Unconnected / unlinked subtrees

[Pollux]

Can 'Ancestris' show unlinked / unconnected subtrees?
Since I've been working on genealogy, this menu item has been very important to me!
When I'm searching, I keep discovering personal data that could fit, but can not yet be linked.
I record this data with my genealogy software so as not to forget it.

[Zurga]

You have two options to see unconnected people :
First is Families groups report (Menu View => Direct reports => Family groups)

Second is to use the Graph view (check the available plugins to add it, then Menu View => Graph ).

Zurga

[Pollux]

Gogle Translator:
Thank you for your hints!
Unfortunately, I'm currently unable to test Ancestris with my GEDCOM file (or a copy of it).
It includes 326,939 people, 116,674 marriages / partnerships, 156,810 notes, 8,039 media, 3,195 sources, 17,179 source locations, ...
There is a lot of data to correct (including imports from at least 20 different sources), for which Ancestris would be ideally suited!
But even after a manual 'cleanup' (media, sources and source locations have been deleted) Ancestris needs over 90 minutes to read in, about 40% of the time for shortcuts!
PC: AMD Ryzen 5 1400, 4x 3.20GHz (8 threads), 64 GB RAM, NVIDIA GeForce GTX 1060 3GB
I would like to get away from my currently used genealogy software!
I've looked at a lot of genealogy programs, Ancestris is the only alternative left!

[Zurga]

Outch : 326,939 people is a lot of people to manage in memory.

Ancestris is actually an in memory program : All the GEDCOM is put in memory.

You can change the memory settings of the program if you open the "ancestis.conf" file in the "etc" directory.
Locate the linebe"ginning by "default_options=" (line 12 normally)
Change the parameter "-J-Xmx1g" (which means Max memory 1Go) for something like -J-Xmx32g (which means use 32Go memory max).

This should improve the behavior of the program.

We have in our todo list the development of a disk-persistent system to allow to manage such number of people.
But, there is a long road to achieve that feature.

Zurga

[arvernes]

Unfortunately, I'm currently unable to test Ancestris with my GEDCOM file (or a copy of it).
It includes 326,939 people, 116,674 marriages / partnerships, 156,810 notes, 8,039 media, 3,195 sources, 17,179 source locations, ...

I guess the problem could be with your "media". If your media are just links to external media, it should work just fine. My test gedcom file for Ancestris has more than 200.000 individuals, 80.000 families. My old laptop loads the gedcom file within not more than 20 seconds.
Check the file ancestris.conf in your .ancestris/trunk/etc directory. Be sure the Xmx parameter is setted to 1g as this one : "-J-Xmx1g"
Francois

[Pollux]

@ Zurga, @ Francois: Thanks for your answers!
You give me different hints to set the Xmx parameter to "-J-Xmx32g" or "-J-Xmx1g".
Sorry for the wrong RAM of 64GB! There are only 32GB.

Ancestris is installed on four Windows 10 computers and is constantly updated:
32GB (desktop), 16GB (mini desktop), 8GB (laptop), 2GB (tablet).

Below is the current header of my GEDCOM:

0 HEAD
1 SOUR "program name"
2 VERS "program version"
2 NAME "program name"
2 CORP "Program Provider"
3 ADDR "Street Information"
4 CONT "Location"
4 CONT Germany
1 DEST AGES
1 DATE "last save"
1 CHAR ANSI
1 SUBM @ SUBM @
1 "File Name" .ged
1 GEDC
2 VERS 5.5.1
2 FORM LINEAGE-LINKED

Ancestris does not recognize a valid file header and specifies an unknown source.

I could read the GEDCOM with a newer version of my current software (which I have licensed, but do not use) and save in "UTF-8". Would that make sense?

The "media" are mostly corpses from imports, so it's not 'real' files, but 'clues' to them.
I will remove ALL and reintegrate my own media as soon as I can use Ancestris!

Sources and source locations have become unmanageable. This confusion results from the most diverse imports of relatives from all over the world!

Other questions are sure to follow!

[Zurga]

I don't have a big GEDCOM file (10 000 people only), it works well with 1Go in memory.
François tells 1Go is sufficient for his file.
You can put the limit to 2 or 4Go to see if it's more usable.

For your header, if the data are exactly what you write here (with double quote), this is indeed a not valid GEDCOM header.
I'm not sure about the information you're looking for with this header.
Best charset is UTF-8, especially if you use 5.5.1 version of GEDCOM : this charset allows any character to be written.

Zurga

[Pollux]

Thanks!
Here is the file header in the original:

0 HEAD
1 SOUR AGES
2 VERS 1.50
2 NAME AGES
2 CORP Daub IT consulting
3 ADDR Glashuetter Weg 105
4 CONT 22889 Tangstedt
4 CONT Germany
1 DEST AGES
1 DATE 3 NOV 2019
1 CHAR ANSI
1 SUBM @ SUBM @
1 FILE Dieter Pollmann 2019 - Genealogie.ged
1 GEDC
2 VERS 5.5.1
2 FORM LINEAGE-LINKED

[Pollux]

I will save a copy of my GEDCOM file (isolated) on my new Win 10-set laptop and make any changes to it, which you and Francois advise me to do.
Step by step ... and, if necessary, manual intervention as described by Frederic in 'Gedcom problem' GED VERS 5.5.1 [change "5.5.1" to "5.5" manually].

[Zurga]

The SUBM line is very strange.
Normally, you have a reference to SUBM record.
@ SUBM @ is a strange reference.
But if you have another line with :
0 @ SUBM @ SUBM
This should works.

AGES is not a recognized software for Ancestris.
We don't do any specific modifications on the GEDCOM for this software.

Please be sure to understand the imports warning : unrecognized source means that the software is unknown or there is nothing specific to do.
Ancestris will manage the GEDCOM file with the standard conversion functions. (Check for invalid dates, check for incomplete tag, create Ancestris association pattern,...)

Zurga