[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: lynx-dev Vlad's: <PRE>, formatting, screen width
|
From: |
Klaus Weide |
|
Subject: |
Re: lynx-dev Vlad's: <PRE>, formatting, screen width |
|
Date: |
Sun, 7 Mar 1999 06:04:25 -0600 (CST) |
On Fri, 5 Mar 1999, Vlad Harchev wrote:
> It seems that it will be useful to allow lynx to interpret tags
> inside <PRE> block preserving the layout of the source in <pre>.
Yes.
> This is supported by at least Netscape (even <HTML>,<HEAD>,<BODY>
> are not required for NS) and kfm ( <HTML>,<HEAD>,<BODY> tags are required
> for document to be parsed as HTML even for file with .html extension).
Whether or not a given documents is parsed as HTML is unrelated to <pre> -
except that, in order to recognize <pre>, a document has to be handled
as text/html.
> Both of them allow tags like <b>,<i> and even <a>.
So does lynx, as already noted elsewhere.
> This feature allows quick pseudo-html-formatting of the pure-text
> documents
> for which no hypertext source in any form isn't availble by inserting tags
> like <b>,<i> around chapter titles, etc using 'sed'. Currently default
> HTML-parser in lynx, when encounters any tag in <pre> block, discards <pre>
> semanics and treats everything like no <pre> was specified.
As far as I know even older lynx versions don't act like that, at least
for harmless inline elements like B, I.
> It seems to me
> that lynx won't have problems or logical inconsistences displaying such html
> files since it's character based. It's also a good solution for documents
> that
> contain tables - the tables can be rendered with some other tools (may be
> even
> on http server based on the name of the user-agen header) to the html and
> placed within <pre> with normal html formatting.
Sure - lynx could do lots of things if combined with the right external tools.
It just seems that, after all, folks are not *that* interested in table
rendering - otherwise we'd hear more about those solutions.
There is a collection of links to table scripts at the bottom of
<http://www.crl.com/%7Esubir/lynx/patches.html> (the links on that page
seem to be mostly of historical interest).
I posted an example script called gettidy.sh that shows how an external filter
can be called from lynx, in a rather convenient way, without any code changes.
(See the lynx-dev archives for December or January)
[ Not much feedback on that - which really makes me think there isn't that
great need for more hooks to external programs. ]
> If there will be such http
> servers, it will be nice to pass the width of the screen in characters in
> user-agen header like 'Lynx/2.8.2dev.12-80 libwww-FM/2.14`, where 80 is the
> width of the screen. It's likely that GNU-oriented sites will support it -
You are too optimistic about what sites may be willing to do to accomodate
lynx.
In any case, if there is a demand for something like this - please not another
lynx-specific hack of headers. The negotiation features of HTTP should be
used for this, since they are just meant for this kind of thing, although
I don't know the current status of standardization.
Start with <http://gewis.win.tue.nl/~koen/conneg/>, follow some links,
look especially for "feature tags". Al Gilman who is on this list may
know more about the current status - I see he's also on the relevant
ietf-medfree
list.
> such formatting can be achived with Perl scripts on server side probably
> with
> the use of nroff.
Klaus