Getting patches out of HTML archives (was Re: EGCS web pages)

[Jeffrey A Law]

  In message < 199810291834.KAA07632@smtp.earthlink.net >you write:
  > >>>>> "Jason" == Jason Molenda <jsm@cygnus.com> writes:
  > 
  >     Jason> However, the one benefit that we gained by not HTML'ifying
  >     Jason> code is that no matter how the user saves the file, they
  >     Jason> can run patch on it.  e.g.  they can save it as "Source"
  >     Jason> (HTML), and patch will work.  Or they can save it as "Plain
  >     Jason> text" and have a good chance of it working correctly.
  > 
  > That's what I meant. I said `cut-and-paste' as a shorthand, but
  > `save-as-source' is what I really do.
Right.  This is precisely what I do with them too.

Now what gets more interesting is extracting a mime-encoded testcase from
the archive.  The only thing that worked was to hack up enough message
headers to make metamail happy.  Unpleasant.

jeff