GenealogyJ

[Anonymous]

imagine you are creating your own tree in the file me.ged
your friend John asks you to create a tree for him and you store it in the file frank.ged.
then you discover that you and frank share a common ancestot, and you want to put your trees together in order to show the whole ancestor's discendance. how would you manage such a situation?

the only solution to the problem that I could manage to find (but haven't tried yet) is to change all INDI , FAM , etc.. numbers so that they aren't the same as in the first file. then copy and paste in the first file. suppose that the INDI numbers of the first file are from 1 to 500 (@I001@ to @I500@), I'm talking of editing (manually?? ) the numbers of the second file so that are from 1000 upwards (@1001 and so on). then also all the FAM numbers and relationships should be changed, so that no one is the same as in the first ged file. then in genj I should be able to create relationships and delete some people that shows up two times.

Do you think it could work? I believe this should be a feature of the next version of genj

in this case "copy and paste" it's not so easy!!

[nmeier]

Hi Stefano

once upon a time we had a merging function that got left behind and needs to be re-introduced. I hope we'll have something helping in this case in the near future.

What's your take on it - should a merge try to find the same people by comparing IDs, names and dates and then suggest a merge? I think the user needs to check the matches but in theory then the data can be put together with persons who don't match possibly getting new ids (to avoid duplicates).

Anything I'm forgetting here? I think some applications also use custom IDs attached to persons to match data when files are returned. Don't know how that's supposed to work though.

I'll have to think about this one


Regards
Nils

[Anonymous]

i think a "merge" function should:
- let you choose manually which INDI and FAM are the same in the 2 files and match
- be also able to do an automatic research of them, but it's always the user who has to confirm it
- in case the IDs are equal (e.g. in 2 files generated by GenJ), generate new numbers for INDI and FAMs who don't match.

another hot point is: which of the 2 files has the correct information about one person? (e.g. the birth date is different, or the wife/husband name don't match). it's plain that merging 2 files with such differences would be more difficult.

(maybe this topic should stay in DEVELOPER )

[nmeier]

for now we can discuss the feature here

yes, merging means identifying who's who and also how to merge the duplicate information. We could offer four modes

• a.ged overwrites b.ged if mismatch
  
• b.ged overwrites a.ged if mismatch
  
• data from a.ged and b.ged is kept and tagged with notes - the user has to do some manually work after the merge (this works fine for a lot of stuff like having two birth dates for example - genj can keep that info and let the user go through those conflicts for resolution afterwards)
  

good input - thanks

/nils

[Udo from Germany]

For me merging GEDCOM files together is a very important and frequent function as I exchange new findings and more detailed attributes of persons or families with other hobby genealogists.

I would like all the GEDCOM data of both objects side-by-side and to be able to check line-by-line which of the data should go into the merged object. (Similar as GRAMPS provides.)