[REBOL] Re: delay, RE: Re: steel update
From: maximo:meteorstudios at: 22-Aug-2003 11:04
> -----Original Message-----
> From: Andrew Martin 
> Sent: Thursday, August 21, 2003 4:29 PM
> But wouldn't it be easier for the author, instead of writing:
> <img src=href<file! "images/img.gif">>
> to write:
> <img src=href<file! "img.gif">>
> And let the software work out the correct path? And if there's two
> "img.gif" files, the software could choose the file that
> closest to the
but that's dangerous, if there are two img.gif on disk, then they should be be different...
letting the software choose which one to use will eventually cause it to choose the bad
The point of the <file!> tag is to have relative paths from absolute paths. This is
because with the menu, I am sharing the site menu and tools menu in 30 different pages.
If everything was trying to use paths as you describe, none of them would work. As a
quick example, I have an index.html in every directory...
But it would only take you 5-10 lines of code to add another tag (or replace this one)
which does what you say... that's one of the strenghts of remark.r you prebuild html
tags and they get executed when they are persistently and recursively encountered...
which takes up less space than putting actual rebol code between the brackets..
> And why not use Rebol values? Like writing:
> <img src=%img.gif>
It actually is possible, it gets converted to a string right at the start of parsing.
I do have an <img!> tag, here's how its used:
<img! %path/file.gif> can't really be easier. ;-)
and for links
<link! "path/file.html" "Press Here"> or <link! %path/file.html "any html stuff <img!
it might be improved to expect an image (and build up the <img> tag by itself) when a
path is supplied for the second argument, like in...
<link! %path/file.html %path/file>