GRAMPS: 5.2.3
Python: 3.11.9
BSDDB: not found
sqlite: 3.42.0 (2.6.0)
LANG: en_GB.UTF-8
OS: Linux
Distribution: 6.4.0-150600.23.25-default
Perhaps unwisely, I have had the habit of creating the odd independent tree snippet along with my main tree. The idea being, to merge them into the main tree when the connections become clear. Unfortunately, (and of course!) with time and without resolution of the connections, I have lost the overview of which snippets are lying around. What would be the best way to identify them individually? Something like a list of unconnected persons to, say the home person, which are grouped together somehow?
I know, donât tell me, it doesnâ't sound like a good idea now. I should have created independent trees and imported them when the connection became clear. It worked well when I was working continually on Gramps and had them upmost in my mindâŚbut after the yearsâŚhmmm.
Cheers
Harvey
I have done the same thing especially for a unique name that is almost 100% guaranteed to be related. I too have little islands of families not yet connected to my tree.
There is the Not Related tool. It will return a list of people that are not connected to the active person.
If you are in the Relationships view, the list of people returned will be people that you cannot click through the view and reach from the active person.
Once the list of people is created, you can select everyone in the list and give them all a Tag. Not Related worked for me.
Do you mean that you have a dozen databases, each with their own tree? Or are they all in one database?
If itâs the latter, you can try Genealogica Grafica. It can show your isolated trees, when you give it a GEDCOM file, but itâs a Windows program that I havenât tested under Wine.
Perhaps we could get the opinion of the managers of the HisKi trees in the Isotammi project? They wouldâve dealt with such snippets on a regular basis.
They are all in one database, and there are a total of 2095, where some 536 individuals are not related to me. This is presumably because I record independent witnesses at family events, or visitors, lodgers, domestic servants on census returns. In addition, there are more substantial tree snippets that will hopefully connect with the main tree me as my research matures.
Of course, the âNot Relatedâ utility is a good starting point. HoweverâŚ
It has no possibility to export or save the results to help keep them in view when working.
The only output it has is an on-screen scrollable alphabetic list of names. Parents are shown, thus indicating the existence of a tree snippet that may, however, be substantially larger. With 536 unrelated people the number of home-made tags (in addition to the 2 provided by default) could become substantial. And the home-made tags would have to be transferable on Gramps version upgrades, which, at the moment, would be quite cumbersome.
It occurs to me that, to be useful, tags would have to be descriptive in order to characterise the origin and purpose of the person or group.
Sorting or grouping by Tag (and Parents?) would be a helpful addition.
âŚ
âŚ
Just a few thoughts. Gramps is a marvellous tool!
Maybe this helps: I wrote a Supertool script that finds all the non-connected subsets in the database. The script will display the number of people in each subset and a sample person from each set. The code is based on the âNot Relatedâ tool.
For example, for the sample database the script finds 105 separate partitions. The largest one has 1844 individuals:
The supertool script mentioned by @kku looks great, and if the number of trees is small, you can pick one ID from the 2nd, and 3rd, etc., and create a related filter around that person, for export to another database, and removal afterwards.
And even without that, you could pick one person from the not related list, and build a related filter around that, just to see what you have. The script is much nices however, especially when you have a lot of isolated trees.
Super. Thanks That works well, at least to get a saved overview of the individuals.
Pity that the parents canât be exportedâŚand Notes in general, instead of just TdDoâŚ
Does the âNot Relatedâ tool consider Associations, or people who merely share Events? Iâm guessing not, but that might be helpful in @ahnâs case:
The script is a Python written for the SuperTool addon tool. SuperTool allows quickly making ad hoc queries on filtered data without the burden of building a lot of interface. SuperTool routes data from the current view through the script to generate tables of results.
SuperTool scripts can be adapted into addons for Gramps: QuickReports, Reports, Tools, filters, Gramplets.
In this particular case, it is even more than that.
Kari created a separate subsets module that is imported and executed by a SuperTool script. SuperTool can be used as a filter of the People you want to process with that module.
It is safe to say that it needs some experimentation. I am unclear how the filtering of People affects the evaluation of connectivity between the remaining people.
Does that filter also filter Families? (e.g., if the Parents are filtered out, is the Parent family ignored, disconnecting the Siblings?)
Hiskitrees are large, from 20 000 ti 150 000 people. Key demand of Subsets is capability to run in modest time. With a tree of 25 000 (from Käkisalmi parishes) runtime was speed, but with tree of 120 000 (St. Maria parish of St. Petersburg) I stopped the run after 10 minutes, nothing seemed to advance.
It would be nice, if you can scrape 1-person islands from result list. They are of no interest in our case.
Export/import of a subtree should be supported by setting resilient IDs or tags and developing an ad-on for matching in import. For long time ago I used Import and Merge-addon. Maybe it could be something to base on? Regards Pekka aka Peter Power
