word1[1, StageVar-1]


-Paul
by paultechm
Mon May 23, 2011 7:12 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Logic to find no. of saturdays and sundays betwen two dates.
Replies: 17
Views: 6746

Re: Logic to find no. of saturdays and sundays betwen two da

saraswati wrote:Please let me know a logic to find out the number of weekends ( Saturdays\Sundays) between two dates.
Use the following parallel function. You have to handle all the special cases

(YearweekFromDate(StringToDate('2011-05-31'))-YearweekFromDate(StringToDate('2011-05-09')))*2

-Paul
by paultechm
Tue May 03, 2011 3:12 pm
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Source table has 3 columns and target table has 1 column
Replies: 5
Views: 3974

Hi, It is altogether a different requirement , This is for converting single row into multiple rows Split these records in the transformer with out any constraint then pivot these links such as Link1: COL_A1 COL_ID COL_A2 COL_CODE COL_A3---------->pivot CoL_A(COL_A1,CoL_A2) COL_ID Col_B COL_CODE Col...
by paultechm
Tue May 03, 2011 8:14 am
Forum: IBM<sup>®</sup> Infosphere DataStage Server Edition
Topic: Source table has 3 columns and target table has 1 column
Replies: 5
Views: 3974

Re: Source table has 3 columns and target table has 1 column

Hello Create 3 stage variables (no need of split and collect the data ) 1)if not(isnull(ColumnA1)) Then ColumnA1 Else if not(isnull(ColumnA2)) Then ColumnA2 Else ColumnA3 Else 0= StgA 2)if not(isnull(ColumnB1)) Then ColumnB1 Else if not(isnull(ColumnB2)) Then ColumnB2 Else ColumnB3 Else 0 = StgB 2)i...
by paultechm
Sat Apr 30, 2011 12:59 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Read Rejected Rows and Format the Rows
Replies: 3
Views: 2893

Re: Read Rejected Rows and Format the Rows

Read as a single varchar column(with out column length) correct each line in a transformer ,use column import to split the data into actual meta data
by paultechm
Wed Apr 27, 2011 6:08 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: How to create a chain b/w 2 Fields
Replies: 2
Views: 2549

Re: How to create a chain b/w 2 Fields

There might be better solution than this ,Try this and let me know Job 1 Create a job to split the actual file based on the columnB(Ref_No) ,one has all the '0000'records with sequncenumber and other has sorted(desc) not like '0000' records surrogatekeygen---->seqfileMater.txt(SEQ,ColumnA,ColumnB) C...
by paultechm
Tue Apr 26, 2011 1:01 pm
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Remove Carriage Return from the end of the file
Replies: 4
Views: 5721

Re: Remove Carriage Return from the end of the file

Hi,
Read the file as a single varchar column(without data length) , remove the blank line based on the length.Use column export to split the records into multiple columns

-Paul
by paultechm
Tue Apr 26, 2011 8:50 am
Forum: IBM<sup>®</sup> DataStage Enterprise Edition (Formerly Parallel Extender/PX)
Topic: Removing Duplicates : unique situation
Replies: 13
Views: 11262

Re: Removing Duplicates : unique situation

Sort the input data based on the column B and enable the key change column Use filter to spilt the input data based on change value '1' or '0' Left Join the '0' stream with '1' stream based on the column B and retrieve the Column A from the '1' Stream . Filter the data based on Stream0.ColumnA- Stre...