MAJW relationship generation
to : Martin Holmes
Minutes: 110
DA noted that we were generating inverse versions of many relationships
in the generation of XML input for the family trees; I figured out that this
was due to the fact that the function being used to retrieve relations for
a person was designed specifically to provide all those relations with the
person as @passive because its main purpose was to supply content
for their person page on the site. I added a parameter so that the inversion
can be turned off if not required.
On Wednesday discussed with DA the possibility of addressing family tree layout issues in various ways, including writing a custom layout algorithm (determined by DA to be too complicated), and the more practical approach of post-processing the dot file.
On Thursday, reworked the XML-for-family-trees code again, this time ensuring that every inverse relationship does get generated; DA can use this to detect clusters more effectively. I also added optional inversion of relationships where the target person is active in the function library, so that if/when we need any combination of active-inflected and passive-inflected relations for anyone, we can get them.