From: Jimmy Thrasher <jjthrash@...>
Date: 2001-09-30T03:00:02+09:00
Subject: [ruby-talk:21827] Re: ANN: HtmlDomGenerator 0.1

> #    You should have received a copy of the GNU General Public License
> #    along with this program; if not, write to the Free Software
> #    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307
USA
>
> No GPL to be found :-)

Hrm.  I thought there was a line in there that says, "but if you didn't,
here's where to look."  Oops.  :)

> 1. It's not clear (to me) what you mean by "a DOM".  Are you planning
> to implement the DOM API?  Or more of a home-grown tree-based
> interface? (In which case you'd be best off avoiding the term "DOM".)

I suppose I could just call it a tree, but it looks like the DOMs I've used.
;)  However, I was planning in future releases to make it jibe with a real
DOM API.  I whipped this up yesterday and today so I could parse HTML files
much more easily than otherwise.

> 2. If this is a general SGML and/or XML tool, I wouldn't use "HTML" in
> the name.

Agreed.  The only thing is, I don't do any DTD checking, so consider this a
non-validating parser that was convenient at the time for what I want to do.
I shouldn't call it anything, by that token.  Or,
GenericMarkupLanguageTreeStructureGenerator.  :)

> 3. "Well-formed HTML" is sort of a contradiction in terms, or at least
> redundant.  HTML has a DTD; therefore, HTML documents are either valid
> (and, incidentally, well-formed) or invalid.  If a document is
> well-formed but not valid HTML, then it isn't HTML.  (I guess there's
> such a thing these days as "well-formed [DTD-less] SGML", though I
> think that only exists as a way to make sure that all XML documents
> are also SGML documents.)

Right, my stuff is currently a quick hack to get a tree structure I can deal
with.  It may improve, if I don't end up using something else.

> 4. I'm not sure what "handling" interleaved tags would entail, other
> than dying with a fatal error :-)

Ask Netscape or IE.  :)  Or HtmlTidy.

> 5. You might get some mileage out of existing Ruby tools in this area,
> including Jim Menard's NQXML, which has a tree parser.

Would NQXML handle HTML <br> tags?  I need something that will handle HTML.

Shall I update the description and names to communicate the lack of
DOM-iness, and lack of validation?  Or does it matter?

Jimmy