I think we can proctice some conversion now.
All we need is a simple thing like Eugene
(efbxt - eugene's first bootstrap xml translator) made
to put XML wrappers around important divisions
that it can detect. If this is simple untagged email,
then it looks like we can pick up paragraphs.
For an html page, it may be just a single tagset
around the whole thing, it doesn't matter.
The important thing to think about is the markup
that identifies each block that that efbxt picks out.
Essentially, efbxt should be thinking of generating
a global 'purple number' for each block it identifies.
Doug's purple number are a search and retrieve items.
For example, when you see the tag DkrId in example
markup from me, it represents a unique namespace
for a specific archive (DKR). For example, if Joe is
running efbxt, then the tagset might be:
<DkrId>
<JoesDkr0>
<message id="001>
<DkrId:p sid="001">
first block
</p>
<DkrId:p sid="002">
second block
</p>
...
</message>
</JoesDkr01DkrId>
This would establsh define a unique purple number for
each block in this document and give its location
as a certain block in a certain message
in a certain subarchive in the total archive namespace.
So the complete address of the second block in this
message is:
DkrIdJoesDkr0001002
This can be used to generate a purple number
unique for each block in the DKR.
Another reason the purple number is important,
one not fully exploited now, is its use in a
copy and paste or a move operation.
Including the purple number provides attribution.
Thanks,
Joe
----- Original Message -----
From: "Grant Bowman" <grant@suse.com>
To: <ohs-dev@bootstrap.org>
Sent: August 27, 2000 10:22 PM
Subject: Re: email to XML Conversion Example 1
> * Eugene Eric Kim <eekim@eekim.com> [000826 04:40]:
> > Not necessary to run my program. It doesn't do anything to included
> > markup right now. Keep in mind that the code I wrote was a quick hack
to
> > get us started, not as a final system. It's clear to me now that we
> > weren't ready to start talking about code.
>
> I don't know that's true. I think if we have actual HTML, XML, XSL and
> Augment samples we can certainly write code to translate, between them,
etc.
>
> Regards,
>
> --
> -- Grant Bowman grant@suse.com
> -- SuSE +1-510-628-3380 x5027
> -- 580 Second Street, Suite 210 fax +1-510-628-3381
> -- Oakland, CA 94607 http://www.suse.com
>
>
This archive was generated by hypermail 2.0.0 : Tue Aug 21 2001 - 17:57:54 PDT