Create a list of the pages that need to be processed manually

[waltster]

In order to generate the manifests with the process I proposed, we will need to maintain a simple array of each page that needs to be converted into IIIF. For example, the URL https://scalar.usc.edu/works/piranesidigitalproject/16-veduta-2822 would be used.

This is because it is trivial to fetch information about the image being rendered in Scalar and linking any annotations on the image using this URL. The old method of identifying the annotations and images without any previous knowledge of the Scalar website by the script is computationally too consuming.

We can accomplish the first stage of this by getting the URLs for the five images that Dr. Britton sent. We can accomplish the long-term goal of this by compiling a list of all URLs for the pages necessary.

[adamskeefer]

@waltster I uploaded a list of the sample images to my branch and what page they correspond to. It is also attached here: ImagePageListSAMPLE.txt

Let me know what you think of the formatting and I will proceed with the whole site!

As far as the images the pages correspond to, would you like me to continue to use the way they are formatted on the repo? It's different than their names on the website, but generally only by one leading 0. For example, Vol17_007 is Vol17_0007 on the site.

[waltster]

@adamskeefer, did you see #3 ? This is the convention Dr. Britton wanted. Let me know if that answers it.

The capitalization can stay the same (since for any comparison we can use a toLowerCase()), so no problem there, but the zeroes and such needs to match.

The mapping looks good, but it needs to be JSON so we can parse it, and so its in an easily-translated format.

Thanks for doing that! I will give the generator a go on this map later. I will then create a map-ingestor. Look out for an issue for this ingestor and we can collab on it.

[adamskeefer]

Gotcha, will get the complete list done asap!

[waltster]

This has been done in the last few commits! Great work!