This article appeared in the March/April 1999 issue of Everton's Genealogical Helper.

The Master Genealogist Tips
By Lee H. Hoffman


Splitting a Dataset the Hard Way
     Some users wonder if they can split off a specific group of persons into a new dataset. The answer, of course, is yes. Usually it is very easy. But sometimes, the group that is desired to split is sometimes a bit hard to define easily. That is what I’d like to discuss this month.
     The normal easy method would be to define a List of Persons (LOP) report that uses a Focus filter defining the desired persons. Then the Secondary Output tab would be selected to choose the Create New Dataset option and that would be it. But there are times when your desired group is harder to define.
     For example, if you want the persons in the dataset for a particular line and their spouses, there could be three steps. First to find the descendants of the line desired, then find the ancestors of the one descendant for which the line is desired. Finally, to create the new dataset for those persons and their spouses.
     The filter for the first report might be:
        ID Number                 Equals                   [?]    OR
        Is a Descendant of    Person Number   [?]   END
     The Secondary Output tab would be selected and the Change Flag option chosen. I have an INTERIM1 Flag set up with a default of N(o) for this purpose and set the option to Y(es) for those selected in this report. I then generate the report. Using my dataset and entering a two great grandfather, the report set the INTERIM Flag for 149 persons to Y(es).
     Now the filter for the second report might be:
       INTERIM Flag              Equals                  Y       AND
       Is a Ancestor of           Person Number  [?]    OR
       ID Number                   Equals                   [?]    END
The Secondary Output tab is again chosen with the same choices except that I use a different Flag -- the INTERIM2 Flag. When this report is generated, I enter my own ID Number in the two places. Then TMG selects the 149 persons with the INTERIM FLAG set to Y(es), finds 498 ancestors of those persons, and finally selects my record to make sure it is included for a total of 5 persons.
     The final run creates the new dataset. The filter for it might be:
       INTERIM2 Flag            Equals                  Y       END
with "Include [X] Spouses chosen at the bottom of the screen.
     The Secondary Output tab is selected and the only option chosen is [X] Create New Dataset. When the report is run, the five person are selected and six spouses are also selected for a total of eleven persons.
      After the report is read and closed, TMG displays an Extract Data Set window asking for the name to be assigned to the new dataset. At this point, I normally choose the Create Directory button and create a new directory/folder although this is not strictly necessary. If you do create the new directory, the path for that directory will be displayed in the place for the name of the new dataset after the directory is created. If you just enter the name of the new dataset without a path for a directory, the new dataset will be created in the TMGW directory. After you enter the new dataset name, click on the Create button and the new dataset will be created.
     You may now choose the new dataset and begin using it. One thing to keep in mind is that this does not remove the person in the new dataset from the old dataset, it just copies them to it.

Splitting a Dataset the Easy Way
Now that we have gone through all this work, I’ll tell you that all this could have been done with just one generation of a report. In that case, the filter might have looked like:
       (  ID Number                  Equals                    [?]      OR
          Is a Descendant of      Person Number   [?]  )  AND
       (  ID Number                  Equals                    [?]      OR
          Is an Ancestor of        Person Number   [?]  )   END
with the Include [X] Spouses option selected.
     Now selecting the Secondary Output tab, choosing the Create New Dataset option and generating the report. TMG will ask for the target ID Numbers and we will use the same persons as in the other runs. A two great-grandfather for the first two filter lines and myself for the last two lines. This will select one person for the first filter line, 148 for the second, one for the third line and 498 for the fourth line for a total of five person and then add the six spouses for a total of eleven persons again.

     The Master Genealogist allows eight filter lines and this is most often more than enough. But there are times when you will need more lines to include all of a somewhat complex filter; but because you can only enter eight filter lines, you will need to split the filter into two parts. You would use one part of the filter to set a Flag. Then you would use the Flag in place of those filter lines in the first run much as I used the INTERIM Flag filter line in the second "hard way" report to replace the first two lines in the "easy way" report. You can do this as much as you need although most runs will only require the "easy" way.


Further information about The Master Genealogist can be obtained from Wholly Genes, Inc., 5144 Flowertuft Ct, Columbia, MD
21075, on the TMG website < > or tollfree at 1-800-982-2103, TMG Tech Support is available at
1-410-715-2260 or by e-mail at <>.

Return to Lee's GH TMG Tips Page

Last revised:

Hit Counter