The U2 Egypt project aims to digitize a series of recently declassified aerial photographs that document the Nile Delta and Nile Valley. Shot by U2 spy planes, this imagery provides much higher resolution images than any currently available imagery (for instance the CORONA satellite imagery that has revolutionized archaeology in this area in the past two decades), and shows the landscape prior to the construction of the Aswan High Dam.

This repo contains the website which contextualizes the images, provides geographical information about them, and describes the project.

Specs

This site is built in Hugo. The command to build the site for the development server, including drafts, is hugo build --minify --cleanDestinationDir -b http://dcdscit2.services.brown.edu/projects/u2egypt/. The command to build the site for production is hugo -e production --minify --cleanDestinationDir. To deploy, you must be on VPN, and rsync files: rsync -avz public username@.server:/var/www/html/projects/u2egypt replacing username and server with the correct values.

Dependencies

The image comparison functionality is by @cloudfour. The map is built with Leaflet and a number of plugins.

Custom shortcodes

To embed a BDR item using Mirador, use the following shortcode:

{{< mirador "1234" >}}

where 1234 is the numeric value for a BDR PID, e.g., the item at https://repository.library.brown.edu/studio/item/bdr:89322/ would be rendered with {{< mirador "89322" >}}. Multiple Mirador embeds can be included in a given page.

To compate two images of equal dimensions:

{{< compare left="804now.jpg" right="804then.jpg" left-alt="Giza now" right-alt="Giza then" label="blah">}}

The shortcode assumes the images are page resources. Multiple image comparisons can be included in a given page.

Kiosk data

When getting updated images and json from Kiosk, put everything in the /assets/kiosk directory.

Cleaning/reformatting

In the project root, run python assets/js/clean_data.py -i assets/kiosk/docs.json -o assets/kiosk/clean_data.json. This script:

1. removes the wrapping u2ers_sites so the values are easier to access
2. removes sites where the limit_view_to field is not null, and where there are no U2 images, which shouldn't be made public
3. transforms the at_periods and g_earth_damage_cause field values into arrays, so Hugo can loop through them appropriately (and use at_periods as a taxonomy)
4. transforms the u2_images_showing_the_site field into an array

Once the data has been cleaned and formatted, we need to add information about the BDR items which are part of the Kiosk dataset. To do so, run

python assets/js/pid_lookup.py \
-s assets/kiosk/clean_data.json \
-l static/bdr.json \
-o assets/kiosk/data.json

Creating geojson

Run the json2geojson.py script: python assets/js/json2geojson.py -i assets/kiosk/data.json -o static/kiosk.geojson to transform the cleaned data into geojson.

There has got to be a less stupid way to make the Kiosk images, which should be global resources, published so the {{ .RelPermalink }} code works, but for now there is a Secret content page, which publishes all the images in the assets directory.

To-dos