next message in archive
next message in thread
previous message in archive
previous message in thread
Index of Subjects
> &am
On Sun, 18 May 1997, Edward Dyer wrote:
> Hello Norman!
>
> Looks fine to me!
>
> But you are right, in a way, it doesn't look right to you :-(
> Here's the long explanation:
>
> The problem is this: There are two phases in retrieving a web document
> (i.e. anything that goes through http:)
>
> For a web document, the &#NNN; is supposed to be interpreted according to
> either ISO-8859-1 (ISO Latin-1) or in the proposed revision to use UniCode
> (which includes ISO 8859-1 as a subset in the first 256 characters) by the
> browser (Lynx if you dial in to CCN, your choice if you go by other
> routes.)
>
> But Lynx uses the setting "Display character set" to decide which 8-bit
> code to send to you to display the e-ague character. In order for that to
> work two things are required: your communications software must be able to
============================================
> map that code to the appropriate glyph on your screen, which actually
=====================================================================
> means the comm program must map the received code to a display font, and
=========================================================================
> the display font must include the correct glyph at the correct code.
====================================================================
That doesn't apply to the samples I sent you as I used the "p" (Print)
command to print to my Chebucto directory and then snipped away the
irrelevant text. Although I may have seen a *copy* of it on my screen,
the actual text I sent you never passed through my communications
software. It went straight from the HTML file, through lynx (via "Print")
to a local text file, got snipped and sent to you.
> On the other hand, I hypothesise that if Lynx is not viewing web pages,
> i.e. using file: access, the first stage of the interpretation does not
> occur, in the same way: Lynx displays the file according to your default
> character set, which is by default USASCII, and the character that is sent
> to you has a different 8-bit value. This is probably a Lynx BUG, the way
> we use it, but one might get some discussion on that. Technically, it is
> probably an undefined behaviour.
>
> THE WORKAROUND: set the default character set to ISO 8859-1, and select
> fonts that will work for you.
I'll try that but am not too hopeful. In fact, I'll try a number of
settings to see if characters are shuffled around even more with others.
Whan I am on CCN, I have an ANSI font loaded into the VGA adapter. No
translation is selected at all -- presumably.
Note that at the library they have a 7-bit text terminal. High ANSI
characters are folded to the low characters:
!"#$%&'()*+,-./01...89:;<=>?@AB...YZ[\]^_`ab...yz{|}~
With a local link, the folded "alphabetic" characters come out in the order:
ABCDEFGHIJKLMNOPQRSTUVWXYZ ... abcdefghijklmnopqrstuvwxyz
when you view my htmlchars.html file. With an "http://www.chebucto.etc."
link, you see something like the sequence: (I may have the wrong
characters swapped but you get the idea)
ABCEDFGIHJKMLNOPQSRTUVWYXZ ... abcedfgihjklmnpoqsrtuvwyxz
> THE DOWNSIDE: people receiving mail from you may see a note that you are
> using a different character set, unless they too are using 8859-1.
>
> THE FIX: make Lynx interpret entities in local HTML files using ISO
> 8859-1/Unicode, since we use local references as a shortcut to what are
> effectively web documents. Does this have any downside? need to
> distinguish between HTML and other document types and views, especially
> binary, and source, but I think only HTML and equivalent (htm, html.fr)
> are interpreted anyway. (Aside: some browsers now support up to 5 digit
> numbers in &#NNNNN; to do Unicode - does Lynx?)
>
> Discussion on the merits to CSuite-Dev@chebucto.ns.ca, please.
>
> Ed Dyer aa146@chebucto.ns.ca (902) H 826-7496 CCN Assistant Postmaster
> http://www.chebucto.ns.ca/~aa146/ W 426-4894 CSuite Technical Workshop
> Religion Page Editor, Chebucto Community Network http://www.chebucto.ns.ca
>
> On Sat, 17 May 1997 af380@chebucto.ns.ca wrote:
>
> > Hello.
> >
> > Has anyone found the cause of the annoying bug that displays some accented
> > characters differently depending on the URL used?
> >
> > I can never be sure when I quote a web page with accented characters on it
> > if I am getting them correctly.
[lexographic defoliation]
Norman De Forest
af380@chebucto.ns.ca
http://www.chebucto.ns.ca/~af380/Profile.html
(A Speech Friendly Site)
.........................................................................
Q. Which is the greater problem in the world today, ignorance or apathy?
A. I don't know and I couldn't care less.
.........................................................................
next message in archive
next message in thread
previous message in archive
previous message in thread
Index of Subjects