On Tue, Jul 8, 2008 at 8:19 AM, Kent Tenney <[EMAIL PROTECTED]> wrote:

>
> I understand that the .leo tree holds files _with_ sentinels, which Leo
> uses
> to build the outline.


No.  The Leo tree is distinct from derived files.  Derived files may, or may
not, contain sentinels.  Internally, a Leo tree is a tree of nodes: this is
the Leo "DOM" (document object model).  Externally, a Leo tree (.leo file)
is an xml file containing <v> and <t> elements, among other things.

*Please* do not confuse all these.  If you do you will simply be confusing
yourself and others.

Now to the issue of "chunking".  You are, of course, free to slurp in a file
without doing anything else.  s = f.read() reads the open file f, and it
would be dead easy to write a script to make a node containing a file.  As I
have said, repeatedly, there are plugins to do this.  I'm tired of
discussing this.

However, it was *your* suggestion that lead to @auto, and I consider @auto
to be a most happy development. In particular, @auto provides an excellent
*starting* point for @shadow.  @auto comes close to recreating the style of
outline that I use in Leo's core.  I've used this style almost unchanged
since day one, ca. 1986.  Converting @auto to @thin or @shadow will then
allow me to make teaks.

Of course, if you don't like what @auto does, you can create your own
starting point.

True, @auto imposes some (very minor!) constraints on the original (pypy)
sources, but those constraints are *essential* if we are going to use Leo as
it is typically used.  If you *ever* convert from the one-node-per-file view
to a more typical one-node-per-class-or-method view you will be forceed to
clean up leading whitespace and underindented comments.

Let's turn now from what is, in essence, a discussion of preferences (how
files are chunked), and look at what @shadow will do regardless of how files
are chunked. @shadow allows you and me to play nicely with projects such as
pypy:

1. You and I can represent the pypy files as *we both, independently* like
and

2. Minimal (or no) changes are needed to the pypy files themselves.

How could it get any better than this?

For example, (and this is important) you would be free to create one giant
node for each pypy file, and the *public* files would be exactly the same
(when you converted the file to @shadow) as *I* would get if I used @auto.
*You* get to represent the files as *you* want, but you and I get the *same*
public pypy files (assuming neither you nor I actually changed the imported
outline).  The only difference between *your* view of the pypy world and
*my* view would be that your *private* files would be different from mine!

Do you see how excellent this is?

Edward

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/leo-editor?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to