api.header.include: document it, and fix its default value

While defining api.header.include worked as expected, its default
value was incorrectly defined.  As a result, by default, the generated
parsers still duplicated the content of the generated header instead
of including it.

* data/skeletons/yacc.c (api.header.include): Fix its default value.
* tests/output.at: Check it.
* doc/bison.texi (%define Summary): Document api.header.include.
While at it, move the definition of api.namespace at the proper
place.
This commit is contained in:
Akim Demaille
2020-06-09 07:08:11 +02:00
parent ae5edcc23b
commit a53c6026cd
5 changed files with 112 additions and 34 deletions

20
NEWS
View File

@@ -7,6 +7,26 @@ GNU Bison NEWS
When installed to be relocatable (via configure --enable-relocatable),
bison will now also look for a relocated m4.
** Bug fixes
*** Include the generated header (yacc.c)
Historically, when --defines was used, bison generated a header and pasted
an exact copy of it into the generated parser implementation file. Since
Bison 3.4 it is possible to specify that the header should be `#include`d,
and how. For instance
%define api.header.include {"parse.h"}
or
%define api.header.include {<parser/parse.h>}
Now api.header.include defaults to `"header-basename"`, as was intended in
Bison 3.4, where `header-basename` is the basename of the generated
header. This is disabled when the generated header is `y.tab.h`, to
comply with Automake's ylwrap.
** New features
*** File prefix mapping

8
TODO
View File

@@ -15,16 +15,11 @@ list_get_end (gl_list_t list)
** Bistromathic
- Hitting tab on a line with a syntax error is ugly
- Be robust to existing ~/.inputrc
- How about not evaluating incomplete lines when the text is not finished
(as shells do).
- Caret diagnostics
** Doc
*** api.header.include
** Questions
*** Java
- Should i18n be part of the Lexer? Currently it's a static method of
@@ -54,9 +49,6 @@ enough.
*** calc.at
Stop hard-coding "Calc". Adjust local.at (look for FIXME).
** Counter example generation
See https://github.com/akimd/bison/pull/15.
** Clean up
Rename endtoken as eoftoken.

View File

@@ -335,7 +335,7 @@ m4_define([b4_header_include_if],
[$2])],
[$2])])
m4_if(b4_spec_header_file, [[y.tab.h]],
m4_if(b4_spec_header_file, [y.tab.h], [],
[b4_percent_define_default([[api.header.include]],
[["@basename(]b4_spec_header_file[@)"]])])

View File

@@ -5922,44 +5922,58 @@ Unaccepted @var{variable}s produce an error. Some of the accepted
@var{variable}s are described below.
@c ================================================== api.namespace
@deffn Directive {%define api.namespace} @{@var{namespace}@}
@c ================================================== api.header.include
@deffn Directive {%define api.header.include} @{"header.h"@}
@deffnx Directive {%define api.header.include} @{<header.h>@}
@itemize
@item Languages(s): C++
@item Languages(s): C (@file{yacc.c})
@item Purpose: Specify the namespace for the parser class.
For example, if you specify:
@item Purpose: Specify how the generated parser should include the generated header.
Historically, when option @option{-D}/@option{--defines} was used,
@command{bison} generated a header and pasted an exact copy of it into the
generated parser implementation file. Since Bison 3.6, it is
@code{#include}d as @samp{"@var{basename}.h"}, instead of duplicated, unless
@var{file} is @samp{y.tab}, see below.
The @code{api.header.include} variable allows to control how the generated
parser @code{#include}s the generated header. For instance:
@example
%define api.namespace @{foo::bar@}
%define api.header.include @{"parse.h"@}
@end example
Bison uses @code{foo::bar} verbatim in references such as:
@noindent
or
@example
foo::bar::parser::semantic_type
%define api.header.include @{<parser/parse.h>@}
@end example
However, to open a namespace, Bison removes any leading @code{::} and then
splits on any remaining occurrences:
Using @code{api.header.include} does not change the name of the generated
header, only how it is included.
@example
namespace foo @{ namespace bar @{
class position;
class location;
@} @}
@end example
To work around limitations of Automake's @command{ylwrap} (which runs
@command{bison} with @option{--yacc}), @code{api.header.include} is
@emph{not} predefined when the output file is @file{y.tab.c}. Define it to
avoid the duplication.
@item Accepted Values:
Any absolute or relative C++ namespace reference without a trailing
@code{"::"}. For example, @code{"foo"} or @code{"::foo::bar"}.
An argument for @code{#include}.
@item Default Value:
@code{yy}, unless you used the obsolete @samp{%name-prefix "@var{prefix}"}
directive.
@samp{"@var{header-basename}"}, unless the header file is @file{y.tab.h},
where @var{header-basename} is the name of the generated header, without
directory part. For instance with @command{bison -d calc/parse.y},
@code{api.header.include} defaults to @samp{"parse.h"}, not
@samp{"calc/parse.h"}.
@item History:
Introduced in Bison 3.4. Defaults to @samp{"@var{basename}.h"} since Bison
3.7, unless the header file is @file{y.tab.h}.
@end itemize
@end deffn
@c api.namespace
@c api.header.include
@c ================================================== api.location.file
@@ -6043,6 +6057,46 @@ Introduced in Bison 2.7 for C++ and Java, in Bison 3.4 for C.
@end deffn
@c ================================================== api.namespace
@deffn Directive {%define api.namespace} @{@var{namespace}@}
@itemize
@item Languages(s): C++
@item Purpose: Specify the namespace for the parser class.
For example, if you specify:
@example
%define api.namespace @{foo::bar@}
@end example
Bison uses @code{foo::bar} verbatim in references such as:
@example
foo::bar::parser::semantic_type
@end example
However, to open a namespace, Bison removes any leading @code{::} and then
splits on any remaining occurrences:
@example
namespace foo @{ namespace bar @{
class position;
class location;
@} @}
@end example
@item Accepted Values:
Any absolute or relative C++ namespace reference without a trailing
@code{"::"}. For example, @code{"foo"} or @code{"::foo::bar"}.
@item Default Value:
@code{yy}, unless you used the obsolete @samp{%name-prefix "@var{prefix}"}
directive.
@end itemize
@end deffn
@c api.namespace
@c ================================================== api.parser.class
@deffn Directive {%define api.parser.class} @{@var{name}@}
@itemize @bullet

View File

@@ -30,9 +30,10 @@ m4_define([AT_CHECK_FILES],
[], [$1
])])
# AT_CHECK_OUTPUT(INPUT-FILE, [DIRECTIVES], [FLAGS], EXPECTED-FILES, [STATUS],
# [ADDITIONAL-TESTS], [PRE-TESTS])
# -----------------------------------------------------------------------------
# AT_CHECK_OUTPUT($1 = INPUT-FILE, $2 = [DIRECTIVES], $3 = [FLAGS],
# $4 = EXPECTED-FILES, $5 = [STATUS],
# $6 = [ADDITIONAL-TESTS], $7 = [PRE-TESTS])
# -----------------------------------------------------------------
m4_define([AT_CHECK_OUTPUT],
[AT_SETUP([[Output files: ]$2 $3])[
]$7[
@@ -73,8 +74,19 @@ AT_CHECK_OUTPUT([foo.y], [], [-dv >&-],
[], [],
[AT_CHECK([[case "$PREBISON" in *valgrind*) exit 77;; esac]])])
# Check how api.header.include.
AT_CHECK_OUTPUT([foo.y], [], [-dv -o foo.c],
[foo.c foo.h foo.output])
[foo.c foo.h foo.output], [],
[AT_CHECK([grep '#include "foo.h"' foo.c], [0], [ignore])])
AT_CHECK_OUTPUT([foo.y], [], [-dv -y],
[y.output y.tab.c y.tab.h], [],
[AT_CHECK([grep '#include "y.tab.h"' y.tab.c], [1], [ignore])])
AT_CHECK_OUTPUT([foo.y], [%define api.header.include {"./foo.h"}], [-dv -y],
[y.output y.tab.c y.tab.h], [],
[AT_CHECK([grep '#include "./foo.h"' y.tab.c], [0], [ignore])])
AT_CHECK_OUTPUT([foo.y], [], [-dv -o foo.tab.c],
[foo.output foo.tab.c foo.tab.h])