Resolve mismatches between Seshat polity year ranges and shape dataset year ranges

[edwardchalstrey1]

Where the shape has a Seshat ID, check the polity start and end years of the shapes dataset matches those from the core_polity table Compile a list of problems for the RAs to handle They will do this QA themselves

• [x] email the RAs with some examples on where the shape start/end year doesn't match the polity shape and end year
  • Use the db to make a spreadsheet where that's the case
  • Link to a few examples on the dev site
  • In some cases multiple polities are linked to the same seshat ID, so highlight this

[edwardchalstrey1]

Would a better solution be to use the Seshat polity years and just show the first or last shapes for longer (forwards in backwards in time) instead of leaving it blank. Where shape dataset polity extends beyond the seshat polity range I think this should just be cut off in the viewer.

The problem with this approach might be that on the world map we'd still be using the shape dataset years.

TODO:

• [x] Ensure the displayed years on the polity map match the polity years
• [x] Modify the world map so that it shows only Seshat polities by default
• [x] with the years being the Seshat years, not shape dataset years
• [x] Implement a dropdown to select the Shape dataset to load, the options for now can be "Seshat" and "Bennet et al 2024 (partially linked to Seshat)"
  • This can later be updated to include additional shape datasets
  • [ ] Somehow have a way to describe the dropdown options
• [x] Fix tests
• [x] What's going on with British Empire II (years still don't match page)
  • Matches the "intrinsic duration"
• [ ] Switching between Bennet and Seshat linked should just add or remove the non-linked shapes in the same way as the show polities checkbox
• Suppress the click-ability of polities with "empty" pages
  • the polity start year > -10,000 (the default value)
  • Note: this might not be worth doing, depends on whether the latest db actually has these missing pages
• [x] Use a recent db dump to get more linked polities (maybe do this before the above)
• [ ] Make the world map Seshat linked polities use the Seshat years when "Bennett: Seshat linked" selected? Wait for email response

[edwardchalstrey1]

Meeting Wednedsay 20th March 2024

• This will likely be solved on the data side, so we should not attempt to fix it with a code fudge

[edwardchalstrey1]

Hover over capitals sometimes hidden behind shapes - I think this will go away after the shape dataset/ polity dataset years mismatch goes away