> Le 17 janv. 2015 à 21:21, John Horigan <j...@glyphic.com> a écrit :
>
> Hi Akim,
Hi John
Thanks for the report! Let's keep this public: cc'd to bug-bison.
> I tried bison 3.0.3 on my OSX Yosemite system with XCode 6.1.1
>
> I ran ./configure && make && make check
>
> Test 203 failed because it declared a loop variable inside a for loop.
I don't see that. Can you pinpoint the problem?
> So I ran make distclean && ./configure CC=c99 && make && make check
>
> During make there were four warnings:
> src/symtab.h:76:14: warning: ISO C forbids forward references to 'enum' types
> [-Wpedantic]
ok.
> src/graphviz.c:116:35: warning: positional arguments are not supported by ISO
> C [-Wformat-non-iso]
ok.
> lib/printf-frexp.c:72:3: warning: use of GNU statement expression extension
> [-Wgnu-statement-expression]
This is from gnulib, I'm sure it's used only when it's safe to use.
> lib/abitset.c:789:24: warning: using extended field designator is an
> extension [-Wextended-offsetof]
It's been ok for years, so I'll ignore that warning.
> Test 203 still fails. But now I don't understand why. I have attached
> testsuite.log
The relevant bit is:
./headers.at:262: $CXX $CXXFLAGS $CPPFLAGS -c -o cxx-only.o cxx-only.cc
stderr:
g++: warning: couldn't understand kern.osversion '14.0.0
stdout:
./headers.at:262: $CXX $CXXFLAGS $CPPFLAGS $LDFLAGS c-only.o cxx-only.o -o
c-and-cxx ||
exit 77
--- /dev/null 2015-01-17 11:45:15.000000000 -0800
+++ /Users/john/Projects/bison-3.0.3/tests/testsuite.dir/at-groups/203/stderr
2015-01-17 11:45:18.000000000 -0800
@@ -0,0 +1 @@
+g++: warning: couldn't understand kern.osversion '14.0.0
stdout:
203. headers.at:112: 203. Several parsers (headers.at:112): FAILED
(headers.at:262)
Maybe you need to reinstall g++ after you upgraded to Yosemite?
http://stackoverflow.com/questions/26486163/gcc-4-9-1-in-os-x-yosemite-gcc-warning-couldn-t-understand-kern-osversion-1
I'll install this:
commit 6ee80d5417e1a7493d69f55305da0941bb40e105
Author: Akim Demaille <a...@lrde.epita.fr>
Date: Sun Jan 18 15:18:38 2015 +0100
build: fix some warnings
Reported by John Horigan.
* src/graphviz.c, src/symtab.h: Address compiler warnings.
diff --git a/THANKS b/THANKS
index 875ac96..26bea0a 100644
--- a/THANKS
+++ b/THANKS
@@ -66,6 +66,7 @@ Jim Kent jk...@arch.sel.sony.com
Jim Meyering j...@meyering.net
Joel E. Denny joelde...@joeldenny.org
Johan van Selst joh...@stack.nl
+John Horigan j...@glyphic.com
Jonathan Fabrizio jonathan.fabri...@lrde.epita.fr
Jonathan Nieder jrnie...@gmail.com
Juan Manuel Guerrero juan.guerr...@gmx.de
diff --git a/src/graphviz.c b/src/graphviz.c
index e7611b0..663cf37 100644
--- a/src/graphviz.c
+++ b/src/graphviz.c
@@ -113,8 +113,8 @@ conclude_red (struct obstack *out, int source, rule_number
ruleno,
with n the source state and m the rule number. This is because we
don't want all the reductions bearing a same rule number to point to
the same state, since that is not the desired format. */
- fprintf (fout, " %1$d -> \"%1$dR%2$d%3$s\" [",
- source, ruleno, ed);
+ fprintf (fout, " %d -> \"%dR%d%s\" [",
+ source, source, ruleno, ed);
/* (The lookahead tokens have been added to the beginning of the
obstack, in the caller function.) */
diff --git a/src/symtab.h b/src/symtab.h
index aa6d65b..f92407f 100644
--- a/src/symtab.h
+++ b/src/symtab.h
@@ -73,12 +73,12 @@ typedef enum
declared,
} status;
-typedef enum code_props_type code_props_type;
enum code_props_type
{
destructor = 0,
printer = 1,
};
+typedef enum code_props_type code_props_type;
enum { CODE_PROPS_SIZE = 2 };
