On 1/15/22 08:02, Akim Demaille wrote:
most of the time it suffices to include something like
%code provides {
int yylex (void);
void yyerror (const char *);
}
How about if we mention this in the "Calling Conventions" section of the
doc? Although the info is present elsewhere, it wasn't in that section
which is sort of where I expected it.
I installed the attached to do that; please feel free to revise/improve/etc.
From 8b96c82b05c68b17c64577be52349b6beae5f655 Mon Sep 17 00:00:00 2001
From: Paul Eggert <[email protected]>
Date: Sat, 15 Jan 2022 11:16:08 -0800
Subject: [PATCH] doc: improve calling-convention doc
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
* doc/bison.texi (Calling Convention, Error Reporting Function):
Suggest ‘%code provides’ to declare yylex and yyerror.
Prompted by this email thread:
https://lists.gnu.org/r/bug-bison/2022-01/msg00002.html
---
doc/bison.texi | 55 +++++++++++++++++++++++++++++++++++++++-----------
1 file changed, 43 insertions(+), 12 deletions(-)
diff --git a/doc/bison.texi b/doc/bison.texi
index 92c8d74e..69c92c0b 100644
--- a/doc/bison.texi
+++ b/doc/bison.texi
@@ -7516,22 +7516,41 @@ numeric code for that character is also the code for the token kind. So
@code{unsigned char} to avoid sign-extension. The null character must not
be used this way, because its code is zero and that signifies end-of-input.
-Here is an example showing these things:
+A simple program might use the following declaration:
@example
+%code provides @{
+ int yylex (void);
+@}
+@end example
+
+@noindent
+and the following definition, either in the grammar file itself or in some
+other module that has @code{#include "y.tab.h"}:
+
+@example
+#include <stdio.h>
+
int
yylex (void)
@{
- @dots{}
- if (c == EOF) /* Detect end-of-input. */
- return YYEOF;
- @dots{}
- else if (c == '+' || c == '-')
- return c; /* Assume token kind for '+' is '+'. */
- @dots{}
- else
- return INT; /* Return the kind of the token. */
- @dots{}
+ for (;;)
+ @{
+ int c = getchar ();
+ if (c == EOF)
+ return YYEOF; /* Report end-of-input. */
+ if (c == '+' || c == '-')
+ return c; /* Assume token kind for '+' is '+'. */
+ if ('0' <= c && c <= '9')
+ @{
+ yylval = c - '0';
+ while ('0' <= (c = getchar ()) && c <= '9')
+ yylval = yylval * 10 + (c - '0');
+ ungetc (c, stdin);
+ return INT; /* Return the kind of the token. */
+ @}
+ @dots{}
+ @}
@}
@end example
@@ -7809,10 +7828,22 @@ In some cases diagnostics like @w{@code{"syntax error"}} are
translated automatically from English to some other language before
they are passed to @code{yyerror}. @xref{Internationalization}.
-The following definition suffices in simple programs:
+A simple program might use the following declaration:
+
+@example
+%code provides @{
+ void yyerror (char const *);
+@}
+@end example
+
+@noindent
+and the following definition, either in the grammar file itself or in some
+other module that has @code{#include "y.tab.h"}:
@example
@group
+#include <stdio.h>
+
void
yyerror (char const *s)
@{
--
2.32.0
