In bug #64155, Deri pointed to this snippet of font/devpdf/devpdf.am

          if test -f $$f; then \
            $(INSTALL_DATA) $$f $(DESTDIR)$(devpdffontdir)/$$f; \
          else \
            $(INSTALL_DATA) $(top_builddir)/font/devpdf/$$f \
              $(DESTDIR)$(devpdffontdir)/$$f; \
          fi; \

as the culprit of an install problem he experienced: "I had a corrupt copy of U-TR in my build directory so this code copied the corrupt file rather than the correct file in ./font/devpdf."

This snippet in question dates back to commit 87046b294, in which gropdf was first added to the groff tree.  The current version of the snippet is substantively unchanged from then.  (Some variables have been renamed, and the code was relocated from its original font/devpdf/Makefile.sub.)

Git credits this commit to Werner (and he may have been responsible for the updates to the Texinfo manual and other integration tasks) but the ChangeLog entry credits Deri, and Deri's name appears as the author of the new source files, including Makefile.sub.

This historical tour is relevant because when Deri encountered the corrupt-file propagation in 2024, he asked of the offending lines of code, "What is the purpose of this?", revealing that even the code's original author no longer recalls what it was intended to do.

So from the viewpoint of today, does this code do something useful, and the corrupted file getting copied was collateral damage?  Or should it be revamped?