Delete duplicate row

[nicolai_moller]

Hi

I'm trying to delete duplicate rows (activities) where the fields Start, End and Status are all empty. So if an activity exists more than once (like Activity A), the empty ones should be deleted (like Activity_ID 001). But only if all 3 fields are empty

Activity	Start	End	Status	Activity_ID
A				001
A	01-01-2014	05-05-2014	Completed	002
B				003
C				004
D	05-05-2015		Waiting	005

Thanks

[anbu1984]

Load * From Table

Where Len(Trim(Start)) = 0 And Len(Trim(End)) = 0 And Len(Trim(Status)) = 0

[#3]

Hi

Have a look this one.

Thanks

[nicolai_moller]

your suggestion deletes all empty rows. Only rows that duplicate AND empty should be deleted.

Activity A exists twice and one row (Activity_ID 001) is completely empty, so that row should be deleted.

[anbu1984]

Temp:

Load * From Table

Join(Temp)

Load Activity,Count(Activity) As Cnt Resident Temp;

Final:

NoConcatenate

Load * Resident Temp

Where Len(Trim(Start)) = 0 And Len(Trim(End)) = 0 And Len(Trim(Status)) = 0 And Cnt > 1;

Drop Table Temp;

[Anonymous]

modify Where statement.

Where Cnt=1 or (Len(Trim(Start))>0 AND Len(Trim(End))>0 AND Len(Trim(Status))>0)

[jonathandienst · Accepted solution]

Hi

This works, but makes the assumptions that there is only one populated line, but there may be one or more unpopulated lines:

Data:

LOAD Activity,

  Start,

  End,

  Status,

  Activity_ID,

  If(Len(Start) = 0 And Len(End) = 0 And Len(Status) = 0, Null(), Activity_ID) As TActivity_ID

Inline

[

  Activity,Start,End,Status,Activity_ID

  A,,,,001

  A,01-01-2014,05-05-2014,Completed,002

  B,,,,003

  C,,,,004

  D,05-05-2015,,Waiting,005

];

Results:

NoConcatenate

LOAD Activity,

  Max(Start) As Start,

  Max(End) As End,

  MaxString(Status) As Status,

  Num(Max(Alt(TActivity_ID, Activity_ID), '000')) As Activity_ID

Resident Data

Group By Activity;

DROP Table Data;

The TActivity_ID field stores the Activity_ID of the populated field. The Alt() selects this value if it exists.

See attached

Jonathan

[#8]

Hi,

PFA,

By using staight table, you can acheive your requirement.

Warm Regards,

Joshmi