Are these the best set of bitset macros in the world or what!?

[BartC]

its ok for those of with US (or near US) keyboards to look down our
noses at trigraphs and <iso646.h> but many europeans (and the rest of
the world!) have many important characters (to C programmers) missing
from their keyboards

Which characters are missing?

I remember even my first computer keyboard (a teletype) had pretty much all
of the characters that might be needed. What it didn't have was lower case,
which you'd think would be more of an issue for C code.

 

[BruceS]

It never ceases to amaze me.  I sorta dread to some day go back and see
whether I was as insufferable when I was just starting out.  Hmm.

I think a better question would be whether you spent much time
trolling a group's regulars, and if so, how successful you were.

 

[Ben Bacarisse]

I don't think Eric was saying that there was no problem, but that
standardising the solution in C was unnecessary. In those days one
accepted that there would be problems transferring course code between
systems so I can image solving the problem with a source-code character
mapping utility that had nothing to do with C.

Standardising this mapping helps, but why do it in C? The problem
exists for many languages so a better solution might have been a
portable representation for any text that needs characters not easily
typed (or represented) on some system. C would then be trigraph free,
and there would be a single solution suitable for many texts.

Which characters are missing?

IIRC the most common omissions were {, }, [ and ]. These were not
directly available on many keyboard layouts (particularly Scandinavian
and French keyboard). PC keyboards typically had these as AltGr keys
on the numbers but the problem pre-dates such sophisticated things!

I remember even my first computer keyboard (a teletype) had pretty
much all of the characters that might be needed. What it didn't have
was lower case, which you'd think would be more of an issue for C
code.

Yes, it would be. I never had to write C with such a thing. BCPL, yes,
but then BCPL's keywords were not lowercase.

Algol 68 addressed this issue by not specifying how the abstract token
of the language were to be represented in the source. One could choose,
to some extent, how keywords were written. "UPPER stropping" (WHILE,
IF, etc) and "POINT stropping" .while. .if. were both common as was
using one or two quotes. Most symbols such as the multiplication
operator had alternate an spelling (often as another symbol but also as
a keyword) which could be typed on almost any system.

 

[Nick Keighley]

Which characters are missing?

I remember even my first computer keyboard (a teletype) had pretty much all
of the characters that might be needed. What it didn't have was lower case,
which you'd think would be more of an issue for C code.

some european languages have extra letters and have to lose other
characters to allow for them. I think []{} were troublesome. The only
pictures of Danish keyboards I can find seem to have them- though some
keys have three or four different meanings. Danish typewriters show
three extra letters with lots of punctuation missing. But I suspect
even an American typewriter would probably be inadequate for C.

You need to find someone who uses one of these languages and ask them!

 

[lawrence.jones]

its ok for those of with US (or near US) keyboards to look down our
noses at trigraphs and <iso646.h> but many europeans (and the rest of
the world!) have many important characters (to C programmers) missing
from their keyboards I believe the Danes in particular held out for
trigraphs. Of course life presumably became better with PCs where we
can map whatever we like to say the function keys.

Trigraphs and <iso646.h> were political solutions to political problems;
they were never a good technical solution and they were never necessary
(there was almost always some reasonably convenient way to type
characters that would be interpreted correctly, even if they looked a
little strange [not that they looked any stranger than trigraphs do]).
The Danes (and others) had a *philosophical* objection to an
*International* Standard that apparently required characters that were
not part of the international character set standard but were in the US
character set standard.

 

[lawrence.jones]

Which characters are missing?

The ones that there are trigraphs for, of course!

 

[Marcin Grzegorczyk]

I just want to know, if he's not willing to work with other people or reuse
anyone else's code, how he expects to make enough money to hire contractors?

It occurred to me too, but I decided not to tackle the subject, because
I realized any such suggestion would likely bruise his ego enough to
provoke another rant. (Apparently it didn't matter; he stopped posting
here anyway.)

BTW, those following along with my fascination with pathological narcissism
on the Internet may find it interesting to consider the implications of
"Brian"'s obsession with ways in which he can discard competing authorities
and have full ownership of and control of other people's time and creativity.

Yeah, I found him sort of fascinating, too. A rockstar programmer
attitude if I ever saw one. Well, it seems he's gone now, but I did
find some of his responses truly amusing :->

 

[Marcin Grzegorczyk]

Trigraphs and <iso646.h> were political solutions to political problems;
they were never a good technical solution and they were never necessary
(there was almost always some reasonably convenient way to type
characters that would be interpreted correctly, even if they looked a
little strange [not that they looked any stranger than trigraphs do]).
The Danes (and others) had a *philosophical* objection to an
*International* Standard that apparently required characters that were
not part of the international character set standard but were in the US
character set standard.

On a side note, now that ISO 10646 has been around for 17 years, I
wonder if it might be the time to deprecate trigraphs...

 

[Ian Collins]

Trigraphs and <iso646.h> were political solutions to political problems;
they were never a good technical solution and they were never necessary
(there was almost always some reasonably convenient way to type
characters that would be interpreted correctly, even if they looked a
little strange [not that they looked any stranger than trigraphs do]).
The Danes (and others) had a *philosophical* objection to an
*International* Standard that apparently required characters that were
not part of the international character set standard but were in the US
character set standard.

On a side note, now that ISO 10646 has been around for 17 years, I
wonder if it might be the time to deprecate trigraphs...

Both the compilers I use (gcc and Sun cc) disable them by default. I
guess others do as well, so they are effectively deprecated.