Re: lynx-dev looking for alternative method for printing a web page

[David Woolley]

> 
> This msg is a tip as well as a request.
> 
> The tip is this.  For those of you on Unix, with the addition of a few
> programs, you can get very nice output from lynx.  To do this, you need
> 
> 1. perl <URL:http://www.perl.com/>
> 2. ImageMagick <URL:http://www.wizards.dupont.com/cristy/ImageMagick.html>

You don't need this; there are alternative image handling options (although
this may be the best); netpbm can be used.

> 3. html2ps.pl <URL:http://www.tdb.uu.se/~jan/html2ps.html>
> 
> I _think_ that's all you need <grin>

If you have ghostscript, you can generate PDF files from this, complete
with "bookmarks" for the headings - the HTML 4.0 spec was prepared for
PDF this way, although the Adobe tools where used to create the actual
PDF.  With psutils, you should be able to produce A5 booklets, and, with
a bit of stitching and glueing, A5 books.  (Americans: replace A5 with
the result of folding letter, etc., in half.)

As I said, when suggesting this to David Combs, as an alternative to varying
screen widths to get double column, it does assume valid HTML.  I was 
wondering whether tag soup mode internally creates a (very) transient
valid HTML version, and therefore if it might be possible to have an
option to output that.

From the documentation, it looks like it may have difficulty with the
more excessive uses of tables etc.  It doesn't do frames.

I suspect one might need to use the second %s on the print command to
allow a recursive call of Lynx, or wget, to get the file.  If the second
%s is mangled, maybe an environment variable ought to be set to the 
raw URL.