hi
i need to reject some rows in a file, where the column "FRESH" is null value .
i used the command: FRESH is not null in the filter stage but it did not work . please help me with this problem
thank you
deepthi
filter stage
Moderators: chulett, rschirm, roy
Re: filter stage
How did it not work? Did it result in an error?deepthi wrote:hi
i need to reject some rows in a file, where the column "FRESH" is null value .
i used the command: FRESH is not null in the filter stage but it did not work . please help me with this problem
thank you
deepthi
Were you trying to trap the rejected records? If so, did you have a reject link? Did you make sure that the 'Output Reject Records' was set to TRUE (Stage/Properties)?
You don't necessarily need to jump to a transformer. The filter should work. If all you have to do is filter some input, then you are over complicating things by using a transformer. 99% of the time a specific stage will do it faster than a transformer - simply because that is what it was designed to do. A transformer is more dynamic and can do more than one thing, but it seems to me that simple tasks should be relegated to simple stages. I can use a 4x4 truck to drive in a road race (it can drive fast and do many other things), but the guy in the race car is going to beat me on a track - that's what it was built for.
I have nothing against transformers. We use buildops instead, but those are really just another type of transformer and all the same arguments apply. I could filter data with a buildop, but it would be simpler, more straight forward, and probably faster to use the Filter Stage. It also makes your design more self-documenting - in other words, if someone else were to look at your program they should be able to get the general idea of how it works.
Well, that's my 2 cents... HTH.
Brad.
Re: filter stage
initially, when i use filter stage, no rows are rejected which have nulls in a specified column "FRESH", all the rows are passing to the destination file.
i checked the option 'Output Reject Records' to true but it is asking for reject link, i could not include this reject link to the filter stage.
thank you
i checked the option 'Output Reject Records' to true but it is asking for reject link, i could not include this reject link to the filter stage.
thank you
bcarlson wrote:How did it not work? Did it result in an error?deepthi wrote:hi
i need to reject some rows in a file, where the column "FRESH" is null value .
i used the command: FRESH is not null in the filter stage but it did not work . please help me with this problem
thank you
deepthi
Were you trying to trap the rejected records? If so, did you have a reject link? Did you make sure that the 'Output Reject Records' was set to TRUE (Stage/Properties)?
You don't necessarily need to jump to a transformer. The filter should work. If all you have to do is filter some input, then you are over complicating things by using a transformer. 99% of the time a specific stage will do it faster than a transformer - simply because that is what it was designed to do. A transformer is more dynamic and can do more than one thing, but it seems to me that simple tasks should be relegated to simple stages. I can use a 4x4 truck to drive in a road race (it can drive fast and do many other things), but the guy in the race car is going to beat me on a track - that's what it was built for.
I have nothing against transformers. We use buildops instead, but those are really just another type of transformer and all the same arguments apply. I could filter data with a buildop, but it would be simpler, more straight forward, and probably faster to use the Filter Stage. It also makes your design more self-documenting - in other words, if someone else were to look at your program they should be able to get the general idea of how it works.
Well, that's my 2 cents... HTH.
Brad.
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact: