lynx-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: lynx-dev removing HTMLSRC_* stuff: suggest a new setting name, pleas


From: Klaus Weide
Subject: Re: lynx-dev removing HTMLSRC_* stuff: suggest a new setting name, please
Date: Thu, 6 Jan 2000 02:30:47 -0600 (CST)

On Thu, 6 Jan 2000, Vlad Harchev wrote:

> On Wed, 5 Jan 2000, Klaus Weide wrote:
> > Then make it PRETTYSRC_SPEC, pretty please.
> 
>  Prefix PSRC itself is already used in several places to denote pretty source
> view mode.

That's too bad.

But where?  I only see PSRCVIEW_NO_ANCHOR_NUMBERING in lynx.cfg, which
could be renamed as well.  Variable and macro names in the code don't
count here.

> So I chose PSRCSPEC.
> 
> > That reminds me...
> > The description of "pretty source view settings" is completely
> > incomprehensible, unless one already knows how it works.  Does anyone
> > disagree?  Has anyone tried to create or modify their own -prettysrc
> > formatting, going by the lynx.cfg description, and actually understood it?
> > (I know I didn't before I actually started to fiddle with the source.)
> > 
> > One basic problem - it talks about "internal html markup" but never
> > introduces that term or explains what it does.


>  Here is a quote:
> #  The following group of option tells which styles will surround each
> # lexeme. The syntax of option in this group is:
> #HTMLSRC_<LEXEMENAME>:<TAGSPEC>:<TAGSPEC>
> # The first <TAGSPEC> specifies what tags will precede lexemes of that class
> # in the internal html markup. The second - what will be placed (internally)
> # after it.
> <....>
> #         Examples:
> # HTMLSRC_COMM:B I:!I !B
> #         - html comments will be surrounded by <b><i> and </i></b> in the
> #           internal html markup
> # HTMLSRC_ATTRVAL: span.attrval : !span
> #         - values of the attributes will be surrounded by the
> #           <SPAN class=attrval> </SPAN>
> # HTMLSRC_HREF::
> #         - no special html markup will surround hyperlink destinations (
> #           this means that only default color style for hrefs will be applied
> #           to them)
> 
> 
>  "Examples" could give some insight about "internal html markup".

That really doesn't cut it, unless you want this to be some kind or
riddle or puzzle.  Just as I said - the term "internal markup" is
never introduced or explained.

We read about "styles" surrounding "lexemes" and learn that "tags"
can "precede lexemes" and that somehow this option is for letting us
decide how that is to happen.  But all this is in the context of some
"internal html markup".  Well.  You should say what that is!

After all, theses options (or, now: this option) are (/is) about
styling the -prettysrc display.  A kind of style sheet.  It's not an
expected thing that something like that is achieved by some kind of
"internal html markup".

An honest description should say something like "We are abusing HTML
tags by generating an internal stream of them purely for the purpose
or carrying presentational directives".  Or something like that.
[Optionally add "but that's okay for us to do, because..." if you
can come up with something fitting.]

Anyway,
  HTMLSRC_COMM:B I:!I !B
(ignore that I'm still using the old form, please) really stands for
  HTMLSRC_COMM:<B> <I>:</I> </B>
doesn't it?  Why isn't it written that way?  It would make more sense
and probably make the whole thing a bit more obvious.

All you get by introducing your own abbreviated syntax for this, like
you did, is saving on a bit of parsing code - which can't be that bad.
(You could still require that this must be a minimal sub-HTML, only
tags, no attributes etc.)

Using the abbreviated syntax may be shortsighted.  If you used a
HTML-like form instead for the specifications, you could extend the
format to include arbitrary fixed text strings, for example.  Like
  HTMLSRC_HREF:<B>LINK</B>:
Well that doesn't make too much sense for -prettysrc source viewing
alone, but would be nice for more general styles.

   Klaus


reply via email to

[Prev in Thread] Current Thread [Next in Thread]