To test a potential fix for Bug 11480 (Progen import bug with patronyms), it would help to have a sample data file with patronymic names. Is someone able to create and attach a minimal dataset as described in this note? Thanks.
There is a demo version for Windows that can handle up to 50 persons:
Thanks @emyoulation. Fumbled around a bit and created a data set but Gramps fails to import - just hangs forever. I know I’ve seen that secondary bug reported but can’t find it at the moment.
Sill looking for someone to provide a working data set.
There are 2 samples in the Notes for 0005146: ProGen import fails - Gramps - Bugtracker – Free Genealogy Software
The .def file version headers read:
[PRO-GEN]
version=3.0a
type=def
format=2
And a .def ZIP sample (which I have not examined) in the notes for 0006195: Import from pro-gen - Gramps - Bugtracker – Free Genealogy Software
Unfortunately, the 5.2 importer plugin source or Gramps Plug-in Registration does not mention the version of .def file compatible with the importer. Since 09/02/2022, the current version of Pro-Gen is 3.51
Plugin name: Pro-Gen [Importer] Id: im_progen Version: 1.0
Description: Import data from Pro-Gen files
Filename: importprogen.py Location: /home/districtsupport/.local/lib/python3.11/site-packages/gramps/plugins/importer
Audience: Everyone Status: Stable
Plugin name: Pro-Gen lib [Plugin library] Id: libprogen Version: 1.0
Description: Provides common functionality for Pro-Gen import
Filename: libprogen.py Location: /home/districtsupport/.local/lib/python3.11/site-packages/gramps/plugins/lib
Authors: The Gramps project
Email: http://gramps-project.org
Audience: Everyone Status: Stable
Have not discovered the Pro-Gen Export plug-in mentioned in the wiki.
There does not seem to be a .def version check. But the history goes back to 2008 when Kees says in a maillist thread that the Importer was incomplete.
In the GitHub history for the module includes a note about the .def files needing to be Pro-Gen version 3.2.1 or later because it needs to be encoded with the old (CP437) IBM character set :
[Bug 6413 missing from MantisBT ] Fix importing ProGen files
This isn’t a perfect fix because it requires that the ProGen file is encoded with CP437, which is guaranteed only in Version 3.21 and later.
There’s a 2017 Dev maillist discussion about migrating the Pro-Gen importer and exporter from built-in to add-on for the Gramps 5.0 release. It appears that this goal slipped.
Other interesting maillist threads:
Pro-Gen is a Dutch program, partly made by a fellow student, and it can be used in English too, as you can see on this English download page:
https://www.pro-gen.nl/gbhome.htm
Would you be able to download and install that, and create the test data?
Hi @ennoborg I’ve done that, and that’s the data that doesn’t import into Gramps - with latest layout (Progen terminology) it runs into an unhandled exception, and with the 3.0 layout it just hangs forever.
@emyoulation I’ve looked at all the Pro-gen related defects in the past, downloaded data sets from all of them and have similar troubles with each one (recalling from memory).
Could be Gramps import issue, could be a import settings issue, or something else but clearly this is not my area of knowledge so I’m probably spending way more time on it that someone who uses ProGen would.
Others have seemed to reached that same conclusion. And I think that’s why they recommended that it be backed out of the core.
An add-on isn’t subject to the tyranny of the “major release” cycle. (Both in Pro-Gen not affecting enough users to avoid being cut from the priority list and the long cycle times.)
Even if the bugs are not fixed, it is worth moving the plug-in.
I suspect that doing so WITHOUT making it functional felt like “passing the buck”. So that’s why a non-functional plug-in remains in the core.
And it might be why several of the early copyright credits have disappeared.
In Pro-Gen, users can define their own fields, which are stored in those DEF files. And if my memory is right, they can also define GEDCOM tags with them. I also know that their GEDCOMs are pretty straight-forward, as is the program itself, which makes me think that it’s easier to change the GEDCOM importer when needed, and forget, or rather abandon, the Pro-Gen importer.
The program is quite popular here, just like Aldfaer, and I don’t expect many conversions to Gramps, because that requires quite a change in thinking.
The infrequency on Import is probably affected by there being no Export plugin.
This format would probably get more exercise if Pro-Gen users could use Gramps as a tool instead of just an exit strategy.
Dutch tend to be cheap, and Aldfaer is popular because it’s free. Pro-Gen is not free, but it was one of the pioneers, and much easier to use than the competition, which is why it’s still popular too. And in both, data entry is much faster than it is in Gramps, which is one of the reasons why most people won’t even think about a migration. You have a similar situation in the US, where people stick with Brother’s Keeper, a program that I once recommended to my dad, because it was the 1st Windows program that spoke Dutch. The local programs were all DOS programs then.
In Pro-Gen, persons have only two parents, and source citations are simple too, just like on Geneanet, where you just type a single line, and don’t have to think about a hierarchy with a repository, source, and citation. It also gives users a chance to define their own reports, and these are all things why their users avoid Gramps, or more complicated Dutch programs like GensDataPro.
For them, Gramps has no use as a tool.