I want to convert my Reunion for Mac to Gramps. I have solved a lot of problems in the meantime. However, I am now running into a problem for converting sources.
In Reunion I have the following fields:
Title: GEDCOM code - TITL
Author: GEDCOM code - AUTH
Volume/Page: GEDCOM code - PAGE
These fields are in a Reunion ‘template’.
I want the content of Reunion ‘Title’ to appear in Gramps SOURCES ‘Title’ and the content of Reunion ‘Author’ to appear in Gramps SOURCE ‘Author’. This works fine in both cases.
The problem is that in Gramps CITATION ‘Volume/page’ the content of Reunion ‘Volume/page’ doesn’t appears. The content appears in Gramps SOURCES notes. What am I doing wrong? I think the GEDCOM code in Gramps for ‘Volume/page’ is PAGE.
I don’t think you are doing anything wrong, but Reunion might be doing something strange.
When Gramps imports a GEDCOM file, it doesn’t drop information when some data is not in the expected format. Instead the import write the data to a note, so nothing is lost.
Things to look for:
What is the GEDCOM version (tag VERS) in the export from Reunion? When exporting from Reunion do you have any option for choosing GEDCOM version?
TITL and AUTH should always have level 1 as they belong to a source (SOUR) definition.
PAGE should either have level 2 or 3 because the citation refers to a source.
0 HEAD
1 SOUR Reunion
2 VERS V14.0
2 CORP Leister Productions
1 DEST Reunion
1 DATE 7 MAY 2025
1 FILE Klut
1 GEDC
2 VERS 5.5.1
2 FORM LINEAGE-LINKED
1 CHAR UTF-8
0 @I1@ INDI
1 NAME Willem /Klut/
1 SEX M
1 BIRT
2 DATE 25 JAN 1821
2 PLAC Koog
2 SOUR @S3@
1 DEAT
2 DATE 01 AUG 1700
2 PLAC Koog
1 FAMS @F1@
1 CHAN
2 DATE 7 MAY 2025
0 @I2@ INDI
1 NAME Aafje /Smit/
1 SEX F
1 BIRT
2 DATE 7 DEC 1823
2 PLAC Zaandijk
1 DEAT
2 DATE 22 NOV 1700
2 PLAC Koog
1 FAMS @F1@
1 CHAN
2 DATE 7 MAY 2025
0 @F1@ FAM
1 HUSB @I1@
1 WIFE @I2@
1 MARR
2 DATE 31 JAN 1730
2 PLAC Zaandijk
1 RESI
2 DATE 22 NOV 1751
2 PLAC Koog
1 CHAN
2 DATE 7 MAY 2025
0 @S3@ SOUR
1 AUTH Gemeentearchief Zaanstad te Zaandam
1 TITL OA-0133 Burgerlijke stand Zaandijk
2 PAGE inv.nr 5, 31-10-1850, Register van geboortenakten, Burgerlijke Stand Zaandijk, 1843 jan 2 - 1852 dec 29
1 CHAN
2 DATE 29 APR 2025
0 TRLR
Note: Tried changing 1 PAGE to 2 PAGE, this makes no difference.
It looks as though Reunion is writing a better Citation GEDCOM than Gramps. And that Gramps isn’t recognizing that format PAGE part of the SOUR tag fully.
I did some experimenting GEDCOM Extensions .ged2 and Geneanet exporters by @grocanar writes a the Source Citation .ged more correctly. But that doesnt help with import.
If this is GEDCOM v5.5.1, then I disagree. According to the GEDCOM 5.5.1 definition there is no PAGE tag in the SOUR definition. 0 @S3@ SOUR is a source definition, while 2 SOUR @S3@ is a reference to a source.
The PAGE tag belongs in source citation which starts with a source reference.
So I would say, that Gramps writes better GEDCOM.
@Seagram
You could try to change the Reunion GEDCOM file, so it will be similar to the Gramps GEDCOM.
If there’s not too many PAGE lines, I would just use vi to edit the GEDCOM file. If there’s really many I would make a script with sed commands.
vi and sed are command line tools. On the Mac, they can be run in the Terminal application. But if you don’t already have some cli (command line interface) experience, there is a substantial learning curve. We’re assuming that there are so many broken citations that it would take many many hours to fix by hand.
Basically, you need a program to read, write and modify text files. I’m fond of BBedit on the Mac–even the free version is very capable–but there are many other options.
The idea is that you would do a series of ‘search and replace’ operations on the Gedcom file from Reunion and then import the modified file to Gramps. Because the Gedcoms are just text files, you can work out the process by trial and error until you get a satisfactory result.
I don’t know the exact manipulations that will be required. For programmers, the ‘gold standard’ for doing complex text manipulation is “RegEx” (regular expressions). RegEx phrases can be pretty darn hard to decipher…even for people who work with them regularly. OTOH, they can do stuff that seems like magic. BBedit has a built-in RegEx mode. Maybe what you need done doesn’t require that level of power?
My biggest challenge will be that in my case I have to split ‘0 @S3@ SOUR’ into ‘0 @S3@ SOUR’ and ‘2 SOUR @S3@’. And that I have to add ‘2 SOUR @S3@’ part to several events. And that for several hundreds of sources/citations.
It might be good to try with just changing one or two source definition, and then run an import to see if it solves the problem. A manual change for one or two sorce definitions can easily be made in textEdit. If that works, then it’s time to make the big update of the gedcom file.
I did that and it works fine. Now I have to decide whether to transfer all sources/citations. I was hoping there would be a script or something for this. So I will have to transfer everything manually.
It seems like enabling the PAGE tag (in the n SOUR @<XREF:SOUR>@ tag structure) in the libgedcom.py module would be another option. But there’s more of a learning curve there.
Each source has multiple citations but each citation has 1 source. Each citation consists of a PAGE a NOTE (which contains a url) and a DATE (which contains the date when the citation was viewed).
By the way, the comments about Reunion’s odd Citation GEDCOM export points out something very specific:
This problem mostly affects users who used the free-form Citation rather than filled in the structured fields for Citations in Reunion. And from the text of your sample “PAGE” data, that’s what is happening here.
Then the inv.nr 5, 31-10-1850, Register van geboortenakten, Burgerlijke Stand Zaandijk, 1843 jan 2 - 1852 dec 29 in the Page field seems doubly strange.
The full citations read ‘OA-0133 Burgerlijke stand Zaandijk’ + ‘inv.nr 5, 31-10-1850, Register van geboortenakten, Burgerlijke Stand Zaandijk, 1843 Jan 2 - 1852 Dec 29’