Re: Proposition on advanced URL features (#3)

Mirsad Todorovac ([email protected])
Tue, 28 Nov 1995 19:48:29 +0100 (MET)


Dan writes:
>
> These are interesting ideas, and they come up from time to time
> on this list, but they don't seem to merit deployment for one
> reason or another. I'd like to see folks hack up support for them
> and tell us how it goes. Heck -- maybe Henrik will hack them into
> the library!
>
> > 1. URL doc.html##123 should mean:
> > "Start display of rendered document at line 123 of resulting
> > document text (not the source).
>
> Err.. what's the definition of a line in HTML?
>
> In plain text, it's much more clear.

Probably it makes sense only there.

>
> In any case I'd suggest something a little more mnemonic, like maybe:
>
> foo.txt#line=123
>
> Of course that takes the = character away from the anchor name syntax,
> but I for one won't miss it.
>
> doc.html#count=3;element=H2

It gives me some feeling of uncertanty. It puts a hard task on a parser, to
decide whether it is anchor name or line addressing. With ## there are
far less things to worry about.

> > 3. URL: doc.html##/foobar/[n]
> >
> > "Set top of display to the first [n-th] occurance of word
> > (desirable regexp) 'foobar' inside rendered doc.html".
>
> doc.html#search=foobar
> doc.txt#search=foobar
> doc#count=3;search=foobar

Dunno. ``;'' is used e.g. in ftp scheme to designate type of file. It would
probably create a lot of confusion if we use it here. Maybe use of comma sign
would be better?

>
> >Please note use of #/ extension in this case:
>
> I think you will lose badly if you attempt to use / in fragment
> identifiers.

How about \/ , like in lots of programs where / is used to delimit search
pattern ?? -- Mirsad