LAE

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon



Hi,
Does MDA have some Crosstab node that I can use?

Best Regards,

Natanael Moreira
ZON

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: rboccuzzi

Here, I built a CrossTab (or pivot table) Node for you, and put in the example found on wikipedia.

I hope this helps.

Cheers
Rich
Attachments:
CrossTab.brg

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Thanks.

Can you tell me how can I create my own library with my own nodes?
I'm asking this because the sample you send to me is related with a library called InFlow, and I can't copy this crosstab node to my project (turns out error).
Can I create and add nodes to a library of my own?

Best Regards,
Natanael Moreira
ZON

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: rboccuzzi

Sure. If you go to BRE's help menu, and choose BRE Help, and then find the topic category on the left menu bar, Libraries, there is a bunch of information there on working with and creating libraries. Most of your questions should be answered there.

To get my sample into your library, you can just go to the library view in my provided graph, and copy the node, and then open your library, and paste it in, and you should be all set. You can also open my sample and go to edit base libraries (again, described in the BRE Help mentioned above), you can remove the inFlow library as well.

Cheers
Rich

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Hi,



I've execute the node you gave me but it turned out error.

****** Errors in node CrossTab__TP_COM

--------- Errlog for Pivot__Data_To_Names
Error while starting execution of node 'CrossTab+-+TP_COM.'
No executor for type:

************************************
Do you know why? I'm using two field that are never null. Cant understand.

Best Regards,

Natanael Moreira
ZON

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: Tim Meagher

Quick question:
What version of the LAE/LD do you have?
I notice that this appears to make use the "Pivot - Data To Names" node.
This node is available in versions 4.5.0 and later.
If you have an earlier version this node not work.

Regards,
Tim.

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

My version is 4.1.7.0 build 279
Is not the version.

Natanael Moreira
ZON

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: rboccuzzi

Natanael, can you contact your support channel and get yourself updated to LAE 4.5? It has a lot of new features and improvements over your existing version, such as InFlow reporting. This example will work with that version of the LAE, as it makes use of some of the new nodes that are also available as of 4.5.

Cheers
Rich

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon


Hi,

I've installed the new version, but now the error in the CrossTab node is:
---------------------------------------------------------------------
--------- Errlog for Pivot__Data_To_Names.GroupByPath.DataToNamesWithGr oupBy
2011-07-08 17:36:24.692; ERROR: Input length = 1
id: 0 chain: 0 group: 0
javaDetail: com.lavastorm.brain.io.data.brd.BrdRecordOutput.wr iteMetadata(BrdRecordOutput.java@401)

2011-07-08 17:36:24.723; ERROR: Error occurred while attempting to open output (0) : "GroupedData"
id: 13 chain: 0 group: 0
javaExceptionDetail: java.nio.charset.UnmappableCharacterException: 'Input length = 1';
stack trace:
java.nio.charset.CoderResult.throwException(CoderR esult.java@261)
com.lavastorm.util.JERROR.THROW(JERROR.java@571)
com.lavastorm.util.JERROR.VERIFY(JERROR.java@669)
com.lavastorm.brain.io.data.brd.BrdRecordOutput.wr iteMetadata(BrdRecordOutput.java@401)
com.lavastorm.brain.io.data.brd.BrdRecordOutput.op en(BrdRecordOutput.java@137)
com.lavastorm.io.data.WrappingRecordIo.open(Wrappi ngRecordIo.java@59)
com.lavastorm.brain.node.AbstractNode.openOutput(A bstractNode.java@522)
com.lavastorm.brain.node.core.DataToNamesWithGroup By.processAll(DataToNamesWithGroupBy.java@228)
com.lavastorm.brain.node.NodeWrapper.nodeProcess(N odeWrapper.java@407)
com.lavastorm.brain.node.NodeWrapper.run(NodeWrapp er.java@598)
com.lavastorm.brain.node.NodeWrapper.main(NodeWrap per.java@1122)
javaDetail: com.lavastorm.brain.node.AbstractNode.openOutput(A bstractNode.java@525)
dataDetail:
GroupedData(in) row: 0
columns: ['BA', 'java.lang.String'], ['__metadataNames__', 'java.lang.String'], ['__groupedData__', 'java.lang.String']
Metadata(in) row: 1
columns: ['__NamesMetadata__', 'java.lang.String'], ['__DataType__', 'java.lang.String']
records:
[Null, Null]
PivotedData(out) row: 0
columns: ['BA', 'java.lang.String'], ['REDE FIXA', 'java.lang.Integer'], ['FACTUR�VEL', 'java.lang.Integer'], ['GRATUITOS', 'java.lang.Integer']

2011-07-08 17:36:24.738; ERROR: Node failed.
id: 15 chain: 14 group: 0
javaDetail: com.lavastorm.brain.node.NodeWrapper.run(NodeWrapp er.java@623)
--------------------------------------------------------------------

A bit long! I can't understand it. Can you help?

Best Regards,
Natanael Moreira
ZON

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: Tim Meagher

Hey,

Not 100% sure from that error trace, I would need to see the data that is being pumped through the node.

I would guess though that this is occurring as the data that you are trying to pivot is unicode data. While unicode data is allowed within records, only String field names are allowed in the metadata (in the character set that you are using).

So without knowing fully, I would guess that you need to convert to String all of the the data within the fields that you wish to pivot to become the field metadata.

Tim.

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

Hey,

Thanks for the fast answer, but both fields I'm using in pivot are string.
One field is the Customer Billing Account. The other field tells me if the call were to "Fixed Line", "Chargable" or "Free".

I don't any issue in the content of these fields.

Best Regards,
Natanael Moreira
ZON

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: Tim Meagher

Ok,

Are you able to post a sample of what you are working with?

It looks to me like the problem is with the field name: "FACTUR�VEL".

I thought that provided the data was in String format coming in, and not unicode format, that the pivot node could handle extended ascii characters being mapped to the field names. I could be wrong on this though.

Tim.

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: Tim Meagher

Hi,

Sorry I meant a sample graph - i.e. sample node + sample data.

However, if that brd file is being fed in directly to the Pivot - Data To Names node, if you can just tell me the values of the parameters:
NamesField
DataField
GroupBy
OnDuplicate
NameSet

I can work with this.

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: zon

I realized something when you talked about "Pivot Data To Names".

I was still using the CrossTab that you previously sent.
So, I will use the Pivot Data To Names, and check if there is still errors.

Natanael

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: Tim Meagher

Hi,

So it unfortunately looks like the node (crosstab node, pivot - data to names node, or any Java based node for that matter) cannot handle extended ASCII characters in its field names.

I will raise an internal bug report about this to get it looked at.

In the meantime, for this set of data, you should be able to workaround this issue using the filter node in the attached graph.

Obviously, this is just a very temporary workaround, since it specifically only handles the '�' character. A more general workaround would be to check if the character code is > 127 & then handle specially before trying to pass it to the cross tab node.

Tim.
Attachments:
CrossTab.brg

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: burke

I keep getting "Cannot find base library inFlow".

Note: This was originally posted by an inactive account. Content was preserved by moving under an admin account.

Originally posted by: Tim Meagher

Hi,

The previously attached graph was constructed with LAE 4.5.
As of 4.5, the inFlow library is included in the LAE installation.
It is also included as a base library by default when you create a graph.

If the graph doesn't use any inFlow nodes, or features that were introduced on the 4.5.x branch, you can safely ignore error messages about a missing inFlow library - if the graph uses nodes from the inFlow library you will get a subsequent error message stating that the graph has nodes with a bad type.

In this case, however, the graph uses the "Pivot - Data To Names" node - which was also released in LAE 4.5.
If you see the error message that the inFlow library cannot be found it implies that you are using an earlier version of the LAE (prior to 4.5.0), therefore the attached graph with the "Pivot - Data To Names" node will not work for you.

You would need to upgrade to LAE 4.5.x to be able to use this graph.

Tim.