bug-grep
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [bug #29325] Undefined Symbol 'rpl_wctob'


From: Jim Meyering
Subject: Re: [bug #29325] Undefined Symbol 'rpl_wctob'
Date: Thu, 25 Mar 2010 11:41:32 +0100

Dagobert Michelsen wrote:
>> In case you'd like to test, here's a copy of the latest:
>>
>>  http://meyering.net/grep/tmp/grep-2.6.7-64540.tar.xz
>
> This works, I get an almost-clean compile with the following warnings:

Thanks for testing that.

>>   CC       main.o
>> "main.c", line 1628: warning: argument #2 is incompatible with
>> prototype:
>>         prototype: pointer to pointer to char : "../lib/getopt.h",
>> line 241
>>         argument : pointer to const pointer to char
>>   AR       libgrep.a
>>   CC       grep.o
>> "grep.c", line 10: warning: void function cannot return value
>> "grep.c", line 16: warning: void function cannot return value
>> "grep.c", line 22: warning: void function cannot return value
>> "egrep.c", line 7: warning: void function cannot return value

These are easily addressed.  Patch below.

> I used -features=extensions as suggested in
>    <http://savannah.gnu.org/bugs/?24813>
>
> However, the test char-class-multibyte is failing:
...
>> FAIL: char-class-multibyte (exit: 1)
>> ====================================
>>
>> out1-none exp1 differ: char 2, line 1
>> out2-none exp2 differ: char 2, line 1

You can help the test suite produce more/better
information if you install gnu diff and rerun "make check".
When a diff program is available that supports "-u", it uses
that rather than "cmp", which prints what you see above.

>> SKIP: euc-mb (exit: 77)
>> =======================
>>
>> ????
>> euc-mb: skipped test: EUC-JP locale seems not to work
>>
>> SKIP: fmbtest.sh (exit: 77)
>> ===========================
>
> Is this a problem with the testsuite or a real problem?

I doubt it's a testsuite problem, but that is a possibility.

It may be due to a bug in grep, or in gnulib's emulation
of some missing-in-solaris-8 function, or a bug in Solaris 8.
Hard to tell without a debugger.
Solaris 8 is old enough that it may well be a problem
with their locale support.

>From 480565bcb5a7ad7bc9b32cc8d53d1d2413f06cfb Mon Sep 17 00:00:00 2001
From: Jim Meyering <address@hidden>
Date: Thu, 25 Mar 2010 11:36:50 +0100
Subject: [PATCH] build: avoid warnings about unnecessary use of "return"

* src/grep.c (Gcompile, Ecompile, Acompile): Do not use "return X",
not even when X is a function returning void, in a function
returning void.  This avoids warnings from Sun Studio 11,
reported by Dagobert Michelsen.
* src/egrep.c (Ecompile): Likewise.
---
 src/egrep.c |    2 +-
 src/grep.c  |   12 ++++++------
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/src/egrep.c b/src/egrep.c
index 80595fe..f6a1d59 100644
--- a/src/egrep.c
+++ b/src/egrep.c
@@ -4,7 +4,7 @@
 static void
 Ecompile (char const *pattern, size_t size)
 {
-  return GEAcompile (pattern, size, RE_SYNTAX_POSIX_EGREP | 
RE_NO_EMPTY_RANGES);
+  GEAcompile (pattern, size, RE_SYNTAX_POSIX_EGREP | RE_NO_EMPTY_RANGES);
 }

 struct matcher const matchers[] = {
diff --git a/src/grep.c b/src/grep.c
index 8ef4968..1a4166c 100644
--- a/src/grep.c
+++ b/src/grep.c
@@ -4,22 +4,22 @@
 static void
 Gcompile (char const *pattern, size_t size)
 {
-  return GEAcompile (pattern, size,
-                    RE_SYNTAX_GREP
-                    | RE_HAT_LISTS_NOT_NEWLINE
-                    | RE_NO_EMPTY_RANGES);
+  GEAcompile (pattern, size,
+             RE_SYNTAX_GREP
+             | RE_HAT_LISTS_NOT_NEWLINE
+             | RE_NO_EMPTY_RANGES);
 }

 static void
 Ecompile (char const *pattern, size_t size)
 {
-  return GEAcompile (pattern, size, RE_SYNTAX_POSIX_EGREP | 
RE_NO_EMPTY_RANGES);
+  GEAcompile (pattern, size, RE_SYNTAX_POSIX_EGREP | RE_NO_EMPTY_RANGES);
 }

 static void
 Acompile (char const *pattern, size_t size)
 {
-  return GEAcompile (pattern, size, RE_SYNTAX_AWK);
+  GEAcompile (pattern, size, RE_SYNTAX_AWK);
 }

 struct matcher const matchers[] = {
--
1.7.0.3.435.g097f4




reply via email to

[Prev in Thread] Current Thread [Next in Thread]