lynx-dev
[Top][All Lists]
Advanced

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

Re: lynx-dev javascript? was Re: [...] Lynx will be soon an old ...


From: Martin McCormick
Subject: Re: lynx-dev javascript? was Re: [...] Lynx will be soon an old ...
Date: Sun, 28 Apr 2002 21:01:42 -0500

        I went back and re-did the experiment I did with the
packet capture from the javascript web site and I discovered that
I was wrong about one particular observation.  When one gets to
the relative links, the lynx browser does not generate any
out-bound traffic at all.  This is probably because it does not
see what one is on as a valid link.

        As a test, I tried the Debian Linux web site which always
works in lynx.  Clicking on a link either seems to take one to
that part of the page if it is part of the presently-loaded page
or the browser generates the GET command if the link is somewhere
else.  In other words, it works like it should.

        When I go to the Domino-produced site, the first link is
absolute and lynx generates a GET request.  When I click on any
one of the next layer of links that appears, no traffic is
generated.

        Anyone who is interested can try the following link to
test with.

http://home.okstate.edu/Okstate/evp/cis/CISProjTrk.nsf

        The lynx browser sends quite a bit of capability
information to the server which the server simply ignores.

        Here is a capture of what is sent to the okstate site on
a good link.

        I have manually edited the text portion of the packet so
what you see is not necessarily formatted like it truly is.

GET /WebHome2.ns f/pages/16C4DC092F1DE2B586256BA8 00048257?OpenDocument HTTP/
1.0..        
Host: home.okstate.edu..Accept:         
text/html, text/       plain, text/sgml       , */*;q=0.01..Ac        
cept-Encoding: gzip, compress..Accept-Language:         
en..User-Agent:         
Lynx/2.8.5dev.7         
libwww-FM/2.14..         
Referer: http://home.okstate.edu/WebHome2.nsf/RelatedLinksGrape?         

        I was confused last time on which packet I was looking at
and thought something was being generated on the broken links,
but it is not.

        This is actually very good in one respect.  It means that
the lynx browser is not becoming confused and generating garbage
based upon a misunderstanding.

        I have tried the links browser which, for those using
speech synthesizers is spelled l i n k s and is not the browser
that this list is about and it actually does decode the relative
links which is good.  It also does not support SSL URL's or
mailto links.

        The lynx or l y n x browser is still the most mature
text-based browser as far as I can tell.

        This little exercise has at least for me, made it easier
to understand what is not happening.  

        All the information is there to put the link together.

        Now for another question.  Are relative links peculiar to
ecmascript?

        This is beginning to look to me more like a separate
issue.

        I know that the developers of the Mindleaders.com web
site state that their content is native html and I know that if
somebody sends me a page from that site, it comes through clearly
in lynx with the usual limitations about tables, etc that we know
about.  It's just the links that don't work.

        There is a login form that also doesn't work, but one can
fill it out.  It just can't be submitted.

        For those still following this rambling, thank you for
your patience.  If I can figure out what to do, we will have the
same browser we have right now except that it will be able to
follow more links and be useful when it encounters what it can
read.

Martin McCormick WB5AGZ  Stillwater, OK 
OSU Center for Computing and Information Services Network Operations Group

; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden

reply via email to

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