US and Worldwide: +1 (866) 660-7555

Search Forums:

Type: Posts; User: darrell.nelson; Keyword(s):

Page 1 of 16 1 2 3 4

Search Forums: Search took 0.12 seconds.

  1. Replies
    13
    Views
    617

    If coauthors is a multi-valued dimension then I...

    If coauthors is a multi-valued dimension then I think you will need a bridge table between that dimension and your fact table. This the exact same concept as a many-to-many mapping relationship.
  2. Replies
    5
    Views
    287

    Would it make sense to just concatenate the three...

    Would it make sense to just concatenate the three fields into one Match_Field, i.e., "KentClark2/1/1968", before going into the Fuzzy Match step? That way you would only have one Match_Score to deal...
  3. Replies
    13
    Views
    617

    If your authors and coauthors exist in the same...

    If your authors and coauthors exist in the same db table you can create a VIEW of dim_author and call it dim_coauthor. In the view, you should rename all the fields such that it looks like two...
  4. Replies
    13
    Views
    617

    Thanks for providing the additional information...

    Thanks for providing the additional information (which I haven't had time to look at yet). However, I can say that I think you need to make a correction to your data structure. What you are proposing...
  5. Yes, snowflake schemas are supported. Follow this...

    Yes, snowflake schemas are supported. Follow this link for sample XML from a star schema...

    http://mondrian.pentaho.com/documentation/schema.php#Star_schemas
  6. Replies
    13
    Views
    617

    Can you post some samples of the data you have...

    Can you post some samples of the data you have extracted from the html?
  7. Replies
    13
    Views
    617

    Some questions to ponder before giving advice: ...

    Some questions to ponder before giving advice:

    What is your end goal? (Putting data into a datamart is not a goal!) Do you have any "facts", i.e., metrics, that you want to calculate and...
  8. Replies
    6
    Views
    373

    Why do you need to use StAX?

    Why do you need to use StAX?
  9. Replies
    3
    Views
    226

    Your assumption that you will "get the max value"...

    Your assumption that you will "get the max value" when denormalising is incorrect. You are grouping by timestamp but your groups have multiple values for the same key, such as:
    ...
  10. Replies
    2
    Views
    180

    Try the Value mapper step. Also, the Select...

    Try the Value mapper step. Also, the Select values step can be used to change the data type from String to Integer or vice versa.
  11. The DLU step needs those extra fields to work...

    The DLU step needs those extra fields to work properly even for a pure Type I dimension. If you hit the SQL button it will generate the SQL needed to add them to your dimension table.
  12. It's not a KETTLE option; it's a Java option. I...

    It's not a KETTLE option; it's a Java option. I believe you can set an environment variable, PENTAHO_JAVA_OPTIONS or something like that, and the sh/bat files will set up the JVM accordingly.
    ...
  13. I have used the Java SimpleDateFormat class for...

    I have used the Java SimpleDateFormat class for special formatting of timestamps. Formatting after switching time zones can be implemented in a JavaScript step like this if the incoming rows provide...
  14. Sure. I set constants in the Calculator step...

    Sure. I set constants in the Calculator step fairly often. I like that you can set the "Remove" column to "Y" so those values don't show up as fields in the output rows.
  15. Internally, PDI uses a Java Date object which...

    Internally, PDI uses a Java Date object which bases its epoch time on the number of milliseconds since 1970.

    Perhaps a simpler way would be to use a User Defined Java Expression step with the Java...
  16. Well, I have actually led you down the wrong path...

    Well, I have actually led you down the wrong path a little but it can be fixed. What I was hoping to do was have a query that looks something like this and use an input field to populate the...
  17. Replies
    6
    Views
    371

    There is a bug in PDI 4.2.1 that is causing your...

    There is a bug in PDI 4.2.1 that is causing your error. It was reported in JIRAs http://jira.pentaho.com/browse/PDI-7109 and http://jira.pentaho.com/browse/PDI-8110. It was fixed in PDI 4.3.0.1 and...
  18. Here is what I was thinking along the line of two...

    Here is what I was thinking along the line of two transformations.

    T1 currently does this:

    Generate a random number of rows (my example just generates up to 1000 rows but you would substitute...
  19. I'm not sure I understand where the "data set of...

    I'm not sure I understand where the "data set of 814" is coming from or why you want to chop it up in blocks of 100 but wouldn't it make more sense to pass into the transformation the "9" as a...
  20. Replies
    6
    Views
    371

    What step to use for table input? Try the Table...

    What step to use for table input? Try the Table input​ step.
  21. Replies
    6
    Views
    371

    Yes, it is possible. Generating the XML is pretty...

    Yes, it is possible. Generating the XML is pretty easy but I did have to use an XSL Transformation to add the DOCTYPE declaration and standalone attribute.

    13324


    <?xml version="1.0"...
  22. According to the Mondrian Documentation...

    According to the Mondrian Documentation (http://mondrian.pentaho.com/documentation/mdx.php): "Format strings use Visual Basic formatting syntax;"

    A good resource for the VB format syntax can be...
  23. What rules engine implementation does it use?

    What rules engine implementation does it use?
  24. Replies
    2
    Views
    189

    The Get System Info step may be able to give you...

    The Get System Info step may be able to give you the answer. Using the "system date (fixed)" info type will result in a constant date for the life of the transformation. You can also use the...
  25. To expand on what mcampbell wrote, you basically...

    To expand on what mcampbell wrote, you basically have either a star schema or a snowflake schema. If you have a dimensional database model (described in depth in the writings of Ralph Kimball, et al)...
  26. I don't believe dividing by 100 can be...

    I don't believe dividing by 100 can be accomplished through the format specifier.
  27. I concur with that diagnosis. What I do to get...

    I concur with that diagnosis. What I do to get around that is "lock" my computer instead of logging off. That keeps the drives mapped for scheduled tasks.
  28. Yes.

    Yes.
  29. Can you give a little more information on the...

    Can you give a little more information on the project scope and business rules you are trying to implement? A transformation could be created in 2 minutes that takes your 1 row of example input and...
  30. Replies
    8
    Views
    432

    Have you thought about: Setting up a web...

    Have you thought about:

    Setting up a web service (perhaps with OpenRules or JBoss Drools) and calling that service from a transformation?
    Embedding a rules engine in a User Defined Java Class?
  31. Replies
    8
    Views
    366

    I don't believe that is the case. If you a enter...

    I don't believe that is the case. If you a enter table name that doesn't exist, hitting the SQL button will offer to create it.
  32. Another version is attached. This one will detect...

    Another version is attached. This one will detect an edge at either a blank line or a new file. In addition, duplicate blank lines are handled properly. I added another input file to test the new...
  33. Replies
    8
    Views
    366

    Assuming you have the Excel input step set up and...

    Assuming you have the Excel input step set up and those input fields are passed on down the line to a Table output step, hitting the SQL button on the Table output step will give you the recommended...
  34. The Mail job entry supports variable...

    The Mail job entry supports variable substitution. Note the little gray/red boxes to the right of the input fields.

    Follow this link for more info: http://wiki.pentaho.com/display/EAI/.07+Variables
  35. I remember having a little trouble with that...

    I remember having a little trouble with that conversion but your sample data currently runs fine with my latest transformation. What version of Spoon are you running? What version of Java?
  36. I see. The key to solving the problem is finding...

    I see. The key to solving the problem is finding the "edge", i.e., identifying the boundaries between the sets of key-value pairs. I have attached a modified version of my sample transformation that,...
  37. The Row denormaliser step is the answer. You can...

    The Row denormaliser step is the answer. You can even specify a value to insert if the field is missing.
  38. Yes that makes sense. I believe the jars are...

    Yes that makes sense. I believe the jars are added to the Java classpath at JVM startup so will you have to restart when another jar is added in order for it to be "found".
  39. Replies
    2
    Views
    235

    Try using the Dimension lookup/update step. You...

    Try using the Dimension lookup/update step. You can read about it here: http://wiki.pentaho.com/display/EAI/Dimension+Lookup-Update
  40. i prefer it as well. I gave that option a few...

    i prefer it as well. I gave that option a few posts ago but the caller still seemed to want JavaScript.
  41. Even the JavaScript conversion can be vastly...

    Even the JavaScript conversion can be vastly simplified (down to a one-liner!) with that Apache POI Util class:

    var converted_AD_time = org.apache.poi.hpsf.Util.filetimeToDate(AD_time);
  42. Replies
    2
    Views
    247

    Try using the Filter rows step to separate the...

    Try using the Filter rows step to separate the aging from the non-aging passwords.

    http://wiki.pentaho.com/display/EAI/Filter+Rows
  43. Yep, here's an even easier and more efficient way...

    Yep, here's an even easier and more efficient way to convert Active Directory time values to Kettle/Java Date values. Since Apache POI is already an integral part of Kettle, a User Defined Java...
  44. I found another way to handle the conversion! I...

    I found another way to handle the conversion! I knew someone had already done the math on this but I didn't find it on my first cut. I'll post a sample using this class tomorrow if someone hasn't...
  45. You are encountering three different time...

    You are encountering three different time formats, which are roughly:
    1. AD is using the number of 100-nanosecond intervals since 1/1/1601
    2. Excel is using the number of days since 1/1/1900
    3....
  46. I have attached a modified version of your...

    I have attached a modified version of your transformation that parses your attributes field using the Split fields to rows and Split fields steps. Here is a preview of the output. If you want the...
  47. Hmmm. You don't seem to have JSON-formatted data...

    Hmmm. You don't seem to have JSON-formatted data (basically, name-value pairs separated by colons and enclosed in curly braces) so it stands to reason that the Json input step won't handle it very...
  48. Replies
    1
    Views
    181

    I may have a few answers to your questions. 1....

    I may have a few answers to your questions.
    1. Columns outside the schema should not be fetched. The should not be in the SQL statements.
    2. You can "put load" on a separate DB server by enabling a...
  49. 1) Your are not specifying a 'Format' string so...

    1) Your are not specifying a 'Format' string so no currency formatting will be inserted.
    2) The 'Group' symbol should be either a '.' or a ',' but you are not providing a 'Format' string so...
  50. That seems unusually slow. How wide are the rows?...

    That seems unusually slow. How wide are the rows? Are you dependent on a slow network?

    For further troubleshooting, try reading the source data from Oracle and dumping it to a local CSV file. Then...
Results 1 to 50 of 794
Page 1 of 16 1 2 3 4