12:50 p.m.
The FSF are coming gently to the conclusion, again, that the guillemets are
not parentheses. I implemented that insight in the various non-Latin-1
charsets a while ago, but never got around to doing it in the editor.
lisp/ChangeLog addition:
2006-07-29 Aidan Kehoe <kehoea(a)parhasard.net>
* iso8859-1.el:
Move the symbol table modification to three lines at the end of
syntax.c.
src/ChangeLog addition:
2006-07-29 Aidan Kehoe <kehoea(a)parhasard.net>
* syntax.c (Fsyntax_table_p):
Make the docstring better reflect the code.
* syntax.c (define_standard_syntax):
Take a const UExtbyte * as the first argument, not a const char *
* syntax.c (complex_vars_of_syntax):
Use a macro instead of repeating code; don't redundantly set set
the syntax of the alphanumeric characters to ?w; define syntax for
Latin 1 characters here instead of in Lisp; guillemets are string
delimiters, not parentheses.
XEmacs Trunk source patch:
Diff command: cvs -q diff -u
Files affected: src/syntax.c lisp/iso8859-1.el
Index: lisp/iso8859-1.el
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/lisp/iso8859-1.el,v
retrieving revision 1.3
diff -u -u -r1.3 iso8859-1.el
--- lisp/iso8859-1.el 2001/04/12 18:21:28 1.3
+++ lisp/iso8859-1.el 2006/07/29 11:29:54
＠＠ -1,6 +1,6 ＠＠
-;;; iso8859-1.el --- Set case and syntax tables for Latin 1
+;;; iso8859-1.el --- Set syntax table for Latin 1
-;; Copyright (C) 1992, 1997 Free Software Foundation, Inc.
+;; Copyright (C) 1992, 1997, 2006 Free Software Foundation, Inc.
;; Author: Jamie Zawinski <jwz(a)jwz.org>
;; Created: 19-aug-92
＠＠ -24,124 +24,16 ＠＠
;; Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
;; 02111-1307, USA.
-;;; Synched up with: Not synched
+;;; Synched up with: Not in FSF.
;;; Commentary:
;; created by jwz, 19-aug-92.
-;; Sets the case and syntax tables for the ISO-8859/1 character set.
+;; Sets the case table for the ISO-8859/1 character set.
+;; Used to set the syntax table.
;;; Code:
-(let ((table (standard-syntax-table)))
- ;;
- ;; The symbol characters
- ;;
- (modify-syntax-entry ?\240 "_" table) ; nobreakspace
- (modify-syntax-entry ?\241 "." table) ; exclamdown
- (modify-syntax-entry ?\242 "_" table) ; cent
- (modify-syntax-entry ?\243 "_" table) ; sterling
- (modify-syntax-entry ?\244 "_" table) ; currency
- (modify-syntax-entry ?\245 "_" table) ; yen
- (modify-syntax-entry ?\246 "_" table) ; brokenbar
- (modify-syntax-entry ?\247 "_" table) ; section
- (modify-syntax-entry ?\250 "_" table) ; diaeresis
- (modify-syntax-entry ?\251 "_" table) ; copyright
- (modify-syntax-entry ?\252 "_" table) ; ordfeminine
- (modify-syntax-entry ?\253 "(\273" table) ; guillemotleft
- (modify-syntax-entry ?\254 "_" table) ; notsign
- (modify-syntax-entry ?\255 "_" table) ; hyphen
- (modify-syntax-entry ?\256 "_" table) ; registered
- (modify-syntax-entry ?\257 "_" table) ; macron
- (modify-syntax-entry ?\260 "_" table) ; degree
- (modify-syntax-entry ?\261 "_" table) ; plusminus
- (modify-syntax-entry ?\262 "_" table) ; twosuperior
- (modify-syntax-entry ?\263 "_" table) ; threesuperior
- (modify-syntax-entry ?\264 "_" table) ; acute
- (modify-syntax-entry ?\265 "_" table) ; mu
- (modify-syntax-entry ?\266 "_" table) ; paragraph
- (modify-syntax-entry ?\267 "_" table) ; periodcentered
- (modify-syntax-entry ?\270 "_" table) ; cedilla
- (modify-syntax-entry ?\271 "_" table) ; onesuperior
- (modify-syntax-entry ?\272 "_" table) ; masculine
- (modify-syntax-entry ?\273 ")\253" table) ; guillemotright
- (modify-syntax-entry ?\274 "_" table) ; onequarter
- (modify-syntax-entry ?\275 "_" table) ; onehalf
- (modify-syntax-entry ?\276 "_" table) ; threequarters
- (modify-syntax-entry ?\277 "_" table) ; questiondown
- ;;
- ;; the upper-case characters (plus "multiply" and "ssharp")
- ;;
- (modify-syntax-entry ?\300 "w" table) ; Agrave
- (modify-syntax-entry ?\301 "w" table) ; Aacute
- (modify-syntax-entry ?\302 "w" table) ; Acircumflex
- (modify-syntax-entry ?\303 "w" table) ; Atilde
- (modify-syntax-entry ?\304 "w" table) ; Adiaeresis
- (modify-syntax-entry ?\305 "w" table) ; Aring
- (modify-syntax-entry ?\306 "w" table) ; AE
- (modify-syntax-entry ?\307 "w" table) ; Ccedilla
- (modify-syntax-entry ?\310 "w" table) ; Egrave
- (modify-syntax-entry ?\311 "w" table) ; Eacute
- (modify-syntax-entry ?\312 "w" table) ; Ecircumflex
- (modify-syntax-entry ?\313 "w" table) ; Ediaeresis
- (modify-syntax-entry ?\314 "w" table) ; Igrave
- (modify-syntax-entry ?\315 "w" table) ; Iacute
- (modify-syntax-entry ?\316 "w" table) ; Icircumflex
- (modify-syntax-entry ?\317 "w" table) ; Idiaeresis
- (modify-syntax-entry ?\320 "w" table) ; ETH
- (modify-syntax-entry ?\321 "w" table) ; Ntilde
- (modify-syntax-entry ?\322 "w" table) ; Ograve
- (modify-syntax-entry ?\323 "w" table) ; Oacute
- (modify-syntax-entry ?\324 "w" table) ; Ocircumflex
- (modify-syntax-entry ?\325 "w" table) ; Otilde
- (modify-syntax-entry ?\326 "w" table) ; Odiaeresis
- (modify-syntax-entry ?\327 "_" table) ; multiply
- (modify-syntax-entry ?\330 "w" table) ; Ooblique
- (modify-syntax-entry ?\331 "w" table) ; Ugrave
- (modify-syntax-entry ?\332 "w" table) ; Uacute
- (modify-syntax-entry ?\333 "w" table) ; Ucircumflex
- (modify-syntax-entry ?\334 "w" table) ; Udiaeresis
- (modify-syntax-entry ?\335 "w" table) ; Yacute
- (modify-syntax-entry ?\336 "w" table) ; THORN
- (modify-syntax-entry ?\337 "w" table) ; ssharp
- ;;
- ;; the lower-case characters (plus "division" and "ydiaeresis")
- ;;
- (modify-syntax-entry ?\340 "w" table) ; agrave
- (modify-syntax-entry ?\341 "w" table) ; aacute
- (modify-syntax-entry ?\342 "w" table) ; acircumflex
- (modify-syntax-entry ?\343 "w" table) ; atilde
- (modify-syntax-entry ?\344 "w" table) ; adiaeresis
- (modify-syntax-entry ?\345 "w" table) ; aring
- (modify-syntax-entry ?\346 "w" table) ; ae
- (modify-syntax-entry ?\347 "w" table) ; ccedilla
- (modify-syntax-entry ?\350 "w" table) ; egrave
- (modify-syntax-entry ?\351 "w" table) ; eacute
- (modify-syntax-entry ?\352 "w" table) ; ecircumflex
- (modify-syntax-entry ?\353 "w" table) ; ediaeresis
- (modify-syntax-entry ?\354 "w" table) ; igrave
- (modify-syntax-entry ?\355 "w" table) ; iacute
- (modify-syntax-entry ?\356 "w" table) ; icircumflex
- (modify-syntax-entry ?\357 "w" table) ; idiaeresis
- (modify-syntax-entry ?\360 "w" table) ; eth
- (modify-syntax-entry ?\361 "w" table) ; ntilde
- (modify-syntax-entry ?\362 "w" table) ; ograve
- (modify-syntax-entry ?\363 "w" table) ; oacute
- (modify-syntax-entry ?\364 "w" table) ; ocircumflex
- (modify-syntax-entry ?\365 "w" table) ; otilde
- (modify-syntax-entry ?\366 "w" table) ; odiaeresis
- (modify-syntax-entry ?\367 "_" table) ; division
- (modify-syntax-entry ?\370 "w" table) ; ooblique
- (modify-syntax-entry ?\371 "w" table) ; ugrave
- (modify-syntax-entry ?\372 "w" table) ; uacute
- (modify-syntax-entry ?\373 "w" table) ; ucircumflex
- (modify-syntax-entry ?\374 "w" table) ; udiaeresis
- (modify-syntax-entry ?\375 "w" table) ; yacute
- (modify-syntax-entry ?\376 "w" table) ; thorn
- (modify-syntax-entry ?\377 "w" table) ; ydiaeresis
- )
-
-
(defconst iso8859/1-case-table nil
"The case table for ISO-8859/1 characters.")
Index: src/syntax.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/syntax.c,v
retrieving revision 1.28
diff -u -u -r1.28 syntax.c
--- src/syntax.c 2006/02/22 03:30:08 1.28
+++ src/syntax.c 2006/07/29 11:29:58
＠＠ -146,7 +146,6 ＠＠
DEFUN ("syntax-table-p", Fsyntax_table_p, 1, 1, 0, /*
Return t if OBJECT is a syntax table.
-Any vector of 256 elements will do.
*/
(object))
{
＠＠ -2451,7 +2450,7 ＠＠
}
static void
-define_standard_syntax (const char *p, enum syntaxcode syn)
+define_standard_syntax (const UExtbyte *p, enum syntaxcode syn)
{
for (; *p; p++)
Fput_char_table (make_char (*p), make_int (syn), Vstandard_syntax_table);
＠＠ -2461,10 +2460,19 ＠＠
complex_vars_of_syntax (void)
{
Ichar i;
- const char *p;
- /* Set this now, so first buffer creation can refer to it. */
- /* Make it nil before calling copy-syntax-table
- so that copy-syntax-table will know not to try to copy from garbage */
+ const UExtbyte *p; /* Latin-1, not internal format. */
+
+#define SET_RANGE_SYNTAX(start, end, syntax) \
+ do { \
+ for (i = start; i <= end; i++) \
+ Fput_char_table(make_char(i), make_int(syntax), \
+ Vstandard_syntax_table); \
+ } while (0)
+
+ /* Set this now, so first buffer creation can refer to it.
+
+ Make it nil before calling copy-syntax-table so that copy-syntax-table
+ will know not to try to copy from garbage */
Vstandard_syntax_table = Qnil;
Vstandard_syntax_table = Fcopy_syntax_table (Qnil);
staticpro (&Vstandard_syntax_table);
＠＠ -2476,19 +2484,18 ＠＠
Smax);
staticpro (&Vsyntax_designator_chars_string);
+ /* Default character syntax is word. */
set_char_table_default (Vstandard_syntax_table, make_int (Sword));
+
+ /* Control 0; treat as punctuation */
+ SET_RANGE_SYNTAX(0, 32, Spunct);
- for (i = 0; i <= 32; i++) /* Control 0 plus SPACE */
- Fput_char_table (make_char (i), make_int (Swhitespace),
- Vstandard_syntax_table);
- for (i = 127; i <= 159; i++) /* DEL plus Control 1 */
- Fput_char_table (make_char (i), make_int (Swhitespace),
- Vstandard_syntax_table);
-
- define_standard_syntax ("abcdefghijklmnopqrstuvwxyz"
- "ABCDEFGHIJKLMNOPQRSTUVWXYZ"
- "0123456789"
- "$%", Sword);
+ /* The whitespace--overwriting some of the above changes. */
+ define_standard_syntax(" \t\015\014", Swhitespace);
+
+ /* DEL plus Control 1 */
+ SET_RANGE_SYNTAX(127, 159, Spunct);
+
define_standard_syntax ("\"", Sstring);
define_standard_syntax ("\\", Sescape);
define_standard_syntax ("_-+*/&|<>=", Ssymbol);
＠＠ -2503,4 +2510,18 ＠＠
Fcons (make_int (Sclose), make_char (p[0])),
Vstandard_syntax_table);
}
+
+ /* Latin 1 "symbols." This contrasts with the FSF, where they're word
+ constituents. */
+ SET_RANGE_SYNTAX(0240, 0277, Ssymbol);
+
+ /* The guillemets. These are not parentheses, in contrast to what the old
+ code did. */
+ define_standard_syntax("\253\273", Sstring);
+
+ /* The inverted exclamation mark, and the multiplication and division
+ signs. */
+ define_standard_syntax("\241\327\367", Spunct);
+
+#undef SET_RANGE_SYNTAX
}
--
Santa Maradona, priez pour moi!
1:50 p.m.
Just a comment:
>>>> "Aidan" == Aidan Kehoe
<kehoea(a)parhasard.net> writes:
+ /* Control 0; treat as punctuation */
+ SET_RANGE_SYNTAX(0, 32, Spunct);
+ /* DEL plus Control 1 */
+ SET_RANGE_SYNTAX(127, 159, Spunct);
There really oughtta be a better way, don't you think? We really
ought to have an Scntrl class. (Not seriously suggesting doing
something about it now, that's obviously a can of worms.)
By the way, what's the logic for Spunct? Something like, these are
clearly word boundaries, but they can't be whitespace because we don't
want them "catenating up" with "real" whitespace?
Why is ASCII SPACE included in the Control 0 range? That's neither
useful nor correct; SPACE is a graphic character (even though it's not
printing, or maybe I have that distinction backwards, my ISO standards
are at the office).
--
School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
2 p.m.
Ar an triochadú lá de mí Iúil, scríobh Stephen J. Turnbull:
Just a comment:
>>>>> "Aidan" == Aidan Kehoe <kehoea(a)parhasard.net>
writes:
+ /* Control 0; treat as punctuation */
+ SET_RANGE_SYNTAX(0, 32, Spunct);
+ /* DEL plus Control 1 */
+ SET_RANGE_SYNTAX(127, 159, Spunct);
There really oughtta be a better way, don't you think? We really
ought to have an Scntrl class. (Not seriously suggesting doing
something about it now, that's obviously a can of worms.)
By the way, what's the logic for Spunct? Something like, these are
clearly word boundaries, but they can't be whitespace because we don't
want them "catenating up" with "real" whitespace?
More like, they’re not whitespace, and GNU have gone for punctuation. Weak,
I know.
Why is ASCII SPACE included in the Control 0 range? That's
neither
useful nor correct; SPACE is a graphic character (even though it's not
printing, or maybe I have that distinction backwards, my ISO standards
are at the office).
I copied the range from the existing code, to make the chance of something
going wrong smaller; it actually doesn’t matter, since we make it whitespace
a couple of lines down.
ECMA-006.pdf says:
“6.2 Control characters
The control characters of the C0 set shall be represented by bit
combinations as specified in annex A. [so, 0-31 inclusive.]
6.3 Character SPACE
The acronym of the character SPACE is SP and it is represented by bit
combination 2/0.
This character is a graphic character, it has a visual representation
consisting of the absence of a graphic symbol.
6.4 Graphic characters
The 94 bit combinations 2/1 to 7/14 are used for the representation of
graphic characters as specified in 6.4.1, 6.4.2 and 6.4.3.
All graphic characters shall be spacing characters, that is, they
cause the active position to advance by one character position.”
(That does raise the question of whether double-width characters are graphic
characters, which I don’t really want to deal with right now.)
--
Santa Maradona, priez pour moi!
2:16 p.m.
Ar an triochadú lá de mí Iúil, scríobh Aidan Kehoe:
> + /* Control 0; treat as punctuation */
> + SET_RANGE_SYNTAX(0, 32, Spunct);
> + /* DEL plus Control 1 */
> + SET_RANGE_SYNTAX(127, 159, Spunct);
>
> There really oughtta be a better way, don't you think? We really
> ought to have an Scntrl class. (Not seriously suggesting doing
> something about it now, that's obviously a can of worms.)
>
> By the way, what's the logic for Spunct? Something like, these are
> clearly word boundaries, but they can't be whitespace because we don't
> want them "catenating up" with "real" whitespace?
More like, they’re not whitespace, and GNU have gone for punctuation. Weak,
I know.
Ach, sorry, GNU haven’t gone for punctuation for the second range of control
characters, just the first. Their logic for moving to punctuation
syntax--David Kastrup says ‘I think “punctuation” would make more sense.
This designates stuff that is neither a component of words nor of symbols,
and not whitespace.’--holds just as well for the second range of control
characters as for the first, and that control-1 wasn’t included was an
artefact of Stallman doing the commit.
--
Santa Maradona, priez pour moi!
6467
days inactive
6474
days old
4 comments
2 participants
participants (2)
-
Aidan Kehoe -
Stephen J. Turnbull