Most of this script works fine for jessie, but some of it doesn't and
the script is actually quite long and complex. Ivan notes he found it
easier to develop it as a cohesive unit that didn't need to worry about
semaphores after doing A2SERVER, and that's obviously going to be
particularly true since he was the only one working on it at the time.
Well, now that's changing we'll want a little more structure. I won't
be breaking things out into their own files right now, but I'm going to
begin organizing things based on where they should go as I re-enable
them.
Basically, we'll port to jessie block-by-block.
The only major difference I noticed that we use is the fenced code block
format for GFM is different (using the backtick instead of the tilde).
If I missed anything, feel free to reopen the bug and tell me what, or
just submit a patch yourself!
Notably this may cause issues diffing adtpro.sh.txt against the
upstreadm adtpro.sh script. Ours had mixed coding styles, and I figured
having to diff with whitespace ignored was a better solution than
dealing with mixed styles either in the file or in the tree for the
majority of people who aren't likely using the same editor I am.
You can now replace http://appleii.ivanx.com/a2cloud/ for script
downloads just as you can with A2SERVER by setting A2CLOUD_SCRIPT_URL in
your shell's environment. As with A2SERVER, this doesn't affect the
canned files in http://appleii.ivanx.com/a2cloud/files/ because these
aren't really part of the source code. Certainly it makes sense to
mirror these, but GitHub isn't the place to do that.
Adding everything to the tree was a more manual process than it was for
A2SERVER because there are a lot mroe files in Ivan's tree, and they're
not all used anymore. More commits were made in the process, but the
end result is that I don't have to go back and spend a bunch of time
undoing my changes for all the URLs that I shouldn't have changed it in
the first place. Also, it gave me a chance to read through the A2CLOUD
setup script, which was a needed exercise.
Still needed is cleaning up of whitespace to ensure everything is
consistent and adding modelines for vim (and humans who need to know how
to set up editors that aren't vim.) Those can come as batch commits,
and will follow soon.
