Thanks in advance-
~~~~~~~~~~
Please respond via the newsgroup - all addresses are ficticious.
> I have Pine 3.96 running on a home linux system
I don't know specifically about linux, but I use PINE on various unix
and non-unix systems. Note that my answer here does _not_ apply to
PC-Pine, for which, see the PINE Q&A:
http://www.washington.edu/pine/QandA/ , specifically, question 4.7.
>- every so often, I get
> the "Cannot display ISO-8859-1 (or whatever) character set. Your system
> is set for the US-ASCII character set. Some characters may not ... " etc.
Your configuration tells it that. If in fact your environment supports
iso-8859-1, you should configure PINE accordingly.
> My question is, where does pine get the information about what character
> set is in use?
In use by what?
If you're asking about the incoming mail, PINE gets the information
from the incoming mail header (where else?).
If you're asking about PINE's understanding of what character set
_you_'re using, then the answer is that you're supposed to tell it,
via the configuration.
> In Linux, I can use the fontconfig utility to choose from
> a variety of screen fonts, and as far as I can tell the default character
> set is Latin-1, but these changes never seem to get communicated to Pine.
I don't believe there is a general mechanism for PINE to find out what
character set an underlying terminal environment thinks it is using.
It's up to you to define one.
There are defined mechanisms (ISO-2022) for an equipment to support
multiple character sets, and to be switched programmatically between
them, but this mechanism isn't very widely used, apart from some
specific codes (iso-2022-jp, iso-2022-kr). Otherwise, PINE just issues
a caution if there is a discrepancy between the message's charset and
(what it thinks) your display charset is. If, in fact, you change your
environment's charset to the one that the message wants to use, without
bothering to tell PINE, then it will work, and you can disregard PINE's
alert. For the nitty details, see the PINE technical notes, e.g
http://www.washington.edu/pine/tech-notes.4.02/low-level.html#char-set
As for outgoing mail, PINE has a helpful feature of looking at your
composed mail and if it doesn't contain any high-bit-set characters then
it advertises the character set as us-ascii. That's helpful to your
recipients, not particularly to you ;-)
What's annoying is that one receives large numbers of incoming mails
(composed by mail user agents other than PINE) that claim to be needing
koi8-r, iso-8859-2, iso-2022-jp etc. in spite of the fact that they
contain nothing more than us-ascii. PINE then nags one about possible
difficulties in reading this mail, when there are in fact no problems
with it. Maybe PINE should check the displayed mail for high-bit-set
characters too, and suppress the warning. Just a thought.
For a more ambitious approach, PINE could learn some useful tricks from
the latest versions of Lynx, but I guess the handling of international
character sets hasn't got such a high priority for PINE's own user
community, and it's understandable that they'd set their priorities
accordingly.
--
"Lynx entsorgt Designschrott aller Art gründlich und umweltfreundlich."
- Uwe Waldmann
> I have Pine 3.96 running on a home linux system - every so often, I get
> the "Cannot display ISO-8859-1 (or whatever) character set. Your system
> is set for the US-ASCII character set. Some characters may not ... " etc.
> My question is, where does pine get the information about what character
> set is in use? In Linux, I can use the fontconfig utility to choose from
> a variety of screen fonts, and as far as I can tell the default character
> set is Latin-1, but these changes never seem to get communicated to Pine.
As such, Pine doesxnot know what your hardware display is
physically capable of at any given moment. Instead, in your
configuration is a field (toward the bottom), "character-set." The
normal default is "US-ASCII." When a message comes in, Pine examines
the headers for some indication that the message may be in a character
set other than that which Pine is told about in your configuration. If
there is a mismatch, Pine displays the warning. This is strictly a
software situation which has nothing to do with what your screen may
physically set to via, say, a fontconfig utility in Linux or MODE CON
commands in DOS.
--
Paul <pob...@access.digex.net>
..........................................................
Paul O. Bartlett, P.O. Box 857, Vienna, VA 22183-0857, USA
Finger, keyserver, or WWW for PGP public key
Home Page: http://www.access.digex.net/~pobart
: For a more ambitious approach, PINE could learn some useful tricks from
: the latest versions of Lynx,
That would be great. :-)
Interestingly, though Pine displays the ISO/US-ASCII warning, it uses
Latin-1 characters internally, to represent non-breaking spaces. (These
are displayed from the internal HTML viewer, and in Help.) This is fine
when I'm using Pine in an xterm, but the characters appear as "a"'s with
acute accents when I run Pine on my Linux console, where I use an IBM PC-
style character set (CP 437). I found the place where the "nbsp" entity
was defined in the Pine source, and I tried changing it to a plain space;
but that caused the blank line between paragraphs in the HTML viewer to
disappear. (I then tried changing it to 0xff, which is a space in CP 437,
but that showed up as a question mark.) I'm still looking for a good
solution to this.
Apart from that quibble, I'm delighted with Pine 4.02A, having upgraded
from 3.95. The new features are amazing.
--
William McBrine | http://www.clark.net/~wmcbrine/
wmcb...@clark.net | ./\./\./\./\./\./\./\./\./\./\.