gmane.comp.parsers.bison.bugs

bison-2.5.1_rc2 released [beta]

Akim Demaille — 2012-05-25T17:19:31

The Bison team is very happy to announce the forthcoming release
of Bison 2.5.1, which addresses quite a few issues from Bison 2.5.

Here are the compressed sources:
 ftp://alpha.gnu.org/gnu/bison/bison-2.5.1_rc2.tar.gz   (2.7MB)
 ftp://alpha.gnu.org/gnu/bison/bison-2.5.1_rc2.tar.xz   (1.5MB)

Here are the GPG detached signatures[*]:
 ftp://alpha.gnu.org/gnu/bison/bison-2.5.1_rc2.tar.gz.sig
 ftp://alpha.gnu.org/gnu/bison/bison-2.5.1_rc2.tar.xz.sig

Use a mirror for higher download bandwidth:
 http://www.gnu.org/order/ftp.html

[*] Use a .sig file to verify that the corresponding file (without the
.sig suffix) is intact.  First, be sure to download both the .sig file
and the corresponding tarball.  Then, run a command like this:

 gpg --verify bison-2.5.1_rc2.tar.gz.sig

If that command fails because you don't have the required public key,
then run this command to import it:

 gpg --keyserver keys.gnupg.net --recv-keys 78D5264E

and rerun the 'gpg --verify' command.

This release was bootstrapped with the following tools:
 Autoconf 2.69
 Automake 1.12
 Flex 2.5.35
 Gettext 0.18.1
 Gnulib v0.0-7415-g338dd5d

NEWS

* Changes in version 2.5.1_rc2 (2012-05-23):

** Future changes:

 The next major release will drop support for generating parsers in K&R C.

** yacc.c: YYBACKUP works as expected.

** glr.c improvements:

*** Location support is eliminated when not requested:

 GLR parsers used to include location-related code even when locations were
 not requested, and therefore not even usable.

*** __attribute__ is preserved:

 __attribute__ is no longer disabled when __STRICT_ANSI__ is defined (i.e.,
 when -std is passed to GCC).

** lalr1.java: several fixes:

 The Java parser no longer throws ArrayIndexOutOfBoundsException if the
 first token leads to a syntax error.  Some minor clean ups.

** Changes for C++:

*** C++11 compatibility:

 C and C++ parsers use "nullptr" instead of "0" when __cplusplus is 201103L
 or higher.

*** Header guards

 The header files such as "parser.hh", "location.hh", etc. used a constant
 name for preprocessor guards, for instance:

 #ifndef BISON_LOCATION_HH
 # define BISON_LOCATION_HH
 ...
 #endif // !BISON_LOCATION_HH

 The inclusion guard is now computed from "PREFIX/FILE-NAME", where lower
 case characters are converted to upper case, and series of
 non-alphanumerical characters are converted to an underscore.

 With "bison -o lang++/parser.cc", "location.hh" would now include:

 #ifndef YY_LANG_LOCATION_HH
 # define YY_LANG_LOCATION_HH
 ...
 #endif // !YY_LANG_LOCATION_HH

*** C++ locations:

 The position and location constructors (and their initialize methods)
 accept new arguments for line and column.  Several issues in the
 documentation were fixed.

** liby is no longer asking for "rpl_fprintf" on some platforms.

** Changes in the manual:

*** %printer is documented

 The "%printer" directive, supported since at least Bison 1.50, is finally
 documented.  The "mfcalc" example is extended to demonstrate it.

 For consistency with the C skeletons, the C++ parsers now also support
 "yyoutput" (as an alias to "debug_stream ()").

*** Several improvements have been made:

 The layout for grammar excerpts was changed to a more compact scheme.
 Named references are motivated.  The description of the automaton
 description file (*.output) is updated to the current format.  Incorrect
 index entries were fixed.  Some other errors were fixed.

** Building bison:

*** Conflicting prototypes with recent/modified Flex.

 Fixed build problems with the current, unreleased, version of Flex, and
 some modified versions of 2.5.35, which have modified function prototypes.

*** Warnings during the build procedure have been eliminated.

*** Several portability problems in the test suite have been fixed:

 This includes warnings with some compilers, unexpected behavior of tools
 such as diff, warning messages from the test suite itself, etc.

*** The install-pdf target work properly:

 Running "make install-pdf" (or -dvi, -html, -info, and -ps) no longer
 halts in the middle of its course.

Building bison from master fails: "conflicting types for'code_get_leng'"

Stefano Lattarini — 2012-05-13T10:46:59

When I try to bootstrap and build bison from latest master on my Debian
system, I see this failure:

  ...
  CC       src/src_bison-reduce.o
  CC       src/src_bison-relation.o
  CC       src/src_bison-scan-code-c.o
In file included from src/scan-code-c.c:3:0:
src/scan-code.c:789:11: error: conflicting types for 'code_get_leng'
./src/flex-scanner.h:28:1: note: previous declaration of 'code_get_leng' was here
In file included from src/scan-code-c.c:3:0:
src/scan-code.c:2344:11: error: conflicting types for 'code_get_leng'
./src/flex-scanner.h:28:1: note: previous declaration of 'code_get_leng' was here
Makefile:3833: recipe for target 'src/src_bison-scan-code-c.o' failed
make[2]: *** [src/src_bison-scan-code-c.o] Error 1

Attached is the config.log.  I'm using Automake and Autoconf from master, and
flex 2.5.36.  Let me know if you need more information.

Regards,
  Stefano

nuke strncpy

Jim Meyering — 2012-05-03T21:18:28

I noticed that there were some unnecessary uses of strncpy in bison.
The first patch removes them.
Along the way I noticed a 3-line xmalloc+memcpy+zero-one-byte sequence
that can be replaced by a single-line xstrdup use.  That's #2.
And #3 merely regenerates parse-gram.[ch].

[PATCH 1/3] maint: s/strncpy/memcpy/, when equivalent
[PATCH 2/3] maint: simplify parse-gram.y
[PATCH 3/3] maint: regen src/parse-gram.[ch]

Use "git" instead of "CVS" in the description of the bison-patchesmailing list

Stefano Lattarini — 2012-04-01T19:35:01

In the mailing list description here:

  <https://lists.gnu.org/mailman/listinfo/bison-patches>

I read:

  ``If you have found a bug in Bison, or have made an improvement, please
    send a patch, made for the CVS sources, to this mailing list.''

But bison doesn't use CVS anymore (thanks $DEITY :-) so that a correct
description should refer to the "git sources" or "git repository" (and
a link to said repository wouldn0t hurt either, IMHO).

Regards,
  Stefano

[PATCH] master: Doxygen generation broken

Tim Landscheidt — 2012-03-27T22:41:28

Hi,

the transition from recursive makefiles to the present ones
broke Doxygen generation.  The first attached patch fixes
this.

  Then running Doxygen 1.6.1 gave some warnings about obso-
lete options in Doxyfile that don't seem to be important.
Also, Doxygen complained about not being able to find header
files.  First I thought that these were gnulib's, so I added
lib to the INCLUDE_PATH, but Doxygen kept complaining even
about system header files like stddef.h included with angle
brackets, so I ignored it.

  Finally, it revealed a small typo in a Doxygen comment in
src/InadequacyList.h which the other attached patch fixes.

Tim

Bugs at Savannah

Tim Landscheidt — 2012-03-23T20:41:55

Hi,

with 2.5.1 on the horizon, a look at the remaining bugs on
Savannah (https://savannah.gnu.org/support/?group=bison):

- "Testcase #50 failing non-deterministically for bison
  2.4": I assume this was fixed by 47fa57; at least
  lib/subpipe.c is gone anyway :-).

- "Updating to bison 2.4 breaks WebKit": This seems to have
  been a duplicate of
  http://permalink.gmane.org/gmane.comp.parsers.bison.bugs/3078
  and solved on WebKit's side per
  https://bugs.webkit.org/show_bug.cgi?id=22205.

- "YYPOPSTACK used inconsistently": The user infers from
  yacc.c to "garbage collected environments".  The code in
  lalr1.cc looks fine.

- "--no-parser and %no-parser option is broken": Probably
  "Wont Do".

- "ambiguity problem with bison GLR parser": The reporter
  initially misread the manual.

- "Possible bug in overflow handling": After staring for
  half an hour at yacc.c's code paths, I think this is not a
  bug (and hope that saving a few bytes is really worth the
  very "special" memory management Bison uses :-)).

- "parser stack overflow handling + documentation on it":
  The bug's title should be "Document yyoverflow()".

- "C++ template makes it very difficult to override func-
  tions": This is probably a matter of taste.

Tim

[GNU Bison 2.5] testsuite: 229 232 failed

gccbison — 2012-03-13T15:54:39

229: GNU Cim Grammar: Canonical LR(1)                FAILED 
(existing.at:780)
232: GNU pic (Groff 1.18.1) Grammar: Canonical LR(1) FAILED 
(existing.at:1400)
cc1plus: warning: command line option '-Wbad-function-cast' is valid 
for C/ObjC but not for C++ [enabled by default]
cc1plus: warning: command line option '-Wmissing-prototypes' is valid 
for Ada/C/ObjC but not for C++ [enabled by default]
cc1plus: warning: command line option '-Wstrict-prototypes' is valid 
for Ada/C/ObjC but not for C++ [enabled by default]
and multiple of these errors:
input.y:224:7: error: line number out of range [-Werror]
Using gcc (GCC) 4.8.0 20120311 (experimental)
Copyright (C) 2012 Free Software Foundation, Inc.
[GNU Bison 2.5] testsuite: 229 232 failed
Not sure if this a Bison or GCC issue

help

Aravind Thangavel — 2012-03-09T04:53:37

i'm trying to compile a really old c++ code. In which there is a bison
command

bison -d -b exp  -t exp.yacc

generating a file

exp.tab.cacc
exp.tab.hacc

But the next command is this

mv exp.tab.c exp.tab.cc

Generating an error exp.tab.c does not exists.

So what is the solution?

[GNU Bison 2.5] testsuite: 247 failed

Martin & Dunja Zaun — 2012-03-07T03:45:07

$ uname -a
SunOS hoops-s11 5.11 11.0 i86pc i386 i86pc

$ gcc --version
gcc (GCC) 4.5.2
Copyright (C) 2010 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

$ autoconf --version
autoconf (GNU Autoconf) 2.63
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv2+: GNU GPL version 2 or later
<http://gnu.org/licenses/old-licenses/gpl-2.0.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

Written by David J. MacKenzie and Akim Demaille.

"Syntax error before ," in generated .c file

Gizmotechspert — 2012-03-05T13:49:53

dict_parser.c:1068: error: syntax error before ',' token

Generated code:
1065   /* YYCHAR is either YYEMPTY or YYEOF or a valid lookahead symbol.  */
1066   if (yychar == YYEMPTY)
1067     {
1068       YYDPRINTF ((stderr, "Reading a token: "));
1069       yychar = YYLEX;
1070     }

The variable YYDPRINTF is defined and it isn't supposed to throw a syntax
error at this place. No clue what's going wrong or how to fix this issue.

[GNU Bison 2.5] testsuite: 3 4 7 8 16 21 33 36 45 68 69 70 92 93 9495 139 143 145 151 160 220 221 222 228 233 246 failed

Lie Yan — 2012-03-02T12:52:57

master: Dependencies broken

Tim Landscheidt — 2012-02-23T15:18:07

Hi,

on a freshly cloned and bootstrapped copy:

| [tim< at >passepartout ~/src/bison]$ make
|   GEN    examples/calc++/calc.stamp
| extexi: ignoring rpcalc.y
| extexi: ignoring mfcalc.y
| extexi: ignoring calc.h
| extexi: extracting calc++-driver.hh
| extexi: extracting calc++-driver.cc
| extexi: extracting calc++-parser.yy
| extexi: extracting calc++-scanner.ll
| extexi: extracting calc++.cc
|   GEN    examples/calc++/calc++-scanner.ll
| make: *** Keine Regel vorhanden, um das Target »examples/calc++/stack.hh«,
|   benötigt von »all«, zu erstellen.  Schluss.
| [tim< at >passepartout ~/src/bison]$

Continuing ignorantly:

| [tim< at >passepartout ~/src/bison]$ make -k
| make: *** Keine Regel vorhanden, um das Target »examples/calc++/stack.hh«,
|   benötigt von »all«, zu erstellen.
| make: *** Keine Regel vorhanden, um das Target »examples/calc++/position.hh«,
|   benötigt von »all«, zu erstellen.
| make: *** Keine Regel vorhanden, um das Target »examples/calc++/location.hh«,
|   benötigt von »all«, zu erstellen.
|   YACC   examples/calc++/calc++-parser.cc
| /home/tim/src/bison/./tests/bison: line 30: /home/tim/src/bison/src/bison: No such file or directory
| make: *** [examples/calc++/calc++-parser.cc] Fehler 1
|   GEN    examples/mfcalc/mfcalc.stamp
| [...]

So beside the missing rules for examples/calc++/*.hh, there
probably needs to be some dependency on src/bison.

Tim

GLR: ambiguity reports are wrong

Akim Demaille — 2012-02-21T11:20:46

On this grammar:


with input a b c d, the current code reports:


instead of:


This is a regression I introduced in master only with
commit 783aa653f4ca70a75919c8516b950494c612cbfe.
http://lists.gnu.org/archive/html/bison-patches/2008-11/msg00291.html

new test failures

Jim Meyering — 2012-02-17T07:53:45

Hi Akim,

It looks like you're adding tests, and once I got past
the link failure (missing -lm) I see they're failing.
Presuming you're still working there, I'll simply
post the test-suite.log file:

====================================================
   GNU Bison 2.4.669-c119-dirty: ./test-suite.log
====================================================

# TOTAL: 2
# PASS:  0
# SKIP:  0
# XFAIL: 0
# FAIL:  2
# XPASS: 0
# ERROR: 0

.. contents:: :depth: 2

FAIL: examples/calc++/test
==========================

--(end of buffer or a NUL)
--accepting rule at line 46 ("1")
--accepting rule at line 39 ("+")
--accepting rule at line 46 ("2")
--accepting rule at line 40 ("*")
--accepting rule at line 46 ("3")
--accepting rule at line 36 ("
")
--accepting rule at line 54 ("=")
2.1: invalid character
--accepting rule at line 35 (" ")
--accepting rule at line 46 ("1")
2.3: syntax error, unexpected number, expecting end of file
test: FAIL:  1 (expected status: 0, effective: 1
1.1: syntax error, unexpected end of file, expecting ( or identifier or number
test: FAIL:  2 (expected status: 0, effective: 1
Starting parse
Entering state 0
Reducing stack by rule 3 (line 9826):
-> $$ = nterm assignments (-:1.1: )
Stack now 0
Entering state 2
Reading a token: Next token is token "(" (1.1: )
Shifting token "(" (1.1: )
Entering state 4
Reading a token: Next token is token "number" (1.2: 1)
Shifting token "number" (1.2: 1)
Entering state 6
Reducing stack by rule 11 (line 9840):
   $1 = token "number" (1.2: 1)
-> $$ = nterm exp (1.2: 1)
Stack now 0 2 4
Entering state 10
Reading a token: Next token is token "+" (1.3: )
Shifting token "+" (1.3: )
Entering state 13
Reading a token: Next token is token "number" (1.4: 2)
Shifting token "number" (1.4: 2)
Entering state 6
Reducing stack by rule 11 (line 9840):
   $1 = token "number" (1.4: 2)
-> $$ = nterm exp (1.4: 2)
Stack now 0 2 4 10 13
Entering state 19
Reading a token: Next token is token ")" (1.5: )
Reducing stack by rule 5 (line 9834):
   $1 = nterm exp (1.2: 1)
   $2 = token "+" (1.3: )
   $3 = nterm exp (1.4: 2)
-> $$ = nterm exp (1.2-4: 3)
Stack now 0 2 4
Entering state 10
Next token is token ")" (1.5: )
Shifting token ")" (1.5: )
Entering state 16
Reducing stack by rule 9 (line 9838):
   $1 = token "(" (1.1: )
   $2 = nterm exp (1.2-4: 3)
   $3 = token ")" (1.5: )
-> $$ = nterm exp (1.1-5: 3)
Stack now 0 2
Entering state 8
Reading a token: Next token is token "*" (1.7: )
Shifting token "*" (1.7: )
Entering state 14
Reading a token: Next token is token "number" (1.9: 3)
Shifting token "number" (1.9: 3)
Entering state 6
Reducing stack by rule 11 (line 9840):
   $1 = token "number" (1.9: 3)
-> $$ = nterm exp (1.9: 3)
Stack now 0 2 8 14
Entering state 20
Reducing stack by rule 7 (line 9836):
   $1 = nterm exp (1.1-5: 3)
   $2 = token "*" (1.7: )
   $3 = nterm exp (1.9: 3)
-> $$ = nterm exp (1.1-9: 9)
Stack now 0 2
Entering state 8
Reading a token: Next token is token "end of file" (2.1: )
Reducing stack by rule 1 (line 9822):
   $1 = nterm assignments (-:1.1: )
   $2 = nterm exp (1.1-9: 9)
-> $$ = nterm unit (-:1.1-9: )
Stack now 0
Entering state 1
Next token is token "end of file" (2.1: )
Shifting token "end of file" (2.1: )
Entering state 3
Cleanup: popping token "end of file" (2.1: )
Cleanup: popping nterm unit (-:1.1-9: )
cat: out_eff: No such file or directory
test: FAIL:  3 (expected output: 7, effective:
test: PASS:  4
3.1: syntax error, unexpected end of file, expecting ( or identifier or number
test: PASS:  5

FAIL: examples/mfcalc/test
==========================

syntax error
test: FAIL:  1 (expected output: 7, effective: 7
0
syntax error
test: FAIL:  2 (expected output: 7, effective: 7
0
test: FAIL:  3 (expected output: 9, effective:

[PATCH] examples: link mfcalc with -lm for uses of pow, cos, atan,etc.

Jim Meyering — 2012-02-17T07:37:56

Building from cloned sources, I got this:

...
gcc -std=gnu99 -Wall -Wextra -Wno-sign-compare -Wcast-align -Wcast-qual -Wformat
-Wpointer-arith -Wwrite-strings -Wbad-function-cast -Wmissing-declarations -Wmiss
ing-prototypes -Wshadow -Wstrict-prototypes -g -O2   -o examples/mfcalc/mfcalc ex
amples/mfcalc/examples_mfcalc_mfcalc-mfcalc.o lib/libbison.a
examples/mfcalc/examples_mfcalc_mfcalc-mfcalc.o: In function `yyparse':
/h/j/w/co/bison/../../doc/bison.texinfo:2419: undefined reference to `pow'
examples/mfcalc/examples_mfcalc_mfcalc-mfcalc.o:(.rodata+0x88): undefined reference to `atan'
examples/mfcalc/examples_mfcalc_mfcalc-mfcalc.o:(.rodata+0x98): undefined reference to `cos'
examples/mfcalc/examples_mfcalc_mfcalc-mfcalc.o:(.rodata+0xa8): undefined reference to `exp'
examples/mfcalc/examples_mfcalc_mfcalc-mfcalc.o:(.rodata+0xb8): undefined reference to `log'
examples/mfcalc/examples_mfcalc_mfcalc-mfcalc.o:(.rodata+0xc8): undefined reference to `sin'
examples/mfcalc/examples_mfcalc_mfcalc-mfcalc.o:(.rodata+0xd8): undefined reference to `sqrt'
collect2: error: ld returned 1 exit status
make[3]: *** [examples/mfcalc/mfcalc] Error 1
make[3]: *** Waiting for unfinished jobs....

Here's one way to fix it:

From cbef12deb94918d4f24286d498bcedba4938d6fa Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering< at >redhat.com>
Date: Fri, 17 Feb 2012 08:36:57 +0100
Subject: [PATCH] examples: link mfcalc with -lm for uses of pow, cos, atan,
 etc.

* examples/mfcalc/local.mk (examples_mfcalc_mfcalc_LDADD): Define.
---
 examples/mfcalc/local.mk |    1 +
 1 file changed, 1 insertion(+)

diff --git a/examples/mfcalc/local.mk b/examples/mfcalc/local.mk
index aae9e04..e34d8d7 100644
--- a/examples/mfcalc/local.mk
+++ b/examples/mfcalc/local.mk
< at >< at > -49,6 +49,7 < at >< at > mfcalc_sources =\
   $(mfcalc_extracted)

 check_PROGRAMS += examples/mfcalc/mfcalc
+examples_mfcalc_mfcalc_LDADD = -lm
 examples_mfcalc_mfcalc_SOURCES =\
   $(mfcalc_sources)

--
1.7.9.1.232.g1a183

[PATCH] tests: avoid c++ failure due to lack of getenv decl

Jim Meyering — 2012-02-16T10:11:12

I noticed that there was a single test failure when
running "make check" using the latest gcc/g++ from git (4.7.x).

Pushed to master, as obvious:

From 292402a9b2a9fe3440008e42a41b01185d9a4f9d Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering< at >redhat.com>
Date: Thu, 16 Feb 2012 11:03:31 +0100
Subject: [PATCH] tests: avoid c++ failure due to lack of getenv decl

* tests/c++.at (Syntax error as exception): Avoid spurious failure
at least when compiling with g++-4.7.x due to lack of declaration
of getenv.  Include <stdlib.h>.
---
 tests/c++.at |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/tests/c++.at b/tests/c++.at
index d29b6f6..3510498 100644
--- a/tests/c++.at
+++ b/tests/c++.at
< at >< at > -461,6 +461,7 < at >< at > AT_DATA_GRAMMAR([[input.yy]],

 %code
 {
+  #include <stdlib.h>
   int yylex (yy::parser::semantic_type *);
 }

< at >< at > -510,7 +511,7 < at >< at > yy::parser::error (const std::string &m)
 }

 int
-main ()
+main (void)
 {
   yy::parser parser;
   parser.set_debug_level(!!getenv("YYDEBUG"));
--
1.7.9.1.232.g1a183

master: bootstrap doesn't clean up properly

Tim Landscheidt — 2012-02-11T17:50:38

Hi,

on a freshly cloned copy, after "git submodule update --init
&& ./bootstrap && ./configure && make && make check", "git
status" gives:

| # On branch master
| # Untracked files:
| #   (use "git add <file>..." to include in what will be committed)
| #
| #build-aux/config.rpath~
| #build-aux/snippet/
| #lib/glthread/
| #m4/codeset.m4~
| #m4/fcntl-o.m4~
| #m4/gettext.m4~
| #m4/glibc2.m4
| #m4/glibc2.m4~
| #m4/glibc21.m4~
| #m4/iconv.m4~
| #m4/intdiv0.m4
| #m4/intdiv0.m4~
| #m4/intl.m4
| #m4/intl.m4~
| #m4/intldir.m4
| #m4/intldir.m4~
| #m4/intlmacosx.m4~
| #m4/intmax.m4
| #m4/intmax.m4~
| #m4/inttypes-pri.m4~
| #m4/inttypes_h.m4~
| #m4/lcmessage.m4
| #m4/lcmessage.m4~
| #m4/lib-ld.m4~
| #m4/lib-link.m4~
| #m4/lib-prefix.m4~
| #m4/lock.m4
| #m4/lock.m4~
| #m4/longlong.m4~
| #m4/nls.m4~
| #m4/po.m4~
| #m4/printf-posix.m4~
| #m4/progtest.m4~
| #m4/size_max.m4~
| #m4/stdint_h.m4~
| #m4/threadlib.m4~
| #m4/uintmax_t.m4
| #m4/uintmax_t.m4~
| #m4/visibility.m4
| #m4/visibility.m4~
| #m4/wchar_t.m4~
| #m4/wint_t.m4~
| #m4/xsize.m4~
| nothing added to commit but untracked files present (use "git add" to track)

Most of them would probably be hidden by a common
.git/info/exclude of "*~", but:

a) Bison shouldn't rely on that, and
b) there are several *.m4 and directories in the above list.

Tim

"check" table missing comment in bison.m4

Tim Landscheidt — 2012-02-11T01:18:56

Hi,

the "check" table (yycheck_[] in the Java incarnation) has
no comment in bison.m4's b4_integral_parser_tables_map which
leads to a misindentation in the generated Java scanner:

| [...]

| /* YYTABLE[YYPACT[STATE-NUM]].  What to do in state STATE-NUM.  If
|    positive, shift that token.  If negative, reduce the rule which
|    number is the opposite.  If YYTABLE_NINF, syntax error.  */
|   private static final byte yytable_[] = yytable_init();
|   private static final byte[] yytable_init()
|   {
| [...]
|   }

| private static final byte yycheck_[] = yycheck_init();
|   private static final byte[] yycheck_init()
|   {
| [...]
|   }

| /* STOS_[STATE-NUM] -- The (internal number of the) accessing
|    symbol of state STATE-NUM.  */
|   private static final byte yystos_[] = yystos_init();
|   private static final byte[] yystos_init()
|   {
| [...]

Rather than accounting for tables without a comment, I think
it'd be much easier to just document the table :-).

Tim

[PATCH] maint: force "make syntax-check" to pass by skipping failingtests

Jim Meyering — 2012-01-29T20:27:54

There are a few failing syntax-check tests.
For now, I propose to simply skip them.
If folks have time/desire to address the issues
or exempt the offending files, these tests can
easily be reenabled later.

I've just addressed the sole violation of the m4 quote check
in the 2nd patch below:

From bffa42b866e7d4b2d6257904679d77b905ab5fca Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering< at >redhat.com>
Date: Sat, 28 Jan 2012 17:11:43 +0100
Subject: [PATCH 1/2] maint: force "make syntax-check" to pass by skipping
 failing tests

* cfg.mk (local-checks-to-skip): Skip all currently-failing tests.
Remove changelog-check; it's long gone.
---
 cfg.mk |   24 ++++++++++++++++++------
 1 files changed, 18 insertions(+), 6 deletions(-)

diff --git a/cfg.mk b/cfg.mk
index 6b3deb9..2a4d711 100644
--- a/cfg.mk
+++ b/cfg.mk
< at >< at > -33,12 +33,24 < at >< at > url_dir_list = \
   ftp://$(gnu_rel_host)/gnu/bison

 # Tests not to run as part of "make distcheck".
-# Exclude changelog-check here so that there's less churn in ChangeLog
-# files -- otherwise, you'd need to have the upcoming version number
-# at the top of the file for each `make distcheck' run.
-local-checks-to-skip = \
-  changelog-check \
-  sc_immutable_NEWS
+local-checks-to-skip =\
+  sc_immutable_NEWS\
+  sc_bindtextdomain\
+  sc_error_message_period\
+  sc_error_message_uppercase\
+  sc_m4_quote_check\
+  sc_program_name\
+  sc_prohibit_HAVE_MBRTOWC\
+  sc_prohibit_always-defined_macros\
+  sc_prohibit_always_true_header_tests\
+  sc_prohibit_atoi_atof\
+  sc_prohibit_doubled_word\
+  sc_prohibit_empty_lines_at_EOF\
+  sc_prohibit_magic_number_exit\
+  sc_prohibit_quotearg_without_use\
+  sc_prohibit_strcmp\
+  sc_unmarked_diagnostics\
+  sc_useless_cpp_parens

 # The local directory containing the checked-out copy of gnulib used in
 # this release.  Used solely to get a date for the "announcement" target.
--
1.7.9.1.g63eb


From c7313af131e13505c44110c5bb9939ab161d675f Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering< at >redhat.com>
Date: Sun, 29 Jan 2012 21:26:43 +0100
Subject: [PATCH 2/2] maint: reenable sc_m4_quote_check

* cfg.mk (local-checks-to-skip): Reenable sc_m4_quote_check.
* m4/dmalloc.m4: Add quotes.
---
 cfg.mk        |    1 -
 m4/dmalloc.m4 |    2 +-
 2 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/cfg.mk b/cfg.mk
index 2a4d711..515ac25 100644
--- a/cfg.mk
+++ b/cfg.mk
< at >< at > -38,7 +38,6 < at >< at > local-checks-to-skip =\
   sc_bindtextdomain\
   sc_error_message_period\
   sc_error_message_uppercase\
-  sc_m4_quote_check\
   sc_program_name\
   sc_prohibit_HAVE_MBRTOWC\
   sc_prohibit_always-defined_macros\
diff --git a/m4/dmalloc.m4 b/m4/dmalloc.m4
index f724670..90e72f6 100644
--- a/m4/dmalloc.m4
+++ b/m4/dmalloc.m4
< at >< at > -12,7 +12,7 < at >< at > AC_ARG_WITH(dmalloc,
                           http://www.dmalloc.com/dmalloc.tar.gz],
 [if test "$withval" = yes; then
   AC_MSG_RESULT(yes)
-  AC_DEFINE(WITH_DMALLOC,1,
+  AC_DEFINE([WITH_DMALLOC],1,
             [Define if using the dmalloc debugging malloc package])
   LIBS="$LIBS -ldmalloc"
   LDFLAGS="$LDFLAGS -g"
--
1.7.9.1.g63eb

[PATCH] do not ignore errors like ENOSPC,EIO when writing to stdout

Jim Meyering — 2012-01-29T18:22:57

This is incomplete in that I haven't added a test for it
and haven't updated NEWS -- since its effect is limited,
I'm not sure it's worthy of a NEWS entry.

Before:

    $ bison --print-datadir > /dev/full
    $ bison --print-datadir >&-
    $

After:

    $ bison --print-datadir > /dev/full
    bison: write error: No space left on device
    [Exit 1]
    $ bison --print-datadir >&-
    bison: write error: Bad file descriptor
    [Exit 1]
    $

From 581dfdb383842aced19443db6f910fbd02736e23 Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering< at >redhat.com>
Date: Sun, 29 Jan 2012 12:50:32 +0100
Subject: [PATCH] do not ignore errors like ENOSPC,EIO when writing to stdout

Standard output was never explicitly closed, so we could not
detect failure.  Thus, bison would ignore the errors of writing
to a full file system and getting an I/O error on write, but only
for standard output, e.g., for --print-localedir, --print-datadir,
--help and some verbose output.
Now, "bison --print-datadir > /dev/full" reports the write failure:
bison: write error: No space left on device
Before, it would exit 0 with no diagnostic, implying success.
This is not an issue for "--output=-" or the other FILE-accepting
command-line options, because unlike most other GNU programs,
an output file argument of "-" is treated as the literal "./-",
rather than standard output.
* bootstrap.conf (gnulib_modules): Add closeout.
* src/main.c: Include "closeout.h".
Use atexit to ensure we close stdout.
* .gitignore: Ignore new files pulled in via gnulib-tool.
---
 bootstrap.conf |    2 +-
 lib/.gitignore |    6 ++++++
 m4/.gitignore  |    3 +++
 src/main.c     |    3 +++
 4 files changed, 13 insertions(+), 1 deletions(-)

diff --git a/bootstrap.conf b/bootstrap.conf
index 599b778..757111c 100644
--- a/bootstrap.conf
+++ b/bootstrap.conf
< at >< at > -17,7 +17,7 < at >< at >

 # gnulib modules used by this package.
 gnulib_modules='
-  announce-gen argmatch assert calloc-posix close config-h c-strcase
+  announce-gen argmatch assert calloc-posix close closeout config-h c-strcase
   configmake dirname error extensions fdl fopen-safer gendocs getopt-gnu
   gettext git-version-gen gitlog-to-changelog
   gpl-3.0 hash inttypes isnan javacomp-script
diff --git a/lib/.gitignore b/lib/.gitignore
index 54edced..365b84d 100644
--- a/lib/.gitignore
+++ b/lib/.gitignore
< at >< at > -259,3 +259,9 < at >< at >
 /strerror-override.h
 /strerror_r.c
 /xalloc-oversized.h
+/close-stream.c
+/close-stream.h
+/closeout.c
+/closeout.h
+/fpending.c
+/fpending.h
diff --git a/m4/.gitignore b/m4/.gitignore
index c50ed39..901724d 100644
--- a/m4/.gitignore
+++ b/m4/.gitignore
< at >< at > -166,3 +166,6 < at >< at >
 /raise.m4
 /ssize_t.m4
 /strerror_r.m4
+/close-stream.m4
+/closeout.m4
+/fpending.m4
diff --git a/src/main.c b/src/main.c
index e777ca0..3dbac3a 100644
--- a/src/main.c
+++ b/src/main.c
< at >< at > -28,6 +28,7 < at >< at >
 #include <timevar.h>

 #include "LR0.h"
+#include "closeout.h"
 #include "complain.h"
 #include "conflicts.h"
 #include "derives.h"
< at >< at > -62,6 +63,8 < at >< at > main (int argc, char *argv[])
   (void) bindtextdomain ("bison-runtime", LOCALEDIR);
   (void) textdomain (PACKAGE);

+  atexit (close_stdout);
+
   uniqstrs_new ();
   muscle_init ();

--
1.7.9.1.g63eb

[GNU Bison 2.5] testsuite: 247 failed on Solaris 10 on SPARC(sun4v)

Tomohiro Suzuki — 2012-01-24T06:50:19

Hi, I'm getting a "make check" failure on bison-2.5.
bison-2.4.3 doesn't have the problem.

OS:Solaris 10 on SPARC(sun4v)

I attach the testsuite.log and config.log.

-----
# uname -a
SunOS moore 5.10 Generic_147440-04 sun4v sparc SUNW,SPARC-Enterprise-T1000

# /usr/local/bin/gcc --version
gcc (GCC) 4.2.4
Copyright (C) 2007 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

# make check
.....
Regression tests.

223: Trivial grammars                                ok
224: YYSTYPE typedef                                 ok
225: Early token definitions with --yacc             ok
226: Early token definitions without --yacc          ok
227: Braces parsing                                  ok
228: Duplicate string                                ok
229: Rule Line Numbers                               ok
230: Mixing %token styles                            ok
231: Invalid inputs                                  ok
232: Invalid inputs with {}                          ok
233: Token definitions                               ok
234: Characters Escapes                              ok
235: Web2c Report                                    ok
236: Web2c Actions                                   ok
237: Dancer                                          ok
238: Dancer %glr-parser                              ok
239: Dancer %skeleton "lalr1.cc"                     ok
240: Expecting two tokens                            ok
241: Expecting two tokens %glr-parser                ok
242: Expecting two tokens %skeleton "lalr1.cc"       ok
243: Braced code in declaration in rules section     ok
244: String alias declared after use                 ok
245: Extra lookahead sets in report                  ok
246: Token number in precedence declaration          ok
247: parse-gram.y: LALR = IELR                       FAILED (regression.at:1272)
248: %error-verbose and YYSTACK_USE_ALLOCA           ok
249: %error-verbose overflow                         ok
250: LAC: Exploratory stack                          ok
251: LAC: Memory exhaustion                          ok

C++ Features.

252: Doxygen Public Documentation                    skipped (c++.at:102)
253: Doxygen Private Documentation                   skipped (c++.at:103)
254: Relative namespace references                   ok
255: Absolute namespace references                   ok
256: Syntactically invalid namespace references      ok

Java Calculator.

257: Calculator                                      ok
258: Calculator %error-verbose                       ok
259: Calculator %locations                           ok
260: Calculator %error-verbose %locations            ok
261: Calculator %lex-param { InputStream is }        ok
262: Calculator %error-verbose %lex-param { InputStream is }  ok
263: Calculator %locations %lex-param { InputStream is }  ok
264: Calculator %error-verbose %locations %lex-param { InputStream is }  ok

Java Parameters.

265: Java parser class and package names             ok
266: Java parser class modifiers                     ok
267: Java parser class extends and implements        ok
268: Java %parse-param and %lex-param                ok
269: Java throws specifications                      ok
270: Java stype, position_class and location_class   ok

C++ Type Syntax (GLR).

271: GLR: Resolve ambiguity, impure, no locations    ok
272: GLR: Resolve ambiguity, impure, locations       ok
273: GLR: Resolve ambiguity, pure, no locations      ok
274: GLR: Resolve ambiguity, pure, locations         ok
275: GLR: Merge conflicting parses, impure, no locations ok
276: GLR: Merge conflicting parses, impure, locations ok
277: GLR: Merge conflicting parses, pure, no locations ok
278: GLR: Merge conflicting parses, pure, locations  ok
279: GLR: Verbose messages, resolve ambiguity, impure, no locations ok

GLR Regression Tests

280: Badly Collapsed GLR States                      ok
281: Improper handling of embedded actions and dollar(-N) in GLR parsers ok
282: Improper merging of GLR delayed action sets     ok
283: Duplicate representation of merged trees        ok
284: User destructor for unresolved GLR semantic value ok
285: User destructor after an error during a split parse ok
286: Duplicated user destructor for lookahead        ok
287: Incorrectly initialized location for empty right-hand side in GLR ok
288: No users destructors if stack 0 deleted         ok
289: Corrupted semantic options if user action cuts parse ok
290: Undesirable destructors if user action cuts parse ok
291: Leaked semantic values if user action cuts parse ok
292: Incorrect lookahead during deterministic GLR    ok
293: Incorrect lookahead during nondeterministic GLR ok
294: Leaked semantic values when reporting ambiguity ok
295: Leaked lookahead after nondeterministic parse syntax error ok
296: Uninitialized location when reporting ambiguity ok
297: Missed %merge type warnings when LHS type is declared later ok

Push Parsing Tests

298: Memory Leak for Early Deletion                  ok
299: Multiple impure instances                       ok
300: Unsupported Skeletons                           ok

## ------------- ##
## Test results. ##
## ------------- ##

ERROR: 298 tests were run,
1 failed unexpectedly.
2 tests were skipped.
## -------------------------- ##
## testsuite.log was created. ##
## -------------------------- ##

Please send `tests/testsuite.log' and all information you think might help:

    To: <bug-bison< at >gnu.org>
    Subject: [GNU Bison 2.5] testsuite: 247 failed

You may investigate any problem if you feel able to do so, in which
case the test suite provides a good starting point.  Its output may
be found below `tests/testsuite.dir'.

make[3]: *** [check-local] Error 1
make[3]: Leaving directory `/usr/local/src/bison-2.5/tests'
make[2]: *** [check-am] Error 2
make[2]: Leaving directory `/usr/local/src/bison-2.5/tests'
make[1]: *** [check-recursive] Error 1
make[1]: Leaving directory `/usr/local/src/bison-2.5'
make: *** [check] Error 2
-----