complete refactor of this project here : https://github.com/lmallez/wikidata-tree-generator

wikidata-to-gedcom

use wikidata lineage to generate a gedcom file

Usage

usage: WikidataToGedcom.py [-h] [--max-depth MAX_DEPTH]
                           [--allow-men ALLOW_MEN]
                           [--men-descendants MEN_DESCENDANTS]
                           [--men-ascendants MEN_ASCENDANTS]
                           [--allow-women ALLOW_WOMEN]
                           [--women-descendants WOMEN_DESCENDANTS]
                           [--women-ascendants WOMEN_ASCENDANTS]
                           [--load-parents LOAD_PARENTS]
                           [--load-children LOAD_CHILDREN]
                           [--enable-correction ENABLE_CORRECTION]
                           [--thread-max THREAD_MAX]
                           id {all,desc,asc,linear} output

name	type	default	description
id	str		wikidata id of the entity choose
method	str		{all, asc, desc, linear} method of gedcom generation (see bellow)
output	str		output file name
max_depth	int	10	maximum number of generation differences
allow-mens	bool	True	add men in gedcom
men-descendants	bool	True	add men descendants (can't be activate if allow-men is false)
men-ascendants	bool	True	add men ascendants (can't be activate if allow-men is false)
allow-womens	bool	True	add women in gedcom
women-descendants	bool	True	add women descendants (can't be activate if allow-women is false)
women-ascendants	bool	True	add women ascendants (can't be activate if allow-women is false)
load-parents	bool	False	add the parents of each character
load-children	bool	False	add the children of each character
enable-correction	bool	True	add missing wikidata links
thread-max	int	100	maximum number of threads

methods of gedcom generation	description
all	all the family members (can take a long time to complete)
desc	all the descendants of the root character
asc	all the ascendants of the root character
linear	combine desc and asc

beware of the different options you choose, some characters have a very large number of ancestors on wikidata

Samples

the generated images bellow are not created in this project, they are generated via the gramps software from the output gedcom

• WikidataToGedcom.py Q9696 linear ../samples/Q9696_1.ged

• WikidataToGedcom.py Q9696 linear ../samples/Q9696_2.ged --load-children 1 --load-parents 1

• WikidataToGedcom.py Q82339 desc ../sample Q82339 --allow-women 0

• Q2042 all ../samples/Q2042.ged

TODO

• parse the name and use GIVN and SURN tags
• use more properties of wikidata to fill gedcom (date of birth/death, weddings...)
• find an appropriate way to enter into transactions when a wikidata entity has more than one parent