[Date Prev][Date Next][Subject Prev][Subject Next][ Date Index][ Subject Index]

Re: DELTAGS



Carl Distefano wrote:

> Reply to note from "J. R. Fox"  Mon, 30 Sep 2002
> 17:05:15 -0800
>
> I tried DELTAGS with your sample document, and it works fine -- that
> is, once the doc is converted into HTML! Did you decode the quoted-
> printable portions of the document before running DELTAGS? That's a
> must. Otherwise what you're handing to DELTAGS isn't native
> HTML|XML|SGML, and therefore doesn't constitute a fair test of what
> DELTAGS is designed to do.

Thanks for your reply. No, when I'm looking at that stuff, it's all
Greek to me.

> Try this: Extract the HTML document to a separate XyWrite window,
> everything from  down to . DeFine the entire window with
> DFA, then issue QPD to decode the quoted-printable
> parts ("all the ` =20 ' stuff etc."). Now you've got a real HTML
> document that's fair game for DELTAGS.

You're exactly right, of course. The changes you made last June
resolved most of the issues I see with any frequency.

> Any document that contains quoted-printable codes (=XX, where XX is
> a hex number in range 00-FF), has to be treated this way before
> being fed to DELTAGS.

So, that's the main tip-off I should be looking for ?

> QP encoding is common with HTML and other
> almost-but-not-quite-straight-text docs sent by e-mail; many mail
> programs do such encoding automatically if they detect an HTML tag.
> So, when in doubt, run QPD.

Well, I never had to do this before (and I haven't seen this often), but
I guess that source changed their page creation or transmission
procedures. I must comment here that QPD takes a very long time to
parse a file like that . . . which wasn't an inordinately large one.
Which makes me wonder, in situations where that is the case (and not
just for QPD), is there not some alternate parsing method that could be
chosen, done all in memory and not updating the screen page ? If so,
and it went a whole lot quicker, that would tend to obviate the concern
about users thinking the routine may have locked up.

> I hope this helps. Once you're satisfied that DELTAGS is working as

> advertised, I'd appreciate your putting a short note to the Xy list
> to say so. We like our "public" to be reassured that we're
> responsive on queries about bugs and "features".

Done. DELTAGS works well for me, in nearly every case. This was not a
good example of one of the exceptions. In fact, DELTAGS often works
better for me than trying to Save as Text in Netscape (might not
actually render the page as text at all, sometimes Traps with a c0005
exception), or the well behind-the-times HTML2TXT plugin for
PillarSoft's Enhanced Editor (goes off into Never Never Land and does
not return). I mostly use their E editor for clipboard transit stuff in
OS/2, because I have never been able to get those U2 CLIP frames to work
for me, no matter what I tried.

> OTOH, if you have other docs that illustrate what you believe are
> problems with DELTAGS, by all means send them along and I'll take a
> look.

Thanks, will do. I'm sure there are / will be other HTML things that
create problems for it.

Jordan