bug#32599: 25.2; Feature request: input PUA characters by name

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

The second attempt to post.

Regards

JSB

On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
> Please extend 'insert-char' to allow using the PUA character names and
> codes provided in the format of Unicode named sequences
>
> http://www.unicode.org/reports/tr34/tr34-23.html
>
> but containing only single characters.
>
> Such data are already available for the Medieval Unicode Font Initiative
> specification at
>
> https://bitbucket.org/jsbien/unihistext/src/master/example/
>
> A more radical extension would allow also to input real named sequences,
> but from my point of view this is of much lower importance.
>
> Best regards
>
> Janusz
>
>
> In GNU Emacs 25.2.2 (x86_64-pc-linux-gnu, GTK+ Version 3.22.30)
>  of 2018-07-11, modified by Debian built on x86-ubc-01
> Windowing system distributor 'The X.Org Foundation', version 11.0.12000000
> System Description:	Debian GNU/Linux testing (buster)
>
> Configured using:
>  'configure --build x86_64-linux-gnu --prefix=/usr
>  --sharedstatedir=/var/lib --libexecdir=/usr/lib
>  --localstatedir=/var/lib --infodir=/usr/share/info
>  --mandir=/usr/share/man --with-pop=yes
>  --enable-locallisppath=/etc/emacs25:/etc/emacs:/usr/local/share/emacs/25.2/site-lisp:/usr/local/share/emacs/site-lisp:/usr/share/emacs/25.2/site-lisp:/usr/share/emacs/site-lisp
>  --with-sound=alsa --without-gconf --build x86_64-linux-gnu
>  --prefix=/usr --sharedstatedir=/var/lib --libexecdir=/usr/lib
>  --localstatedir=/var/lib --infodir=/usr/share/info
>  --mandir=/usr/share/man --with-pop=yes
>  --enable-locallisppath=/etc/emacs25:/etc/emacs:/usr/local/share/emacs/25.2/site-lisp:/usr/local/share/emacs/site-lisp:/usr/share/emacs/25.2/site-lisp:/usr/share/emacs/site-lisp
>  --with-sound=alsa --without-gconf --with-x=yes --with-x-toolkit=gtk3
>  --with-toolkit-scroll-bars 'CFLAGS=-g -O2
>  -fdebug-prefix-map=/build/emacs25-cfFROJ/emacs25-25.2+1=. -fstack-protector-strong
>  -Wformat -Werror=format-security -Wall' 'CPPFLAGS=-Wdate-time
>  -D_FORTIFY_SOURCE=2' LDFLAGS=-Wl,-z,relro'
>
> Configured features:
> XPM JPEG TIFF GIF PNG RSVG IMAGEMAGICK SOUND GPM DBUS GSETTINGS NOTIFY
> ACL LIBSELINUX GNUTLS LIBXML2 FREETYPE M17N_FLT LIBOTF XFT ZLIB
> TOOLKIT_SCROLL_BARS GTK3 X11
>
> Important settings:
>   value of $LANG: en_US.UTF-8
>   locale-coding-system: utf-8-unix
>
> Major mode: Group
>
> Minor modes in effect:
>   cursor-sensor-mode: t
>   gnus-undo-mode: t
>   tooltip-mode: t
>   global-eldoc-mode: t
>   electric-indent-mode: t
>   mouse-wheel-mode: t
>   tool-bar-mode: t
>   menu-bar-mode: t
>   file-name-shadow-mode: t
>   global-font-lock-mode: t
>   font-lock-mode: t
>   blink-cursor-mode: t
>   auto-composition-mode: t
>   auto-encryption-mode: t
>   auto-compression-mode: t
>   buffer-read-only: t
>   line-number-mode: t
>   transient-mark-mode: t
>
> Recent messages:
> Sending...
> Mark set [2 times]
> Sending via mail...
> Sending email 
> Sending email done
> Mark set
> Saving file /home/jsbien/Mail/archive/sent/2018-08...
> Wrote /home/jsbien/Mail/archive/sent/2018-08
> Sending...done
> Auto-saving...done
>
> Load-path shadows:
> /usr/share/emacs/25.2/site-lisp/debian-startup hides /usr/share/emacs/site-lisp/debian-startup
> /usr/share/emacs25/site-lisp/cmake-data/cmake-mode hides /usr/share/emacs/site-lisp/cmake-mode
> /usr/share/emacs/site-lisp/rst hides /usr/share/emacs/25.2/lisp/textmodes/rst
> /usr/share/emacs25/site-lisp/latex-cjk-thai/thai-word hides /usr/share/emacs/25.2/lisp/language/thai-word
>
> Features:
> (shadow warnings emacsbug smtpmail mailalias bbdb-pgp bbdb-message
> sendmail nnir sort gnus-cite smiley ansi-color shr-color color url-util
> url-parse url-vars shr dom subr-x browse-url mm-archive mail-extr
> gnus-async gnus-bcklg qp gnus-ml disp-table cursor-sensor nndraft nnmh
> nndoc nnfolder utf-7 bbdb-gnus bbdb-mua bbdb-com crm network-stream nsm
> auth-source cl-seq eieio eieio-core cl-macs starttls gnus-agent
> gnus-srvr gnus-score score-mode nnvirtual gnus-msg gnus-art mm-uu
> mml2015 mm-view mml-smime smime dig mailcap nntp gnus-cache gnus-sum
> gnus-group gnus-undo gnus-start gnus-cloud nnimap nnmail mail-source tls
> gnutls utf7 netrc nnoo parse-time gnus-spec gnus-int gnus-range message
> dired-x dired format-spec rfc822 mml mml-sec password-cache epg
> mm-decode mm-bodies mm-encode mail-parse rfc2231 rfc2047 rfc2045
> ietf-drums mailabbrev gmm-utils mailheader gnus-win gnus gnus-ems
> nnheader gnus-util mail-utils mm-util help-fns mail-prsvr windmove quail
> bbdb bbdb-site timezone server edmacro kmacro mairix cus-edit cus-start
> cus-load wid-edit finder-inf tex-site info debian-el package epg-config
> seq byte-opt gv bytecomp byte-compile cl-extra help-mode easymenu cconv
> cl-loaddefs pcase cl-lib bbdb-loaddefs time-date mule-util tooltip eldoc
> electric uniquify ediff-hook vc-hooks lisp-float-type mwheel x-win
> term/common-win x-dnd tool-bar dnd fontset image regexp-opt fringe
> tabulated-list newcomment elisp-mode lisp-mode prog-mode register page
> menu-bar rfn-eshadow timer select scroll-bar mouse jit-lock font-lock
> syntax facemenu font-core frame cl-generic cham georgian utf-8-lang
> misc-lang vietnamese tibetan thai tai-viet lao korean japanese eucjp-ms
> cp51932 hebrew greek romanian slovak czech european ethiopic indian
> cyrillic chinese charscript case-table epa-hook jka-cmpr-hook help
> simple abbrev minibuffer cl-preloaded nadvice loaddefs button faces
> cus-face macroexp files text-properties overlay sha1 md5 base64 format
> env code-pages mule custom widget hashtable-print-readable backquote
> dbusbind inotify dynamic-setting system-font-setting font-render-setting
> move-toolbar gtk x-toolkit x multi-tty make-network-process emacs)
>
> Memory information:
> ((conses 16 260949 39014)
>  (symbols 48 33546 0)
>  (miscs 40 251 358)
>  (strings 32 65611 13129)
>  (string-bytes 1 2142874)
>  (vectors 16 34262)
>  (vector-slots 8 1334294 209768)
>  (floats 8 577 397)
>  (intervals 56 725 127)
>  (buffers 976 34))

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Robert Pluim]

jsbien@mimuw.edu.pl (Janusz S. Bień) writes:

> On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
>> Please extend 'insert-char' to allow using the PUA character names and
>> codes provided in the format of Unicode named sequences
>>
>> http://www.unicode.org/reports/tr34/tr34-23.html
>>
>> but containing only single characters.

Iʼm not sure I understand the request: insert-char already supports
using character names or code points to specify the character. What's
missing?

Thanks

Robert

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Fri, Aug 31 2018 at 10:05 +0200, rpluim@gmail.com writes:
> jsbien@mimuw.edu.pl (Janusz S. Bień) writes:
>
>> On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
>>> Please extend 'insert-char' to allow using the PUA character names and
>>> codes provided in the format of Unicode named sequences
>>>
>>> http://www.unicode.org/reports/tr34/tr34-23.html
>>>
>>> but containing only single characters.
>
> Iʼm not sure I understand the request: insert-char already supports
> using character names or code points to specify the character. What's
> missing?

You are missing "PUA" in the topic and in the text of my feature
request. PUA is Private Use Area. Cf. e.g.

https://folk.uib.no/hnooh/mufi/

and try to insert a character using such name as 'COMBINING ABBREVIATION
MARK SUPERSCRIPT UR ROUND R FORM'. Cf. also

https://www.unicode.org/mail-arch/unicode-ml/y2018-m08/0096.html

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Robert Pluim]

jsbien@mimuw.edu.pl (Janusz S. Bień) writes:

> On Fri, Aug 31 2018 at 10:05 +0200, rpluim@gmail.com writes:
>> jsbien@mimuw.edu.pl (Janusz S. Bień) writes:
>>
>>> On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
>>>> Please extend 'insert-char' to allow using the PUA character names and
>>>> codes provided in the format of Unicode named sequences
>>>>
>>>> http://www.unicode.org/reports/tr34/tr34-23.html
>>>>
>>>> but containing only single characters.
>>
>> Iʼm not sure I understand the request: insert-char already supports
>> using character names or code points to specify the character. What's
>> missing?
>
> You are missing "PUA" in the topic and in the text of my feature
> request. PUA is Private Use Area. Cf. e.g.
>

I missed that indeed. So what you want is a way of dynamically
extending which named entities can be used with insert-char based on
the type of information here:

https://folk.uib.no/hnooh/mufi/

Robert

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Fri, Aug 31 2018 at 14:34 +0200, rpluim@gmail.com writes:
> jsbien@mimuw.edu.pl (Janusz S. Bień) writes:
>
>> On Fri, Aug 31 2018 at 10:05 +0200, rpluim@gmail.com writes:
>>> jsbien@mimuw.edu.pl (Janusz S. Bień) writes:
>>>
>>>> On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
>>>>> Please extend 'insert-char' to allow using the PUA character names and
>>>>> codes provided in the format of Unicode named sequences
>>>>>
>>>>> http://www.unicode.org/reports/tr34/tr34-23.html
>>>>>
>>>>> but containing only single characters.
>>>
>>> Iʼm not sure I understand the request: insert-char already supports
>>> using character names or code points to specify the character. What's
>>> missing?
>>
>> You are missing "PUA" in the topic and in the text of my feature
>> request. PUA is Private Use Area. Cf. e.g.
>>
>
> I missed that indeed. So what you want is a way of dynamically
> extending which named entities can be used with insert-char based on
> the type of information here:
>
> https://folk.uib.no/hnooh/mufi/

Yes, although it is a slight oversimplification.

First, the MUFI data in a more convenient form are available here:

On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:

[...]

> https://bitbucket.org/jsbien/unihistext/src/master/example/

Secondly, other users may be interested in other sets of PUA characters,
cf.

http://andron-typeforum.xobor.de/t10f13-Towards-a-linguistic-corporate-use-area-LINCUA.html
https://en.wikipedia.org/wiki/ConScript_Unicode_Registry

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Fri, Aug 31 2018 at 14:54 +02, Janusz S. Bień wrote:
> On Fri, Aug 31 2018 at 14:34 +0200, rpluim@gmail.com writes:
>> jsbien@mimuw.edu.pl (Janusz S. Bień) writes:
>>
>>> On Fri, Aug 31 2018 at 10:05 +0200, rpluim@gmail.com writes:
>>>> jsbien@mimuw.edu.pl (Janusz S. Bień) writes:
>>>>
>>>>> On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
>>>>>> Please extend 'insert-char' to allow using the PUA character names and
>>>>>> codes provided in the format of Unicode named sequences
>>>>>>
>>>>>> http://www.unicode.org/reports/tr34/tr34-23.html

[...]

> First, the MUFI data in a more convenient form are available here:
>
> On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
>
> [...]
>
>> https://bitbucket.org/jsbien/unihistext/src/master/example/

If you prefer a file pattern after UnicodeData.txt, you can find it
here:

http://www.kreativekorp.com/charset/PUADATA/PUBLIC/MUFI/

>
> Secondly, other users may be interested in other sets of PUA characters,
> cf.
>
> http://andron-typeforum.xobor.de/t10f13-Towards-a-linguistic-corporate-use-area-LINCUA.html
> https://en.wikipedia.org/wiki/ConScript_Unicode_Registry

or Under-ConScript Unicode Registry:

http://www.kreativekorp.com/ucsur/

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
> Date: Sun, 26 May 2019 10:10:02 +0200
> 
> > First, the MUFI data in a more convenient form are available here:
> >
> > On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
> >
> > [...]
> >
> >> https://bitbucket.org/jsbien/unihistext/src/master/example/
> 
> If you prefer a file pattern after UnicodeData.txt, you can find it
> here:
> 
> http://www.kreativekorp.com/charset/PUADATA/PUBLIC/MUFI/
> 
> >
> > Secondly, other users may be interested in other sets of PUA characters,
> > cf.
> >
> > http://andron-typeforum.xobor.de/t10f13-Towards-a-linguistic-corporate-use-area-LINCUA.html
> > https://en.wikipedia.org/wiki/ConScript_Unicode_Registry
> 
> or Under-ConScript Unicode Registry:
> 
> http://www.kreativekorp.com/ucsur/

The UnicodeData.txt file is compiled into Emacs, but the files you
mention cannot be compiled into it, because they vary, and because
different users might want different lists of characters to be
supported.  So we need to design how this will work.

In addition, I think PUA codepoints aren't really treated as
characters in Emacs, so there's a need for some infrastructure
changes.

Patches welcome.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Sun, May 26 2019 at 17:45 +03, Eli Zaretskii wrote:
>> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
>> Date: Sun, 26 May 2019 10:10:02 +0200
>> 
>> > First, the MUFI data in a more convenient form are available here:
>> >
>> > On Mon, Aug 27 2018 at  9:00 +0200, jsbien@mimuw.edu.pl writes:
>> >
>> > [...]
>> >
>> >> https://bitbucket.org/jsbien/unihistext/src/master/example/
>> 
>> If you prefer a file pattern after UnicodeData.txt, you can find it
>> here:
>> 
>> http://www.kreativekorp.com/charset/PUADATA/PUBLIC/MUFI/
>> 
>> >
>> > Secondly, other users may be interested in other sets of PUA characters,
>> > cf.
>> >
>> > http://andron-typeforum.xobor.de/t10f13-Towards-a-linguistic-corporate-use-area-LINCUA.html
>> > https://en.wikipedia.org/wiki/ConScript_Unicode_Registry
>> 
>> or Under-ConScript Unicode Registry:
>> 
>> http://www.kreativekorp.com/ucsur/
>
> The UnicodeData.txt file is compiled into Emacs,

I know and I'm curious whether it is really needed. Why it cannot be
loaded at the startup? The advantage would be the user can use always
the up-to-date version of UnicodeData.txt (have you noticed that since
7th May we have now Unicode 12.1 because SQUARE ERA NAME REIWA was
added?).

> but the files you mention cannot be compiled into it, because they
> vary, and because different users might want different lists of
> characters to be supported.  So we need to design how this will work.

My naive idea is to "cheat" Emacs by providing it with the extended data
without changing the original logic. Efficiency is less important than
convenience, perhaps you can "advice"  the 'describe-char' function to
look for the data elsewhere.

> In addition, I think PUA codepoints aren't really treated as
> characters in Emacs, so there's a need for some infrastructure
> changes.

I do not propose to support the supplemental PUA planes. For the BMP
this probably boils down to the availability of the property
information. As we have now a pseudo-UnicodeData.txt for the PUA
characters (at least thise I'm interested in) this doesn't seem to me a
big problem).

> Patches welcome.

Unfortunately I'm unable to provide them myself.

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Sun, May 26 2019 at 17:18 +02, Janusz S. Bień wrote:

[...]

>> The UnicodeData.txt file is compiled into Emacs,
>
> I know and I'm curious whether it is really needed.

It was a bad wording as I can guess it is done for efficiency.

How the character data are stored and accessed? I have no qualification
to find the answer myself in the code,

> Why it cannot be loaded at the startup?

Or, better, partially overriden or supplemented? Is is prohibited by the
way the character data are stored?

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
> Cc: 32599@debbugs.gnu.org
> Date: Sun, 26 May 2019 17:48:43 +0200
> 
> >> The UnicodeData.txt file is compiled into Emacs,
> >
> > I know and I'm curious whether it is really needed.
> 
> It was a bad wording as I can guess it is done for efficiency.

Indeed.  Both processing efficiency and (perhaps more importantly)
memory efficiency.  Some of the resulting data is accessed by core
features that must be very efficient: the display engine, the regexp
search functions, etc.

> How the character data are stored and accessed?

The data is stored in char-tables of special kind.  See the files
uni-*.el which are produced as part of the Emacs build process.

> > Why it cannot be loaded at the startup?
> 
> Or, better, partially overriden or supplemented? Is is prohibited by the
> way the character data are stored?

You can override the char-tables at run time, of course, but first you
need to generate the new ones.

And there's the other problem I mentioned: with general treatment of
PUA codepoints.

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
> Cc: 32599@debbugs.gnu.org
> Date: Sun, 26 May 2019 17:18:21 +0200
> 
> > In addition, I think PUA codepoints aren't really treated as
> > characters in Emacs, so there's a need for some infrastructure
> > changes.
> 
> I do not propose to support the supplemental PUA planes. For the BMP
> this probably boils down to the availability of the property
> information. As we have now a pseudo-UnicodeData.txt for the PUA
> characters (at least thise I'm interested in) this doesn't seem to me a
> big problem).

I think we are miscommunicating.  The problems I alluded to start from
the fact that we exclude the PUA codepoints from the character
property database.  This means you cannot change their case and access
their syntax category, for example.  Functions that select a suitable
font also ignore PUA codepoints, IIRC.  Etc. etc. -- Someone™ should
go over all the places where we specify character properties and use
them, and make sure PUA codepoints aren't disregarded.  Otherwise,
just making Emacs know the names of these codepoints will be of very
limited value.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Sun, May 26 2019 at 19:56 +03, Eli Zaretskii wrote:
>> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
>> Cc: 32599@debbugs.gnu.org
>> Date: Sun, 26 May 2019 17:18:21 +0200
>> 
>> > In addition, I think PUA codepoints aren't really treated as
>> > characters in Emacs, so there's a need for some infrastructure
>> > changes.
>> 
>> I do not propose to support the supplemental PUA planes. For the BMP
>> this probably boils down to the availability of the property
>> information. As we have now a pseudo-UnicodeData.txt for the PUA
>> characters (at least thise I'm interested in) this doesn't seem to me a
>> big problem).
>
> I think we are miscommunicating.

I'm afraid it's my fault, looks like I write too briefly.

> The problems I alluded to start from the fact that we exclude the PUA
> codepoints from the character property database.

I understand this but I expected, perhaps incorrectly, that the code for
it is rather simple.

> This means you cannot change their case and access
> their syntax category, for example.  Functions that select a suitable
> font also ignore PUA codepoints, IIRC.

So the problem is that the relevant code occurs in several places.

> Etc. etc. -- Someone™ should
> go over all the places where we specify character properties and use
> them, and make sure PUA codepoints aren't disregarded.

I understand one has to look for the occurences of a function or
constant. Definitely there are tools to make this easy/easier (many
years ago I was experimenting with tags file), now I would expect them
to be quite sophisticated.

How would you approach this problem?

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
> Cc: 32599@debbugs.gnu.org
> Date: Sun, 26 May 2019 19:33:20 +0200
> 
> > The problems I alluded to start from the fact that we exclude the PUA
> > codepoints from the character property database.
> 
> I understand this but I expected, perhaps incorrectly, that the code for
> it is rather simple.

Simple: yes.  But it is also tedious, as there are many characters.
See characters.el.

With PUA codepoints there's one more complication: their attributes
are not standardized, so there should be a way of defining that
dynamically, not as characters.el does for all the other characters
whose attributes are static.

> > This means you cannot change their case and access
> > their syntax category, for example.  Functions that select a suitable
> > font also ignore PUA codepoints, IIRC.
> 
> So the problem is that the relevant code occurs in several places.

No.  Case-fiddling depends on data set up by characters.el, so once
that data is set, all the rest should "just work".  And similarly for
other attributes.

> > Etc. etc. -- Someone™ should
> > go over all the places where we specify character properties and use
> > them, and make sure PUA codepoints aren't disregarded.
> 
> I understand one has to look for the occurences of a function or
> constant. Definitely there are tools to make this easy/easier (many
> years ago I was experimenting with tags file), now I would expect them
> to be quite sophisticated.

I'm not sure such tools will help here.  I think a thorough audit of
the related data and code is needed.

> How would you approach this problem?

I have no idea, sorry.  I mentioned some of the issues in the hope
that will help interested individuals to find the relevant places
easier.  How to modify the code and data in order to allow use of PUA
codepoints, and on top of that to have the properties of each PUA
codepoint determined from external data which isn't known in advance,
is part of the design problem the person who'd like working on this
will have to solve.  Personally, I'm surprised people use PUA for
these purposes, and even more surprised they expect Emacs to support
this.  But that's me.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Sun, May 26 2019 at 21:52 +03, Eli Zaretskii wrote:

[...]

> Personally, I'm surprised people use PUA for
> these purposes,

Which purposes? Cf.

https://en.wikipedia.org/wiki/Private_Use_Areas
http://www.kreativekorp.com/charset/PUADATA/
http://bit.ly/2XVTzRL-LINCUA

> and even more surprised they expect Emacs to support
> this.  But that's me.

PUA characters, especially MUFI, are needed to typeset some
texts. (XeLa)TeX is still a very good typesetting system and Emacs-based
AUCTeX is still a very good tool to use TeX.

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
> Cc: 32599@debbugs.gnu.org
> Date: Mon, 27 May 2019 07:48:38 +0200
> 
> On Sun, May 26 2019 at 21:52 +03, Eli Zaretskii wrote:
> 
> [...]
> 
> > Personally, I'm surprised people use PUA for
> > these purposes,
> 
> Which purposes?

The purposes of mainstream text editing.  Are there other comparable
applications that let users define fonts for PUA codepoints, define
their attributes, and then manipulate those characters as any other?

> > and even more surprised they expect Emacs to support
> > this.  But that's me.
> 
> PUA characters, especially MUFI, are needed to typeset some
> texts. (XeLa)TeX is still a very good typesetting system and Emacs-based
> AUCTeX is still a very good tool to use TeX.

Those are separate projects.  If they need to use non-standard
characters with corresponding non-standard fonts, they could maintain
some add-on packages for Emacs to do that.

Asking Emacs to maintain compatibility to various ad-hoc registries
outside of Unicode is not really reasonable.  How will we know which
PUA codepoints to support?  How would we know which registered
codepoints are stable enough and wide-spread enough to justify their
support in Emacs?  How will we keep track of the changes in these
areas?  This is a task for someone who is an expert in this domain, is
aware of all the important developments there, and who can update
Emacs whenever something important changes.  As things are, we have
trouble even tracking the Unicode Standard: each new edition takes
several hours to incorporate, run the tests, make the necessary
changes, etc.

I guess what I'm saying is that without a dedicated volunteer who
would take care of this issue we can only wish such support will be
added, but we have no real hope it will materialize, except by some
enormous luck.

Of course, if you know someone who could be persuaded to come on board
and work on this now and in the future, I think the feature will be
welcome.

Thanks.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Mon, May 27 2019 at 20:11 +03, Eli Zaretskii wrote:
>> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
>> Cc: 32599@debbugs.gnu.org
>> Date: Mon, 27 May 2019 07:48:38 +0200
>> 
>> On Sun, May 26 2019 at 21:52 +03, Eli Zaretskii wrote:
>> 
>> [...]
>> 
>> > Personally, I'm surprised people use PUA for
>> > these purposes,
>> 
>> Which purposes?
>
> The purposes of mainstream text editing.  Are there other comparable
> applications that let users define fonts for PUA codepoints, define
> their attributes, and then manipulate those characters as any other?

I don't know and don't care, as I use only Emacs for editing.

>> > and even more surprised they expect Emacs to support
>> > this.  But that's me.
>> 
>> PUA characters, especially MUFI, are needed to typeset some
>> texts. (XeLa)TeX is still a very good typesetting system and Emacs-based
>> AUCTeX is still a very good tool to use TeX.
>
> Those are separate projects.  If they need to use non-standard
> characters with corresponding non-standard fonts, they could maintain
> some add-on packages for Emacs to do that.

Thanks for the suggestion. I will contact AUCTeX people and see what thay
think about it.

>
> Asking Emacs to maintain compatibility to various ad-hoc registries
> outside of Unicode is not really reasonable.

Yes, but it is not what I propose. I just would the user to be able to
use his own definition of PUA provided in the form of an additional
UnicodeData.txt (not necessarily as a part of Unicode, it can be perhaps
a different coding system).

[...]

> I guess what I'm saying is that without a dedicated volunteer who
> would take care of this issue we can only wish such support will be
> added, but we have no real hope it will materialize, except by some
> enormous luck.

As I said, we have different things in mind, so I'm not sure the above
statement really applies to my proposal.

> Of course, if you know someone who could be persuaded to come on board
> and work on this now and in the future, I think the feature will be
> welcome.

I don't know such a person now, but this can of course change in the
future.

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
> Cc: 32599@debbugs.gnu.org
> Date: Mon, 27 May 2019 19:39:50 +0200
> 
> > Asking Emacs to maintain compatibility to various ad-hoc registries
> > outside of Unicode is not really reasonable.
> 
> Yes, but it is not what I propose. I just would the user to be able to
> use his own definition of PUA provided in the form of an additional
> UnicodeData.txt (not necessarily as a part of Unicode, it can be perhaps
> a different coding system).

We don't have infrastructure for reliably doing such customizations on
user level.  Even importing a new version of the Unicode Standard
currently requires some manual adaptations, although most of the job
is done by just rebuilding Emacs after downloading a few UCD files.

Apart from making it possible to modify/update the relevant tables
programmatically, someone should figure out which properties are
relevant to PUA codepoints, which scripts they should belong, and also
how to specify fonts for them.  And if you want users to be able to do
this stuff, all those missing infrastructure features should be
exposed as user commands which don't require rebuilding Emacs and
preferably not even restarting it.  Someone who knows enough about the
issue should do this job, for the feature to be available.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Mon, May 27 2019 at 21:45 +03, Eli Zaretskii wrote:
>> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
>> Cc: 32599@debbugs.gnu.org
>> Date: Mon, 27 May 2019 19:39:50 +0200
>> 
>> > Asking Emacs to maintain compatibility to various ad-hoc registries
>> > outside of Unicode is not really reasonable.
>> 
>> Yes, but it is not what I propose. I just would the user to be able to
>> use his own definition of PUA provided in the form of an additional
>> UnicodeData.txt (not necessarily as a part of Unicode, it can be perhaps
>> a different coding system).
>
> We don't have infrastructure for reliably doing such customizations on
> user level.  Even importing a new version of the Unicode Standard
> currently requires some manual adaptations, although most of the job
> is done by just rebuilding Emacs after downloading a few UCD files.

I'm intrigued by the need of manual adaptation.

However my main point today is different:

Why our posts are not available neither at

https://debbugs.gnu.org/cgi/bugreport.cgi?bug=32599

not at

https://groups.google.com/forum/#!msg/gnu.emacs.bug/SK-RO_OG_GQ/BmUzA-23AgAJ

?

This dosn't allow other users to express their opinion on the matter.

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

On May 28, 2019 8:18:13 AM GMT+03:00, jsbien@mimuw.edu.pl wrote:

> However my main point today is different:
> 
> Why our posts are not available neither at
> 
> https://debbugs.gnu.org/cgi/bugreport.cgi?bug=32599
> 
> not at
> 
> https://groups.google.com/forum/#!msg/gnu.emacs.bug/SK-RO_OG_GQ/BmUzA-23AgAJ
> 
> ?
> 
> This dosn't allow other users to express their opinion on the matter.

Our posts do appear on that page, not sure why you don't see them.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Tue, May 28 2019 at  8:34 +03, Eli Zaretskii wrote:
> On May 28, 2019 8:18:13 AM GMT+03:00, jsbien@mimuw.edu.pl wrote:
>
>> However my main point today is different:
>> 
>> Why our posts are not available neither at
>> 
>> https://debbugs.gnu.org/cgi/bugreport.cgi?bug=32599
>> 
>> not at
>> 
>> https://groups.google.com/forum/#!msg/gnu.emacs.bug/SK-RO_OG_GQ/BmUzA-23AgAJ
>> 
>> ?
>> 
>> This dosn't allow other users to express their opinion on the matter.
>
> Our posts do appear on that page, not sure why you don't see them.

You are right, I see it in Chrome, looks like a strange Firefox problem.

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

Let me to approach the problem from another angle.

Input of Unicode characters by name is done by "insert-char" defined in
editfns.c. The code is quite short:

--8<---------------cut here---------------start------------->8---
DEFUN ("insert-char", Finsert_char, Sinsert_char, 1, 3,
       "(list (read-char-by-name \"Insert character (Unicode name or hex): \")\
              (prefix-numeric-value current-prefix-arg)\
              t))",
       doc: /* Insert COUNT copies of CHARACTER.
Interactively, prompt for CHARACTER.  You can specify CHARACTER in one
of these ways:

 - As its Unicode character name, e.g. \"LATIN SMALL LETTER A\".
   Completion is available; if you type a substring of the name
   preceded by an asterisk `*', Emacs shows all names which include
   that substring, not necessarily at the beginning of the name.

[...]


The optional third argument INHERIT, if non-nil, says to inherit text
properties from adjoining text, if those properties are sticky.  If
called interactively, INHERIT is t.  */)
  (Lisp_Object character, Lisp_Object count, Lisp_Object inherit)
{
  int i, stringlen;
  register ptrdiff_t n;
  int c, len;
  unsigned char str[MAX_MULTIBYTE_LENGTH];
  char string[4000];

  CHECK_CHARACTER (character);
  if (NILP (count))
    XSETFASTINT (count, 1);
  else
    CHECK_FIXNUM (count);
  c = XFIXNAT (character);

  if (!NILP (BVAR (current_buffer, enable_multibyte_characters)))
    len = CHAR_STRING (c, str);
  else
    str[0] = c, len = 1;
  if (XFIXNUM (count) <= 0)
    return Qnil;
  if (BUF_BYTES_MAX / len < XFIXNUM (count))
    buffer_overflow ();
  n = XFIXNUM (count) * len;
  stringlen = min (n, sizeof string - sizeof string % len);
  for (i = 0; i < stringlen; i++)
    string[i] = str[i % len];
  while (n > stringlen)
    {
      maybe_quit ();
      if (!NILP (inherit))
	insert_and_inherit (string, stringlen);
      else
	insert (string, stringlen);
      n -= stringlen;
    }
  if (!NILP (inherit))
    insert_and_inherit (string, n);
  else
    insert (string, n);
  return Qnil;
}
--8<---------------cut here---------------end--------------->8---

Which part of the code is responsible for prompting, name input and
consulting the character names list?

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
> Cc: 32599@debbugs.gnu.org
> Date: Wed, 30 Dec 2020 18:49:26 +0100
> 
> Which part of the code is responsible for prompting, name input and
> consulting the character names list?

The interactive spec:

     "(list (read-char-by-name \"Insert character (Unicode name or hex): \")\
	    (prefix-numeric-value current-prefix-arg)\
	    t))",

IOW, the call to read-char-by-name does that.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Wed, Dec 30 2020 at 22:52 +02, Eli Zaretskii wrote:
>> From: jsbien@mimuw.edu.pl (Janusz S. Bień)
>> Cc: 32599@debbugs.gnu.org
>> Date: Wed, 30 Dec 2020 18:49:26 +0100
>> 
>> Which part of the code is responsible for prompting, name input and
>> consulting the character names list?
>
> The interactive spec:
>
>      "(list (read-char-by-name \"Insert character (Unicode name or hex): \")\
> 	    (prefix-numeric-value current-prefix-arg)\
> 	    t))",
>
> IOW, the call to read-char-by-name does that.

Thanks. The answer was quite obvious, I'm slightly ashamed I didn't
noticed it :-(

So the PUA names are to be added to ucs-names in mule-cmds.el. At first
glance it doesn't seem difficult, as I already have them in the form of 
uni-name.el. I will work on it after New Year.

Regards - Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

On December 31, 2020 8:39:03 AM GMT+02:00, jsbien@mimuw.edu.pl wrote:
> So the PUA names are to be added to ucs-names in mule-cmds.el. At
> first
> glance it doesn't seem difficult, as I already have them in the form
> of 
> uni-name.el. I will work on it after New Year.

What would be the advantage of adding names for PUA codepoints?  The disadvantage is clear: bloating the Emacs process memory footprint by some 137,000 strings of some non-descriptive form, like PUA-CHARACTER-nn.

Btw, the names are not generated in mule-cmds.el, they are generated in unidata-gen.el.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Thu, Dec 31 2020 at  9:49 +02, Eli Zaretskii wrote:
> On December 31, 2020 8:39:03 AM GMT+02:00, jsbien@mimuw.edu.pl wrote:
>> So the PUA names are to be added to ucs-names in mule-cmds.el. At
>> first
>> glance it doesn't seem difficult, as I already have them in the form
>> of 
>> uni-name.el. I will work on it after New Year.
>
> What would be the advantage of adding names for PUA codepoints?

To make it clear: I'm not adding names to PUA codepoints, I add to Emacs
the names already in use for some PUA codepoints.

> The disadvantage is clear: bloating the Emacs process memory footprint
> by some 137,000 strings of some non-descriptive form, like
> PUA-CHARACTER-nn.

What do you need PUA-CHARACTER-nn for? I never proposed anything like
this.

Moreover:

There is always price for everything.

BTW, not 137,000 but 739 for MUFI 4.0 and a little more for the current 
version.

>
> Btw, the names are not generated in mule-cmds.el, they are generated
> in unidata-gen.el.

I know that and, as I said in my previous mail, I've generated them already:

--8<---------------cut here---------------start------------->8---
             position: 4051 of 4693 (86%), column: 40
            character:  (displayed as ) (codepoint 59575, #o164267, #xe8b7)
    preferred charset: unicode (Unicode (ISO10646))
code point in charset: 0xE8B7
               syntax: w 	which means: word
             category: L:Left-to-right (strong)
             to input: type "C-x 8 RET e8b7"
          buffer code: #xEE #xA2 #xB7
            file code: #xEE #xA2 #xB7 (encoded by coding system utf-8-unix)
              display: by this font (glyph code)
    xft:-psbk-Junicode-normal-normal-normal-*-15-*-*-*-*-0-iso10646-1 (#x94C)

Character code properties: customize what to show
  name: LATIN SMALL LETTER LONG S WITH FLOURISH
  general-category: Co (Other, Private Use)
  decomposition: (59575) ('')
--8<---------------cut here---------------end--------------->8---

However read-char-by-name takes the names from ucs-names which I have to
update.


Regards - Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

On December 31, 2020 10:14:41 AM GMT+02:00, jsbien@mimuw.edu.pl wrote:
> On Thu, Dec 31 2020 at  9:49 +02, Eli Zaretskii wrote:
> > On December 31, 2020 8:39:03 AM GMT+02:00, jsbien@mimuw.edu.pl
> wrote:
> >> So the PUA names are to be added to ucs-names in mule-cmds.el. At
> >> first
> >> glance it doesn't seem difficult, as I already have them in the
> form
> >> of 
> >> uni-name.el. I will work on it after New Year.
> >
> > What would be the advantage of adding names for PUA codepoints?
> 
> To make it clear: I'm not adding names to PUA codepoints, I add to
> Emacs
> the names already in use for some PUA codepoints.
> 
> > The disadvantage is clear: bloating the Emacs process memory
> footprint
> > by some 137,000 strings of some non-descriptive form, like
> > PUA-CHARACTER-nn.
> 
> What do you need PUA-CHARACTER-nn for? I never proposed anything like
> this.
> 
> Moreover:
> 
> There is always price for everything.
> 
> BTW, not 137,000 but 739 for MUFI 4.0 and a little more for the
> current 
> version.

OK, it wasn't clear that you want to add MUFI codepoints.

However, if that is the intent, I think we should first discuss whether we want to add PUA codepoints defined by the different initiatives, and if so, which ones.  MUFI is not the only such collections, there are others.

We could instead wait until these codepoints are incorporated into the Unicode Standard (as already happened with some MUFI codepoints).  Or we might decide that we want to add some infrastructure which could be used by users to add support for ranges of PUA without adding that by default.

IOW, I think this kind of changes should be discussed first, and the place to discuss them is not here, it's on emacs-devel.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Thu, Dec 31 2020 at 11:06 +02, Eli Zaretskii wrote:

[...]

> OK, it wasn't clear that you want to add MUFI codepoints.
>
> However, if that is the intent, I think we should first discuss
> whether we want to add PUA codepoints defined by the different
> initiatives, and if so, which ones.  MUFI is not the only such
> collections, there are others.

Correct, e.g. the character set of JuniusX font which uses plane 15
(https://github.com/psb1558/Junicode-New/discussions/44)

> We could instead wait until these codepoints are incorporated into the
> Unicode Standard (as already happened with some MUFI codepoints).

Some of them will never be accepted by Unicode because they violate
their principle.

> Or we might decide that we want to add some infrastructure which could
> be used by users to add support for ranges of PUA without adding that
> by default.

That's what I have in mind from the very beginning.

> IOW, I think this kind of changes should be discussed first, and the
> place to discuss them is not here, it's on emacs-devel.

Before starting the discussion on emacs-devel I would like to have a
proof-of-concept and also a quick solution for my own use.

Regards - Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

On December 31, 2020 11:31:32 AM GMT+02:00, jsbien@mimuw.edu.pl wrote:
> On Thu, Dec 31 2020 at 11:06 +02, Eli Zaretskii wrote:
> 
> [...]
> 
> > OK, it wasn't clear that you want to add MUFI codepoints.
> >
> > However, if that is the intent, I think we should first discuss
> > whether we want to add PUA codepoints defined by the different
> > initiatives, and if so, which ones.  MUFI is not the only such
> > collections, there are others.
> 
> Correct, e.g. the character set of JuniusX font which uses plane 15
> (https://github.com/psb1558/Junicode-New/discussions/44)
> 
> > We could instead wait until these codepoints are incorporated into
> the
> > Unicode Standard (as already happened with some MUFI codepoints).
> 
> Some of them will never be accepted by Unicode because they violate
> their principle.
> 
> > Or we might decide that we want to add some infrastructure which
> could
> > be used by users to add support for ranges of PUA without adding
> that
> > by default.
> 
> That's what I have in mind from the very beginning.

Well, your message talked about adding names for these codepoints, not about adding infrastructure.  Apologies for reading that too literally.

bug#32599: 25.2; Feature request: input PUA characters by name

[Lars Ingebrigtsen]

jsbien@mimuw.edu.pl (Janusz S. Bień) writes:

> You are missing "PUA" in the topic and in the text of my feature
> request. PUA is Private Use Area. Cf. e.g.

(I'm going through old bug reports that unfortunately weren't resolved
at the time.)

If I understand correctly, you just want to be able to extend the range
of characters you get when saying `C-x 8 RET TAB'?  That can be done
today by just adding elements to the `ucs-names' hash table.

Is that not sufficient for your needs?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Tue, Apr 26 2022 at 15:09 +02, Lars Ingebrigtsen wrote:
> jsbien@mimuw.edu.pl (Janusz S. Bień) writes:
>
>> You are missing "PUA" in the topic and in the text of my feature
>> request. PUA is Private Use Area. Cf. e.g.
>
> (I'm going through old bug reports that unfortunately weren't resolved
> at the time.)
>
> If I understand correctly, you just want to be able to extend the range
> of characters you get when saying `C-x 8 RET TAB'?

Yes. I've solved it in a quick and dirty way, but it was very cumbersome
and usable only in a specific verion of Emacs:
https://github.com/jsbien/unicode4polish/tree/master/Emacs-MUFI


> That can be done today by just adding elements to the `ucs-names' hash
> table.

Which Emacs version do you mean by "today"?

What do you mean by "just adding"?

What about an example, e.g. U+F159 LATIN ABBREVIATION SIGN SMALL DE?

> Is that not sufficient for your needs?

Perhaps, but I'm not sure yet.

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Lars Ingebrigtsen]

Janusz S. Bień <jsbien@mimuw.edu.pl> writes:

>> That can be done today by just adding elements to the `ucs-names' hash
>> table.
>
> Which Emacs version do you mean by "today"?

Since about 2008.

> What do you mean by "just adding"?
>
> What about an example, e.g. U+F159 LATIN ABBREVIATION SIGN SMALL DE?

(unless ucs-names
  (ucs-names))
(setf (gethash "LATIN ABBREVIATION SIGN SMALL DE" ucs-names) #xF159)

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Tue, Apr 26 2022 at 15:37 +02, Lars Ingebrigtsen wrote:
> Janusz S. Bień <jsbien@mimuw.edu.pl> writes:
>
>>> That can be done today by just adding elements to the `ucs-names' hash
>>> table.
>>
>> Which Emacs version do you mean by "today"?
>
> Since about 2008.
>
>> What do you mean by "just adding"?
>>
>> What about an example, e.g. U+F159 LATIN ABBREVIATION SIGN SMALL DE?
>
> (unless ucs-names
>   (ucs-names))
> (setf (gethash "LATIN ABBREVIATION SIGN SMALL DE" ucs-names) #xF159)

--8<---------------cut here---------------start------------->8---
Debugger entered--Lisp error: (wrong-type-argument hash-table-p nil)
  puthash("LATIN ABBREVIATION SIGN SMALL DE" 61785 nil)
  (let* ((v ucs-names)) (puthash "LATIN ABBREVIATION SIGN SMALL DE" 61785 v))
  (progn (let* ((v ucs-names)) (puthash "LATIN ABBREVIATION SIGN SMALL DE" 61785 v)))
  eval((progn (let* ((v ucs-names)) (puthash "LATIN ABBREVIATION SIGN SMALL DE" 61785 v))) t)
  elisp--eval-last-sexp(t)
  eval-last-sexp(t)
  eval-print-last-sexp(nil)
  funcall-interactively(eval-print-last-sexp nil)
  call-interactively(eval-print-last-sexp nil nil)
  command-execute(eval-print-last-sexp)
--8<---------------cut here---------------end--------------->8---

GNU Emacs 28.1 (build 1, x86_64-pc-linux-gnu, GTK+ Version 3.24.24,
cairo version 1.16.0) compiled locally.

JSB

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Lars Ingebrigtsen]

Janusz S. Bień <jsbien@mimuw.edu.pl> writes:

>> (unless ucs-names
>>   (ucs-names))
>> (setf (gethash "LATIN ABBREVIATION SIGN SMALL DE" ucs-names) #xF159)
>
> Debugger entered--Lisp error: (wrong-type-argument hash-table-p nil)
>   puthash("LATIN ABBREVIATION SIGN SMALL DE" 61785 nil)

Did you eval the `unless' form first?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Tue, Apr 26 2022 at 15:45 +02, Lars Ingebrigtsen wrote:
> Janusz S. Bień <jsbien@mimuw.edu.pl> writes:
>
>>> (unless ucs-names
>>>   (ucs-names))
>>> (setf (gethash "LATIN ABBREVIATION SIGN SMALL DE" ucs-names) #xF159)
>>
>> Debugger entered--Lisp error: (wrong-type-argument hash-table-p nil)
>>   puthash("LATIN ABBREVIATION SIGN SMALL DE" 61785 nil)
>
> Did you eval the `unless' form first?

I thought I did but I used the wrong command :-(

Thank you very much for your solution.

So yes, my original problem is solved and you can close the bug report.

This is however not the end of the story. I would like describe-char to
produce the correct name of the PUA character instead of "the
character’s canonical name and other properties defined by the Unicode
Data Base;". I understand this should posted as a separate feature
request.

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Robert Pluim]

>>>>> On Tue, 26 Apr 2022 17:33:37 +0200, Janusz S. Bień <jsbien@mimuw.edu.pl> said:

    Janusz> On Tue, Apr 26 2022 at 15:45 +02, Lars Ingebrigtsen wrote:
    >> Janusz S. Bień <jsbien@mimuw.edu.pl> writes:
    >> 
    >>>> (unless ucs-names
    >>>> (ucs-names))
    >>>> (setf (gethash "LATIN ABBREVIATION SIGN SMALL DE" ucs-names) #xF159)
    >>> 
    >>> Debugger entered--Lisp error: (wrong-type-argument hash-table-p nil)
    >>> puthash("LATIN ABBREVIATION SIGN SMALL DE" 61785 nil)
    >> 
    >> Did you eval the `unless' form first?

    Janusz> I thought I did but I used the wrong command :-(

    Janusz> Thank you very much for your solution.

    Janusz> So yes, my original problem is solved and you can close the bug report.

    Janusz> This is however not the end of the story. I would like describe-char to
    Janusz> produce the correct name of the PUA character instead of "the
    Janusz> character’s canonical name and other properties defined by the Unicode
    Janusz> Data Base;". I understand this should posted as a separate feature
    Janusz> request.

(put-char-code-property #xf159 'name "LATIN ABBREVIATION SIGN SMALL DE")

(PUA characters donʼt have canonical names).

Robert
-- 

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Tue, Apr 26 2022 at 17:49 +02, Robert Pluim wrote:

[...]

>     Janusz> This is however not the end of the story. I would like describe-char to
>     Janusz> produce the correct name of the PUA character instead of "the
>     Janusz> character’s canonical name and other properties defined by the Unicode
>     Janusz> Data Base;". I understand this should posted as a separate feature
>     Janusz> request.
>
> (put-char-code-property #xf159 'name "LATIN ABBREVIATION SIGN SMALL DE")

Great, thank you very much!

> (PUA characters donʼt have canonical names).

I know, I'm interested only in Medieval Unicode Font Initiative names.

But this is not the end of the story :-)

What is needed to treat the character as a letter? To be precise, I
would forward-word to go to the end of ahb, now it stops after a.

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Eli Zaretskii]

> From: Janusz S. Bień <jsbien@mimuw.edu.pl>
> Date: Tue, 26 Apr 2022 18:09:53 +0200
> Cc: 32599@debbugs.gnu.org, Lars Ingebrigtsen <larsi@gnus.org>
> 
> What is needed to treat the character as a letter? To be precise, I
> would forward-word to go to the end of ahb, now it stops after a.

You need to manually assign that character its properties.  See
characters.el.  You may also need to manually assign some Unicode
properties using put-char-code-property.

bug#32599: 25.2; Feature request: input PUA characters by name

[Janusz S. Bień]

On Tue, Apr 26 2022 at 19:44 +03, Eli Zaretskii wrote:
>> From: Janusz S. Bień <jsbien@mimuw.edu.pl>
>> Date: Tue, 26 Apr 2022 18:09:53 +0200
>> Cc: 32599@debbugs.gnu.org, Lars Ingebrigtsen <larsi@gnus.org>
>> 
>> What is needed to treat the character as a letter? To be precise, I
>> would forward-word to go to the end of ahb, now it stops after a.
>
> You need to manually assign that character its properties.  See
> characters.el.  You may also need to manually assign some Unicode
> properties using put-char-code-property.

Thanks for the advice.

Best regards

Janusz

-- 
             ,   
Janusz S. Bien
emeryt (emeritus)
https://sites.google.com/view/jsbien

bug#32599: 25.2; Feature request: input PUA characters by name

[Lars Ingebrigtsen]

Janusz S. Bień <jsbien@mimuw.edu.pl> writes:

> So yes, my original problem is solved and you can close the bug report.

OK; done.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no