abbrev.c:201: undefined reference to `SYNTAX_ENTRY_FOLLOW_PARENT'

abbrev.c:201: undefined reference to `SYNTAX_ENTRY_FOLLOW_PARENT'

[Brian Millett]

CVS update today has an error in the compilation:

abbrev.o: In function `abbrev_check_chars':
/home/bpm/src/emacs/src/abbrev.c:201: undefined reference to
`SYNTAX_ENTRY_FOLLOW_PARENT'
collect2: ld returned 1 exit status
make[1]: *** [temacs] Error 1
make[1]: Leaving directory `/home/bpm/src/emacs/src'
make: *** [src] Error 2


a cvs diff of the abrev.c shows that 2 days ago the function abbrev_check_chars
was added with the following:

199      /* Copied from SYNTAX in syntax.h, except using FOLLOW_PARENT. */
200 	           Lisp_Object syntax_temp
201  	             = SYNTAX_ENTRY_FOLLOW_PARENT (Vstandard_syntax_table, c);
202  	           if ( (CONSP (syntax_temp)
203  	                 ? (enum syntaxcode) (XINT (XCAR (syntax_temp)) & 0xff)
204 	  	                 : Swhitespace) != Sword ) badchars[nbad++] = c;

But SYNTAX_ENTRY_FOLLOW_PARENT is not in syntax.h

Thanks.
-- 
Brian Millett - [ Thirteen/Psi Cop, "A Spider in the Web"]
"Ours now. All ours."

Re: abbrev.c:201: undefined reference to `SYNTAX_ENTRY_FOLLOW_PARENT'

[Glenn Morris]

Brian Millett wrote:

> CVS update today

Wrong mailing list then.

> But SYNTAX_ENTRY_FOLLOW_PARENT is not in syntax.h

Yes it is, in the CVS trunk. Crystal ball suggests you may be using a
branch without it.

Re: abbrev.c:201: undefined reference to `SYNTAX_ENTRY_FOLLOW_PARENT'

[Brian Millett]

Glenn Morris escribío:
> Brian Millett wrote:
> 
>> CVS update today
> 
> Wrong mailing list then.

Ok, so which list?  Bugs only for Current Stable Release?

> 
>> But SYNTAX_ENTRY_FOLLOW_PARENT is not in syntax.h
> 
> Yes it is, in the CVS trunk. Crystal ball suggests you may be using a
> branch without it.
> 

May be,

:pserver:anonymous@cvs.savannah.gnu.org:/cvsroot/emacs

did a cvs co emacs

Which branch then?

Thanks.

-- 
Brian Millett - [ Londo, "A Voice in the Wilderness I"]
"Now, I go to spread happiness through the rest of the station. It is a
 terrible responsibility, but I have learned to live with it."

Re: abbrev.c:201: undefined reference to `SYNTAX_ENTRY_FOLLOW_PARENT'

[Glenn Morris]

Brian Millett wrote:

> Ok, so which list?  Bugs only for Current Stable Release?

M-x report-emacs-bug does the right thing, and also provides useful
debugging info. Generally, bug-gnu-emacs for releases,
emacs-pretest-bug for CVS. But we're here now...

> :pserver:anonymous@cvs.savannah.gnu.org:/cvsroot/emacs
>
> did a cvs co emacs

Then I'm baffled. That should get you the CVS trunk, and
SYNTAX_ENTRY_FOLLOW_PARENT is sitting there in src/syntax.h, at line
71/86. Do you have a src/CVS/Tag file, and if so what are the
contents?

Does your src/syntax.h look the same as:
http://cvs.savannah.gnu.org/viewvc/emacs/src/syntax.h?root=emacs&view=markup

Re: abbrev.c:201: undefined reference to `SYNTAX_ENTRY_FOLLOW_PARENT'

[Brian Millett]

Glenn Morris escribío:
> Brian Millett wrote:
> 
>> Ok, so which list?  Bugs only for Current Stable Release?
> 
> M-x report-emacs-bug does the right thing, and also provides useful
> debugging info. Generally, bug-gnu-emacs for releases,
> emacs-pretest-bug for CVS. But we're here now...

Thanks for clarifying that.

>> :pserver:anonymous@cvs.savannah.gnu.org:/cvsroot/emacs
>>
>> did a cvs co emacs
> 
> Then I'm baffled. That should get you the CVS trunk, and
> SYNTAX_ENTRY_FOLLOW_PARENT is sitting there in src/syntax.h, at line
> 71/86. Do you have a src/CVS/Tag file, and if so what are the
> contents?

Temacs-unicode-2

> Does your src/syntax.h look the same as:
> http://cvs.savannah.gnu.org/viewvc/emacs/src/syntax.h?root=emacs&view=markup
> 

Almost:
[bpm]$ diff -wruN syntax.h.new syntax.h
--- syntax.h.new        2007-07-19 22:33:53.000000000 -0500
+++ syntax.h    2007-01-26 00:15:07.000000000 -0600
@@ -59,36 +59,13 @@
 /* Set the syntax entry VAL for char C in table TABLE.  */

 #define SET_RAW_SYNTAX_ENTRY(table, c, val)                            \
-  ((((c) & 0xFF) == (c))                                               \
-   ? (XCHAR_TABLE (table)->contents[(unsigned char) (c)] = (val))      \
-   : Faset ((table), make_number (c), (val)))
-
-/* Fetch the syntax entry for char C in syntax table TABLE.
-   This macro is called only when C is less than CHAR_TABLE_ORDINARY_SLOTS.
-   Do inheritance.  */
+  CHAR_TABLE_SET ((table), c, (val))

-#ifdef __GNUC__
-#define SYNTAX_ENTRY_FOLLOW_PARENT(table, c)                   \
-  ({ Lisp_Object _syntax_tbl = (table);                                \
-     Lisp_Object _syntax_temp = XCHAR_TABLE (_syntax_tbl)->contents[(c)]; \
-     while (NILP (_syntax_temp))                               \
-       {                                                       \
-        _syntax_tbl = XCHAR_TABLE (_syntax_tbl)->parent;       \
-        if (NILP (_syntax_tbl))                                \
-          break;                                               \
-        _syntax_temp = XCHAR_TABLE (_syntax_tbl)->contents[(c)]; \
-       }                                                       \
-     _syntax_temp; })
-#else
-extern Lisp_Object syntax_temp;
-extern Lisp_Object syntax_parent_lookup P_ ((Lisp_Object, int));
+/* Set the syntax entry VAL for char-range RANGE in table TABLE.
+   RANGE is a cons (FROM . TO) specifying the range of characters.  */

-#define SYNTAX_ENTRY_FOLLOW_PARENT(table, c)           \
-  (syntax_temp = XCHAR_TABLE (table)->contents[(c)],   \
-   (NILP (syntax_temp)                                 \
-    ? syntax_parent_lookup (table, (c))                        \
-    : syntax_temp))
-#endif
+#define SET_RAW_SYNTAX_ENTRY_RANGE(table, range, val)  \
+  Fset_char_table_range ((table), (range), (val))

 /* SYNTAX_ENTRY fetches the information from the entry for character C
    in syntax table TABLE, or from globally kept data (gl_state).
@@ -106,12 +83,7 @@
 #  define CURRENT_SYNTAX_TABLE current_buffer->syntax_table
 #endif

-#define SYNTAX_ENTRY_INT(c)                            \
-  ((((c) & 0xFF) == (c))                               \
-   ? SYNTAX_ENTRY_FOLLOW_PARENT (CURRENT_SYNTAX_TABLE, \
-                                (unsigned char) (c))   \
-   : Faref (CURRENT_SYNTAX_TABLE,                      \
-           make_number (c)))
+#define SYNTAX_ENTRY_INT(c) CHAR_TABLE_REF (CURRENT_SYNTAX_TABLE, (c))

 /* Extract the information from the entry for character C
    in the current syntax table.  */
@@ -138,6 +110,7 @@
       ? XCDR (_syntax_temp)                                            \
       : Qnil); })
 #else
+extern Lisp_Object syntax_temp;
 #define SYNTAX(c)                                                      \
   (syntax_temp = SYNTAX_ENTRY ((c)),                                   \
    (CONSP (syntax_temp)                                                        \


-- 
Brian Millett - [ Sinclair, "The Coming of Shadows"]
"Stay close to the Vorlon and watch out for the Shadows. They move when
 you're not looking at them."

Re: abbrev.c:201: undefined reference to `SYNTAX_ENTRY_FOLLOW_PARENT'

[Glenn Morris]

Brian Millett wrote:

> Temacs-unicode-2

Right, so this is the unicode-2 branch and not the trunk, making this
the same problem that was reported in emacs-devel today. A unicode-2
developer who knows what has replaced the relevant deleted macro in
this branch will no doubt fix it before long.

Re: abbrev.c:201: undefined reference to `SYNTAX_ENTRY_FOLLOW_PARENT'

[Brian Millett]

Glenn Morris escribío:
> Brian Millett wrote:
> 
>> Temacs-unicode-2
> 
> Right, so this is the unicode-2 branch and not the trunk, making this
> the same problem that was reported in emacs-devel today. A unicode-2
> developer who knows what has replaced the relevant deleted macro in
> this branch will no doubt fix it before long.
> 

I'm sorry, after sending the email, I went back in history and looked at the
command that I used to co the first initial.

co -r emacs-unicode-2 emacs

Thanks, and I am sorry for the run around.  I need to pay more attention.

When I have problems with this branch, should I report them to emacs-devel?

-- 
Brian Millett - [ Lennier and Delenn, "Chrysalis"]
"But what if you're wrong?"
'Then speak well of me when I'm gone.