Hi everyone,
This morning I realized that last night I forgot to mention one of the key reasons why I was talking about the changes to Migrate in 8.1. The reason was that much of the documentation that you'll find out on the Internet about Migrate in 8.x pertains only to 8.0. This means that a lot of that info is out of date and cannot be relied upon! Even the official documentation on D.o has not been completely updated for 8.1. This makes for a confusing and frustrating experience when trying to learn how to build your own migrations.
In general, if you come across a blog post or other doc that mentions "migration templates" then that information pertains to 8.0. Migration templates were changed to the YAML "migration plugins" in 8.1. Hopefully knowing this will help you all filter out any out-of-date info.
Here are some links to some of the pages that I showed:
Mike Ryan's blog post
The migrations are plugins change record
Migrate Plus module
Migrate Tools module
Wordpress Migrate module (dynamic migration builder)
Other resources I didn't show that might be useful:
Another Mike Ryan blog post about the XML and Json parser plugins for Migrate Plus
Migrate Source CSV module (an example of a Migrate Plus source plugin)
@kvanderw
I should have mentioned last night that the Sarnia example site that I showed, https://nutritionhistory.nal.usda.gov/, does exactly what you talked about needing to do for your client. The documents in that Solr core are OCRed print publications. We then offer download links for the document in various formats. Those links are dynamically generated by Views from the documents' ID number using field rewriting.
As an aside, those links don't point to any actual files on the server. When a user clicks on one, the request gets forwarded to a document server which dynamically outputs the file in the requested format.
You may wonder why there is no body text shown on any record page. I don't remember if it was an intentional choice not to display it or if the documents are indexed, but not stored. Of course the latter would mean that the body text is not available in a readable format in the index. If your documents have the body text stored in the index, then there's no reason why it couldn't be displayed.