Re: converting Frame to down-level HTML [ONLY MAP TO <p>]

[jeremy@xxxxxxxxx (Jeremy H. Griffith)]

On Wed, 30 May 2001 09:35:54 +1000, hedley_finger@myob.com.au wrote:

>You can successfully do this up to a point but then come the dreaded lists
>and tables.  You have to somehow detect the first item in a list, and
>decide whether the list is bulleted (<ul>) or numbered (<ol>).  Then you
>have to determine the components of each list item: the [bulleted or
>numbered] element itself <li> and then any <p> elements that are a
>continuation of the list item.  For example, the numbered step in a
>procedure is the <li> list item and the explanatory paragraph under it
>would be a <p> indented paragraph WITHIN the <li> element.
>
>Lists get terribly exciting when you have to determine whether the last
>[bulleted or numbered] para REALLY ends the list, so you can insert an
></ol> or </ul> terminal tag or whether that plain paragraph <p> after the
>last list item <li> is the explanatory continuation that REALLY ends the
>list.

This is indeed challenging, and there's another problem with it.  If
you want to use CSS with the <ol><li>...</ol> (or <ul>) tags, you have
to deduct the amount of indent applied automatically by the browser
from your CSS indent, or you'll see your <li>s sliding away to the
right like elections in Florida.  How much do you deduct?  Why, that's
browser dependent... so you can't have one CSS for all browsers, and
have to play the JS detection game to select among multiple .css files
at runtime.  Ack.  We do support that method, even provide the JS, but
one could hardly call it standards-friendly.  So we generally leave the
lists alone, to be handled as <p> items, which looks better...  The
only time this won't work is if you have to support non-CSS browsers,
which thankfully are becoming an endangered species.  None too soon,
we say... bring on CSS2!  Please!

-- Jeremy H. Griffith, at Omni Systems Inc.
  (jeremy@omsys.com)  http://www.omsys.com/

** To unsubscribe, send a message to majordomo@omsys.com **
** with "unsubscribe framers" (no quotes) in the body.   **