entering Unicode characters as hex numbers not documented

entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

Please add documentation and index entries (for `unicode', `ucs',
`utf-8', etc.) for the function `ucs-insert'!  IMHO this is a very
essential function, and after the release it will become quite
important, I think.


    Werner

Re: entering Unicode characters as hex numbers not documented

[Eli Zaretskii]

> Date: Thu, 01 Jun 2006 09:20:45 +0200 (CEST)
> From: Werner LEMBERG <wl@gnu.org>
> 
> Please add documentation and index entries (for `unicode', `ucs',
> `utf-8', etc.) for the function `ucs-insert'!  IMHO this is a very
> essential function, and after the release it will become quite
> important, I think.

Evidently, someone thinks that it's not worth documenting, since it is
marked as such in etc/NEWS.

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

> > Please add documentation and index entries (for `unicode', `ucs',
> > `utf-8', etc.) for the function `ucs-insert'!  IMHO this is a very
> > essential function, and after the release it will become quite
> > important, I think.
>
> Evidently, someone thinks that it's not worth documenting, since it is
> marked as such in etc/NEWS.

Hmmm.  I don't understand this decision.  How shall Joe User find this
function?  Even I who am quite accustomed to the Emacs source files
had to search for more than 10 minutes because the function's name
neither has `unicode' nor `code' nor `character' in its name, and an
`apropos unicode' doesn't give anything useful.  I didn't give up
because I *knew* that there is such a function but forgot its name.

Additionally, there are a lot of Unicode characters which aren't
accessible with quial input methods.  How do you enter such characters
then if not with `ucs-insert'?


    Werner

Re: entering Unicode characters as hex numbers not documented

[Eli Zaretskii]

> Date: Thu, 01 Jun 2006 16:46:50 +0200 (CEST)
> Cc: emacs-devel@gnu.org
> From: Werner LEMBERG <wl@gnu.org>
> 
> How shall Joe User find this function?

I don't know who decided that and why, but are you saying that we
should document every single command in the manual?

> Even I who am quite accustomed to the Emacs source files
> had to search for more than 10 minutes because the function's name
> neither has `unicode' nor `code' nor `character' in its name, and an
> `apropos unicode' doesn't give anything useful.

If `apropos' doesn't find something, try `apropos-documentation',
which searches the doc strings, not only the symbols' names.

> Additionally, there are a lot of Unicode characters which aren't
> accessible with quial input methods.  How do you enter such characters
> then if not with `ucs-insert'?

Examples will help.  In general, characters that are very popular
should be added to the respective input methods.

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

[-- Attachment #1: Type: Text/Plain, Size: 1731 bytes --]


> > How shall Joe User find this function?
> 
> I don't know who decided that and why, but are you saying that we
> should document every single command in the manual?

Please don't generalize what I've written.  This particular function
is so important IMHO that it deserves an entry in the manual, and it
should be inserted at a very prominent place.  For comparison, even
Windows has a similar function which is well documented -- and these
guys have paid millions of dollars to find out what users need...

> If `apropos' doesn't find something, try `apropos-documentation',
> which searches the doc strings, not only the symbols' names.

Good point, thanks!  I wasn't aware of this function which is
extremely helpful.

> > Additionally, there are a lot of Unicode characters which aren't
> > accessible with quial input methods.  How do you enter such
> > characters then if not with `ucs-insert'?
> 
> Examples will help.

???  What kind of examples do you mean?

> In general, characters that are very popular should be added to the
> respective input methods.

I'm not sure whether we are talking about the same thing.  For
example, I have been asked more than once how to find the Chinese
character with the highest number of strokes.  This particular
character doesn't have a known reading, and it isn't covered by *any*
input methods which are part of Emacs.  The same holds for at least
10000 other Chinese characters, especially the recent additions from
Unicode 4.0.


    Werner


PS: The character in question is U+2A6A5 (�, CNS plane 5, character
    0x7C51), four times the symbol for `dragon'; it consists of 64
    strokes.  Now tell me how you would input this.

[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

[-- Attachment #1: Type: Text/Plain, Size: 168 bytes --]


> PS: The character in question is U+2A6A5 (�, CNS plane 5, character
>     0x7C51),

Uh, oh, Emacs converts this character to U+FFFD.  Here it is again: �


[-- Attachment #2: Type: text/plain, Size: 142 bytes --]

_______________________________________________
Emacs-devel mailing list
Emacs-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-devel

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

> Uh, oh, Emacs converts this character to U+FFFD.  Here it is again: ^[$,3u=^[(B

@#$%^&!  This time it should work: ^[$(K|Q^[(B


    Werner

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

> @#$%^&!  This time it should work: ^[$(K|Q^[(B

Almost :-)


    Werner

Re: entering Unicode characters as hex numbers not documented

[Eli Zaretskii]

> Date: Thu, 01 Jun 2006 09:20:45 +0200 (CEST)
> From: Werner LEMBERG <wl@gnu.org>
> 
> Please add documentation and index entries (for `unicode', `ucs',
> `utf-8', etc.) for the function `ucs-insert'!

Btw, why did you ask for an index entry `utf-8'?  This function has
nothing to do with UTF-8.

Re: entering Unicode characters as hex numbers not documented

[David Kastrup]

Eli Zaretskii <eliz@gnu.org> writes:

>> Date: Thu, 01 Jun 2006 09:20:45 +0200 (CEST)
>> From: Werner LEMBERG <wl@gnu.org>
>> 
>> Please add documentation and index entries (for `unicode', `ucs',
>> `utf-8', etc.) for the function `ucs-insert'!
>
> Btw, why did you ask for an index entry `utf-8'?  This function has
> nothing to do with UTF-8.

Except for the hoi polloi, for which a character set and its encoding
are synonymous.  We are going to have a lot of fun with explaining
that difference in Emacs 23...

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: entering Unicode characters as hex numbers not documented

[Eli Zaretskii]

> Cc: Werner LEMBERG <wl@gnu.org>,  emacs-devel@gnu.org
> From: David Kastrup <dak@gnu.org>
> Date: Thu, 01 Jun 2006 14:57:29 +0200
> 
> > Btw, why did you ask for an index entry `utf-8'?  This function has
> > nothing to do with UTF-8.
> 
> Except for the hoi polloi, for which a character set and its encoding
> are synonymous.

Werner hardly belongs to hoi polloi, as far as character encoding
issues are concerned.

Re: entering Unicode characters as hex numbers not documented

[David Kastrup]

Eli Zaretskii <eliz@gnu.org> writes:

>> Cc: Werner LEMBERG <wl@gnu.org>,  emacs-devel@gnu.org
>> From: David Kastrup <dak@gnu.org>
>> Date: Thu, 01 Jun 2006 14:57:29 +0200
>> 
>> > Btw, why did you ask for an index entry `utf-8'?  This function has
>> > nothing to do with UTF-8.
>> 
>> Except for the hoi polloi, for which a character set and its encoding
>> are synonymous.
>
> Werner hardly belongs to hoi polloi, as far as character encoding
> issues are concerned.

I did not realize that the index was just for Werner.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: entering Unicode characters as hex numbers not documented

[Richard Stallman]

I think the reason for not documenting `ucs-insert' in the Lisp Manual
is that it serves a specialized purpose that will disappear in the
subsequent release.

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

> I think the reason for not documenting `ucs-insert' in the Lisp Manual
> is that it serves a specialized purpose that will disappear in the
> subsequent release.

How that?  A quick alternative to the `ucs' leim input method is
something which is always needed.  Or is there planned a different
method?


    Werner

Re: entering Unicode characters as hex numbers not documented

[Eli Zaretskii]

> Date: Fri, 02 Jun 2006 06:08:28 +0200 (CEST)
> Cc: eliz@gnu.org, emacs-devel@gnu.org
> From: Werner LEMBERG <wl@gnu.org>
> 
> 
> > I think the reason for not documenting `ucs-insert' in the Lisp Manual
> > is that it serves a specialized purpose that will disappear in the
> > subsequent release.
> 
> How that?  A quick alternative to the `ucs' leim input method is
> something which is always needed.  Or is there planned a different
> method?

I think C-q will do what you want in the Unicode Emacs.

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

> > How that?  A quick alternative to the `ucs' leim input method is
> > something which is always needed.  Or is there planned a different
> > method?
>
> I think C-q will do what you want in the Unicode Emacs.

Fine, so after switching to Emacs-Unicode we simply do
s/ucs-insert/C-q/.  No reason to not document it now IMHO.


    Werner

Re: entering Unicode characters as hex numbers not documented

[Kevin Rodgers]

Richard Stallman wrote:
> I think the reason for not documenting `ucs-insert' in the Lisp Manual
> is that it serves a specialized purpose that will disappear in the
> subsequent release.

Subsequent to Emacs 22 (code name Godot)?  That would be Emacs 23 (Son
of Godot).

-- 
Kevin :-)

RE: entering Unicode characters as hex numbers not documented

[Drew Adams]

    Subsequent to Emacs 22 (code name Godot)?  That would be Emacs 23 (Son
    of Godot).
    
That would be Pozzo, whom some have mistaken for Godot.

Re: entering Unicode characters as hex numbers not documented

[Kenichi Handa]

In article <DNEMKBNJBGPAOPIJOOICGEOKDGAA.drew.adams@oracle.com>, "Drew Adams" <drew.adams@oracle.com> writes:

>     Subsequent to Emacs 22 (code name Godot)?  That would be Emacs 23 (Son
>     of Godot).
    
> That would be Pozzo, whom some have mistaken for Godot.

FYI, Emacs' mule feature has it's own version number and
name (C-h v mule-version).  Those of the CVS HEAD are "5.0
(SAKAKI)".  Those of emacs-unicode are "6.0
(HANACHIRUSATO)".  They are named after the chapters of
Japanese classic literature "Genji Monogatari (The Tale of
Genji)".

http://en.wikipedia.org/wiki/The_Tale_of_Genji
http://en.wikipedia.org/wiki/The_Tale_of_Genji#List_of_chapters

---
Kenichi Handa
handa@m17n.org

Re: entering Unicode characters as hex numbers not documented

[Chong Yidong]

Kevin Rodgers <ihs_4664@yahoo.com> writes:

> Richard Stallman wrote:
>> I think the reason for not documenting `ucs-insert' in the Lisp Manual
>> is that it serves a specialized purpose that will disappear in the
>> subsequent release.
>
> Subsequent to Emacs 22 (code name Godot)?

I prefer "Zeno".

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

> Btw, why did you ask for an index entry `utf-8'?  This function has
> nothing to do with UTF-8.

David has answered this.  It's far easier IMHO to provide a `wrong'
index entry than making Joe User learn the distinction between a
character set and an encoding.


    Werner

Re: entering Unicode characters as hex numbers not documented

[Eli Zaretskii]

> Date: Thu, 01 Jun 2006 16:48:19 +0200 (CEST)
> From: Werner LEMBERG <wl@gnu.org>
> Cc: emacs-devel@gnu.org
> 
> > Btw, why did you ask for an index entry `utf-8'?  This function has
> > nothing to do with UTF-8.
> 
> David has answered this.  It's far easier IMHO to provide a `wrong'
> index entry than making Joe User learn the distinction between a
> character set and an encoding.

We shouldn't cater only to the ignorant.  If I type "i utf-8 RET" and
land on the page that describes ucs-insert, I don't want to waste time
looking for information that just isn't there.

Re: entering Unicode characters as hex numbers not documented

[David Kastrup]

Eli Zaretskii <eliz@gnu.org> writes:

>> Date: Thu, 01 Jun 2006 16:48:19 +0200 (CEST)
>> From: Werner LEMBERG <wl@gnu.org>
>> Cc: emacs-devel@gnu.org
>> 
>> > Btw, why did you ask for an index entry `utf-8'?  This function has
>> > nothing to do with UTF-8.
>> 
>> David has answered this.  It's far easier IMHO to provide a `wrong'
>> index entry than making Joe User learn the distinction between a
>> character set and an encoding.
>
> We shouldn't cater only to the ignorant.

It is the ignorant who need to consult a manual in the first place.

> If I type "i utf-8 RET" and land on the page that describes
> ucs-insert, I don't want to waste time looking for information that
> just isn't there.

The time to press `,' is not exactly taxing for a cognoscente.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: entering Unicode characters as hex numbers not documented

[Eli Zaretskii]

> Cc: Werner LEMBERG <wl@gnu.org>,  emacs-devel@gnu.org
> From: David Kastrup <dak@gnu.org>
> Date: Thu, 01 Jun 2006 22:43:20 +0200
> 
> > We shouldn't cater only to the ignorant.
> 
> It is the ignorant who need to consult a manual in the first place.

No, everybody needs to consult a manual.  The only exceptions are
those super-humans who can remember a 600-page manual by heart.

> > If I type "i utf-8 RET" and land on the page that describes
> > ucs-insert, I don't want to waste time looking for information that
> > just isn't there.
> 
> The time to press `,' is not exactly taxing for a cognoscente.

I was talking about the time it takes to read and re-read the node,
searching in vain where it talks about UTF.  I have a bad habit of
trusting the index, so if it says there's something about UTF, I'll
try hard to find it.

Re: entering Unicode characters as hex numbers not documented

[Juri Linkov]

> Please add documentation and index entries (for `unicode', `ucs',
> `utf-8', etc.) for the function `ucs-insert'!  IMHO this is a very
> essential function, and after the release it will become quite
> important, I think.

One alternative to documenting this command is installing the patch that
allows C-q (`quoted-insert') to read Unicode characters as hex numbers and
insert them using `ucs-insert' function.

See http://lists.gnu.org/archive/html/bug-gnu-emacs/2005-06/msg00219.html.

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: entering Unicode characters as hex numbers not documented

[Kenichi Handa]

Werner LEMBERG <wl@gnu.org> writes:

> Additionally, there are a lot of Unicode characters which aren't
> accessible with quial input methods.  How do you enter such characters
> then if not with `ucs-insert'?

At least, you can use the input method "ucs" (leim/quail/uni-input.el).

Input method: ucs (mode line indicator:U)

Input as Unicode: U<hex> or u<hex>, where <hex> is a four-digit hex number.

---
Kenichi Handa
handa@m17n.org

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

> At least, you can use the input method "ucs" (leim/quail/uni-input.el).
>
> Input method: ucs (mode line indicator:U)

Ah, ok.  Perhaps *this* can be documented...


    Werner

Re: entering Unicode characters as hex numbers not documented

[Juri Linkov]

> At least, you can use the input method "ucs" (leim/quail/uni-input.el).
>
> Input method: ucs (mode line indicator:U)
>
> Input as Unicode: U<hex> or u<hex>, where <hex> is a four-digit hex number.

On a par with the recent proposal about \uXXXX and \UXXXXXXXX notation,
shouldn't this input method differentiate between upper- and lower-case
variants, and accept U<hex> with a 8-digit hex number?

-- 
Juri Linkov
http://www.jurta.org/emacs/

Re: entering Unicode characters as hex numbers not documented

[Werner LEMBERG]

> > Input method: ucs (mode line indicator:U)
> >
> > Input as Unicode: U<hex> or u<hex>, where <hex> is a four-digit hex number.
> 
> On a par with the recent proposal about \uXXXX and \UXXXXXXXX
> notation, shouldn't this input method differentiate between upper-
> and lower-case variants, and accept U<hex> with a 8-digit hex
> number?

Hmm, 6 hex digits are sufficient to cover Unicode.  I think that
within that input encoding, *any* number of 1-6 hex digits should be
valid.


    Werner

Re: entering Unicode characters as hex numbers not documented

[Juri Linkov]

>> > Input method: ucs (mode line indicator:U)
>> >
>> > Input as Unicode: U<hex> or u<hex>, where <hex> is a four-digit hex number.
>>
>> On a par with the recent proposal about \uXXXX and \UXXXXXXXX
>> notation, shouldn't this input method differentiate between upper-
>> and lower-case variants, and accept U<hex> with a 8-digit hex
>> number?
>
> Hmm, 6 hex digits are sufficient to cover Unicode.  I think that
> within that input encoding, *any* number of 1-6 hex digits should be
> valid.

If you mean that any non-digit character should terminate the input
like in most other multi-character input methods, then I agree.

-- 
Juri Linkov
http://www.jurta.org/emacs/