DOI / MD5 Resolver

I wrote my own DOI resolver with an associated document database to serve original research papers to students of my classes, in particular old manuscripts that are hard to come by and might not even have a DOI (like many of the foundational papers on relativity by Albert Einstein). The resolver is used in my lecture notes and for exercise assignments to provide links that students can use to download research papers easily.

If you enter a DOI in the search mask, the following can happen:

• The document exists in my database.
  • If the paper is freely available (open access or in the public domain), the PDF file opens directly.
    Example: https://doi.itp3.info/10.1103/physrevlett.45.494
  • If the paper is not freely available, you are asked to enter an access key before you can download the PDF file. This key is made available to students of my classes.
    Example: https://doi.itp3.info/10.1103/PhysRevB.83.035107
• The document does not exist in my database.
  In this case you are redirected via doi.org to the journal’s webpage.
  Example: https://doi.itp3.info/10.1103/PhysRevLett.134.084101

There are some papers in my database that do not have a DOI.
These are identified by the MD5 hash of their PDF file.
Example: https://doi.itp3.info/650aebc80e4c295a6c3d40cd0df42889