Hello,
I am going to be processing a series of book objects consisting of separate image files for each page, and I'm wondering the best way to structure the transfer itself. Each book will be a separate transfer. So, I would wanna structure it like so:
US.NN.FIT.SC.362.1.1.1.19470106/
├── objects
│ ├── US.NN.FIT.SC.362.1.1.1.19470106_001.tif
│ └──US.NN.FIT.SC.362.1.1.1.19470106_002.tif
│ └──US.NN.FIT.SC.362.1.1.1.19470106_003.tif
│ └──etc.
└── metadata
└── metadata.json
The issue I have is how to correctly reference the book itself (the directory) in the metadata. In the documentation, it says that the filename needs to start with objects/, but in this case it's 'objects' directory itself that I want the metadata to refer to. I ran a transfer with the 'filename' as just 'objects' and it worked, but I wondering if that's an ok way to do this.
The example in the documentation uses this:
directoryTransfer/
├── CoastNews-1964-01-02
│ ├── page-01.jpg
│ └── page-02.jpg
├── CoastNews-1964-01-09
│ ├── page-01.jpg
│ └── page-02.jpg
└── metadata
└── metadata.csv
That makes sense when you have multiple book-like objects, but in my case, I'm only including a single book per transfer, so it would seem redundant to have something like:
US.NN.FIT.SC.362.1.1.1.19470106/
├── US.NN.FIT.SC.362.1.1.1.19470106
│ ├── US.NN.FIT.SC.362.1.1.1.19470106_001.tif
│ └──US.NN.FIT.SC.362.1.1.1.19470106_002.tif
└── metadata
└── metadata.csv
Is there a good general practice to follow here?
Thanks,
Joseph Anderson