This week I’ve been taking the beginning steps of locating, downloading, and uploading the individual file images for all items within the legacy Hoagy Carmichael site. These beginning steps involve downloading all of the METS files associated with each item; these METS files contain the unique PURLs for each jpeg image of all pages related to each item that is viewable to a user. For instance, one set of lyrics written by Hoagy Carmichael might be comprised of two pages in total — each of these pages is represented by its own image file (a jpeg, in this case), that can be downloaded and uploaded into the new Omeka site in aggregate using a batch uploading plugin connected to the Omeka site.
Some of this process has been time-consuming and tedious, especially when I was manually locating each METS file listed in the EAD document, searching for the file via a web browser, then downloading the METS file onto my personal machine. I was doing this for about a week — all week — but Nick decided it would be much more effective to automate this process. He showed me how to download a homebrew package called ‘wget’ through the command line, and then provided a command that could take in a csv file that contains one column of file names, and a second column of METS PURLs associated with each file name, and automatically download all the METS files listed in the .csv file.
Once the METS files are downloaded, I can take that list of files and run an XQuery against each item PURL and extract all image PURLs from each item METS file. Once this is accomplished the entire new list of image PURLs can be saved as a file, then converted to a whole new .csv file, that can then be used in the batch upload process on Omeka. This all took another full week to figure out, and I will have to begin learning the full batch upload process next week. Batch uploading will likely include not just the jpeg images associated with each viewable item, but also the appropriate metadata for each of these items. I am both excited and daunted by the amount of information I’ll soon be uploading in aggregate onto the Omeka site.
Learning how to do automated steps is intimidating — I find this is because the smallest of details can derail the entire process for hours. However, as I’ve learned with other technical lessons, this is the best way to learn how to do anything — make a lot of mistakes, learn from those mistakes, do better next time.
Also in the next few weeks, I would like to enhance the CSS files associated with the Hoagy Omeka theme in order to customize it with appropriate images. This will also take some time to figure out, since CSS code is yet another format I will have to brush up on. However, this will also be invaluable practice for dealing with digital exhibitions, and I’m happy to explore.