Re: [PATCH 0/2] getaddrinfo: Add AI_NUMERICSERV and AI_NUMERICHOST flag support

: expected > EAI_NONAME, got -8 > Test case pass=3, host=google.org, service=ldap failed: expected EAI_NONAME, > got -8 > FAIL test-getaddrinfo (exit status: 4) > > Where -8 is EAI_SERVICE. This is on a Debian 12 system. > > This failure makes no sense to me. Any idea? &g

Re: [PATCH 0/2] getaddrinfo: Add AI_NUMERICSERV and AI_NUMERICHOST flag support

Bruno Haible wrote: > KO Myung-Hun wrote: >> S["REPLACE_GETADDRINFO"]="1" >> S["HAVE_DECL_GETADDRINFO"]="0" >> D["HAVE_GETADDRINFO"]=" 0" >> D["HAVE_DECL_GETADDRINFO"]=" 0" > &g

Re: [PATCH 0/2] getaddrinfo: Add AI_NUMERICSERV and AI_NUMERICHOST flag support

Bruno Haible wrote: > KO Myung-Hun wrote: >> I've pulled the latest gnulib. And I've got the following error when >> testing getaddrinfo module: >> >> ./../gllib/netdb.h:745:1: error: 'getaddrinfo' was not declared in this >> scop

Re: [PATCH 0/2] getaddrinfo: Add AI_NUMERICSERV and AI_NUMERICHOST flag support

~~ ./../gllib/netdb.h:757:1: note: 'gnulib::freeaddrinfo' declared here 757 | _GL_CXXALIAS_RPL (freeaddrinfo, void, (struct addrinfo *ai)); | ^~~~ make[4]: *** [Makefile:3189: test-netdb-h-c++.o] Error 1 - -- KO Myung-Hun Korean OS/2 User Community : https://www.os2.kr/

Re: [PATCH 0/2] getaddrinfo: Add AI_NUMERICSERV and AI_NUMERICHOST flag support

Hi/2. Bruno Haible wrote: > KO Myung-Hun wrote: >> These are the patches to support AI_NUMERICSERV and AI_NUMERICHOST flag >> to getaddrinfo(). >> >> Review, please... >> >> [PATCH 1/2] getaddrinfo: Add AI_NUMERICSERV to the supported flags >> [PA

Re: [PATCH 0/2] getaddrinfo: Add AI_NUMERICSERV and AI_NUMERICHOST flag support

Hi/2. Bruno Haible wrote: > KO Myung-Hun wrote: >> These are the patches to support AI_NUMERICSERV and AI_NUMERICHOST flag >> to getaddrinfo(). >> >> Review, please... >> >> [PATCH 1/2] getaddrinfo: Add AI_NUMERICSERV to the supported flags >> [PA

Re: [PATCH] popen: set gl_cv_func_popen_works to no on OS/2

Bruno Haible wrote: > KO Myung-Hun wrote: >>> If you set gl_cv_func_popen_works to 'no', it will activate the workaround >>> in lib/popen.c. Does this workaround work? Does 'test-[p]open' succeed? >> >> Do you mean 'test-popen'?

Re: [PATCH] popen: set gl_cv_func_popen_works to no on OS/2

.. (cached) no > > Bruno > If the test is run, 'configure' is blocked until pressing Ctrl-C or Ctrl-Break. Or do you mean modification of the test source codes to return an error if OS/2? -- KO Myung-Hun Korean OS/2 User Community : https://www.os2.kr/

[PATCH 2/2] getaddrinfo: Add AI_NUMERICHOST flag support

* lib/getaddrinfo.c (getaddrinfo): Check AI_NUMERICHOST flag, too. * lib/netdb.in.h: Define AI_NUMERICHOST if not defined yet. --- lib/getaddrinfo.c | 4 +++- lib/netdb.in.h| 2 +- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/lib/getaddrinfo.c b/lib/getaddrinfo.c index 542add

[PATCH 1/2] getaddrinfo: Add AI_NUMERICSERV to the supported flags

* lib/getaddrinfo.c (getaddrinfo): Check AI_NUMERICSERV flag, too. --- lib/getaddrinfo.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/getaddrinfo.c b/lib/getaddrinfo.c index cd045c5121..542addc670 100644 --- a/lib/getaddrinfo.c +++ b/lib/getaddrinfo.c @@ -213,7 +213,7 @@

[PATCH 0/2] getaddrinfo: Add AI_NUMERICSERV and AI_NUMERICHOST flag support

Hi/2. These are the patches to support AI_NUMERICSERV and AI_NUMERICHOST flag to getaddrinfo(). Review, please... [PATCH 1/2] getaddrinfo: Add AI_NUMERICSERV to the supported flags [PATCH 2/2] getaddrinfo: Add AI_NUMERICHOST flag support

[PATCH] popen: set gl_cv_func_popen_works to no on OS/2

On OS/2, popen() does not work with closed stdios. And configure hangs on with the test. * m4/popen.m4 (gl_FUNC_POPEN): Set gl_cv_func_popen_works to no on OS/2. --- m4/popen.m4 | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/m4/popen.m4 b/m4/popen.m4 index a68c1e606f..b301

Re: [PATCH v2 1/2] wait-process: Use waitpid() from LIBCx on OS/2 kLIBC if possible

Bruno Haible wrote: > KO Myung-Hun wrote: >>>> * lib/wait-process.c (klibc_waitpid) [kLIBC]: New function. >>>> (waitpid) [kLIBC]: Define it to klibc_waitpid. >>> >>> The code looks good, but lacks comments. In the ChangeLog or git commit >>>

Re: [PATCH v2 1/2] wait-process: Use waitpid() from LIBCx on OS/2 kLIBC if possible

Hi/2. Bruno Haible wrote: > KO Myung-Hun wrote: >> LIBCx is the kLIBC extension library which provides some enhanced >> features to OS/2 kLIBC. >> >> It provides its own waitpid() to support its spawn2(). Use it if >> possible. >> >> * lib/wait-pr

[PATCH v2 2/2] spawn-pipe: Support a directory argument of create_pipe() on OS/2 kLIBC

LIBCx, which is a kLIBC extension library, provides spawn2() to support this feature. If LIBCx is available, then support a directory argument using spawn2() of LIBCx. * lib/os2-spawn.c (spawnpvech): New function. * lib/os2-spawn.h (spawnpvech): New declaration. * lib/spawn-pipe.c (create_pipe) [k

[PATCH v2 1/2] wait-process: Use waitpid() from LIBCx on OS/2 kLIBC if possible

LIBCx is the kLIBC extension library which provides some enhanced features to OS/2 kLIBC. It provides its own waitpid() to support its spawn2(). Use it if possible. * lib/wait-process.c (klibc_waitpid) [kLIBC]: New function. (waitpid) [kLIBC]: Define it to klibc_waitpid. --- lib/wait-process.c |

[PATCH v2 0/2] Add support of a dir arg of create_pipe() on OS/2 kLIBC

Hi/2. These are v2 patches to support a directory change of create_pipe() on OS/2 kLIBC. Review, please... [PATCH v2 1/2] wait-process: Use waitpid() from LIBCx on OS/2 kLIBC [PATCH v2 2/2] spawn-pipe: Support a directory argument of

Re: [PATCH 0/2] Add support of a dir arg of create_pipe() on OS/2 kLIBC

Hi/2. Bruno Haible wrote: > Hi, > > KO Myung-Hun wrote: >> These patches allow to use a directory argument of create_pipe() on >> OS/2 kLIBC. > > Four things I don't understand: > > - You still don't have posix_spawn support for this platform? >

[PATCH 0/2] Add support of a dir arg of create_pipe() on OS/2 kLIBC

Hi/2. These patches allow to use a directory argument of create_pipe() on OS/2 kLIBC. Review, please... [PATCH 1/2] wait, waitid, waitpid: Use ones from LIBCx on OS/2 kLIBC [PATCH 2/2] spawn-pipe: Support a directory argument of create_pipe()

[PATCH 1/2] wait, waitid, waitpid: Use ones from LIBCx on OS/2 kLIBC if possible

LIBCx is the kLIBC extension library which provides some enhanced features to OS/2 kLIBC. It provides its own wait(), waitid() and waitpid() to support its spawn2(). Use them if possible. * lib/sys_wait.in.h (rpl_wait, rpl_waitid, rpl_waitpid): New declaration. * lib/wait.c: New file. * lib/wait

[PATCH 2/2] spawn-pipe: Support a directory argument of create_pipe() on OS/2 kLIBC

LIBCx, which is a kLIBC extension library, provides spawn2() to support this feature. If LIBCx is available, then support a directory argument using spawn2() of LIBCx. * lib/spawn-pipe.c (create_pipe) [kLIBC]: Use spawn2() of LIBCx to support a directory argument if available. * m4/spawn-pipe.m4 (

Re: [PATCH] init.sh: Consider OS/2-style path

runo Haible > > test-framework-sh: Improve portability to native Windows and OS/2. > Reported by KO Myung-Hun in > <https://lists.gnu.org/archive/html/bug-gnulib/2024-01/msg00037.html>. > * tests/init.sh (setup_): Test for an absolute directory name li

[PATCH] init.sh: Consider OS/2-style path

tests/init.sh (setup_): Use $PATH_SEPARATOR as a path separator, and consider an OS/2-style absolute path. --- tests/init.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/init.sh b/tests/init.sh index aef5eea808..8b5dc2b23e 100644 --- a/tests/init.sh +++ b/tests/i

[PATCH] init.sh: Consider OS/2-style path

tests/init.sh (setup_): Use $PATH_SEPARATOR as a path separator, and consider an OS/2-style absolute path. --- tests/init.sh | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/init.sh b/tests/init.sh index 0494097e9f..2a88865972 100644 --- a/tests/init.sh +++ b/tests/i

Re: [PATCH] fdopendir: Use Windows code path on OS/2 kLIBC

Hi/2. Bruno Haible wrote: > KO Myung-Hun wrote: >> * lib/closedir.c (closedir): Use Windows code path. >> * lib/dirent.in.h (_gl_register_dirp_fd, _gl_unregister_dirp_fd): >> Remove. >> * lib/dirfd.c (_gl_register_dirp_fd, _gl_unregister_dirp_fd): Remove. >&g

[PATCH] stdlib: Cast putenv() on OS/2 kLIBC

On OS/2 kLIBC, the first parameter of putenv () is `const char *string' not `char *string'. * lib/stdlib.in.h (putenv) [kLIBC]: Cast with _GL_CXXALIAS_SYS_CAST(). --- lib/stdlib.in.h | 8 1 file changed, 8 insertions(+) diff --git a/lib/stdlib.in.h b/lib/stdlib.in.h index b9838c7d5e..b2

[PATCH] fdopendir: Use Windows code path on OS/2 kLIBC

* lib/closedir.c (closedir): Use Windows code path. * lib/dirent.in.h (_gl_register_dirp_fd, _gl_unregister_dirp_fd): Remove. * lib/dirfd.c (_gl_register_dirp_fd, _gl_unregister_dirp_fd): Remove. (dirfd): Use Windows code path. * lib/fdopendir.c (fdopendir): Use Windows code path. * lib/opendir.c (

Re: fdopendir: Fix fd leak and test failure on native Windows

Hi/2. Bruno Haible wrote: > KO Myung-Hun wrote: >>> dirfd is specified at >>> https://pubs.opengroup.org/onlinepubs/9699919799/functions/dirfd.html >>> Which says that if "The dirp argument does not refer to a valid directory >>> stream" &g

Re: fdopendir: Fix fd leak and test failure on native Windows

Bruno Haible wrote: > KO Myung-Hun wrote: >>> @@ -78,11 +82,17 @@ _gl_unregister_dirp_fd (int fd) >>> int >>> dirfd (DIR *dir_p) >>> { >>> +#if GNULIB_defined_DIR >>> + int fd = dir_p->fd_to_close; >> >> dirfd() allo

Re: fdopendir: Fix fd leak and test failure on native Windows

a file descriptor gets allocated > and not closed. > > This patch fixes it, by adding a 'fd_to_close' member. > > For kLIBC, KO Myung-Hun had added a similar mechanism, by keeping a list > of assocations between DIR objects and file descriptors. I believe > overriding

Re: [PATCH 4/7] stdlib: putenv() needs a cast on OS/2 kLIBC

by the maintainer of kLIBC. See details here: http://trac.netlabs.org/libc/ticket/387#comment:2 Would you mind applying this patch ? -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under OS/2 Warp 4 for Korean with FixPak #15 In VirtualBox v6.1.10 on Intel Core i7-3615QM 2.30GHz with 8GB RAM Korean OS/2 User Community : http://www.os2.kr/

Re: [PATCH 4/7] stdlib: putenv() needs a cast on OS/2 kLIBC

har *string' is more proper, because putenv() does not modify the content of `string'. Thanks as always! -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under OS/2 Warp 4 for Korean with FixPak #15 In VirtualBox v6.1.10 on Intel Core i7-3615QM 2.30GHz with 8GB RAM Korean OS/2 User Community : http://www.os2.kr/

Re: [PATCH 1/7] spawn: Use special invocation for on OS/2 kLIBC

Hi, Bruno. Bruno Haible wrote: > Hi KO, > > Thanks, I applied patches 1, 2, 3, 5, 6, 7, slightly modified. Some of the > things that you put into the git commit message actually better belong in > the code, as a comment. > Ok, I'll do so later. Thanks for your effort!

Re: [PATCH 5/7] zerosize-ptr: Fix compilation on OS/2 kLIBC

not > the right test here. And since you write "OS/2 kLIBC has > and mprotect() but not mmap()", defined __KLIBC__ implies !HAVE_MMAP. > I saw that. So I tested HAVE_MMAP only on kLIBC. Anyway I agree that just defined __KLIBC__ is enough. Thanks! -- KO Myung-Hun Using Mo

[PATCH 2/7] yield: Fix linkage to threadlib on OS/2 kLIBC

* lib/sched_yield.c (sched_yield) [OS/2 kLIBC]: New. * modules/yield (Depends-on): Add sched_yield. --- lib/sched_yield.c | 13 + modules/yield | 1 + 2 files changed, 14 insertions(+) diff --git a/lib/sched_yield.c b/lib/sched_yield.c index 0a0ff3c9c..880711405 100644 --- a/lib/

[PATCH 7/7] spawn-pipe: Fix SIGSEGV on OS/2 kLIBC

* lib/spawn-pipe.c [OS/2 kLIBC]: Check directory correctly if it means a current directory. --- lib/spawn-pipe.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/spawn-pipe.c b/lib/spawn-pipe.c index 641e9216f..0fe86ffe2 100644 --- a/lib/spawn-pipe.c +++ b/lib/spawn-pipe.c @

[PATCH 5/7] zerosize-ptr: Fix compilation on OS/2 kLIBC

OS/2 kLIBC has and mprotect() but not mmap(). As a result, it's not enough to test with HAVE_SYS_MMAN_H and HAVE_MPROTECT. It's necessary to test with HAVE_MMAP on OS/2 kLIBC. * tests/zerosize-ptr.h [OS/2 kLIBC]: Test with HAVE_MMAP, too. (zerosize_ptr) [OS/2 kLIBC]: Likewise. --- tests/zerosize

[PATCH 6/7] dirent: Redefine dirfd macro as a rpl function always

On OS/2 kLIBC, dirfd is defined as a macro not a function. Because of this, dirfd is not redefined as rpl function in C++ mode. As a result, the following error occurs: - ./../gllib/dirent.h:689:1: error: 'dirfd' was not declared in this scope; did you mean 'gnulib::dirfd'? 689 | _GL_CXXALI

[PATCH 4/7] stdlib: putenv() needs a cast on OS/2 kLIBC

On OS/2 kLIBC, the first parameter of putenv () is `const char *string' not `char *string'. * lib/stdlib.in.h: Use _GL_CXXALIAS_SYS_CAST() for putenv(). --- lib/stdlib.in.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/lib/stdlib.in.h b/lib/stdlib.in.h index 49fc44e14..a0

[PATCH 3/7] stddef: Fix compilation for max_align_t on OS/2 kLIBC

On OS/2 kLIBC, `max_align_t' is defined only in C++ mode. * lib/stddef.in.h [OS/2 kLIBC]: Include in C++ mode. --- lib/stddef.in.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/lib/stddef.in.h b/lib/stddef.in.h index 77147e9bc..c415cbb85 100644 --- a/lib/stddef.in.h +++ b

[PATCH 1/7] spawn: Use special invocation for on OS/2 kLIBC

On OS/2 kLIBC, includes . Then -> -> -> are included by GNULIB. In this situation, `struct sched_param' is not yet defined. * lib/spawn.in.h: Use special invocation to include properly on OS/2 kLIBC. --- lib/spawn.in.h | 22 -- 1 file changed, 20 insertions(+), 2 deletio

Re: refactor common code of 'execute' and 'spawn-pipe'

006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037*?" > -# define SHELL_SPACE_CHARS " > \001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037" > -#else > -# define SHELL_SPECIAL_CHARS "" > -# define SHELL_SPACE_CHARS "" > -#endif > +#define SHELL_SPECIAL_CHARS "\"\\ > \001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037*?" > +#define SHELL_SPACE_CHARS " > \001\002\003\004\005\006\007\010\011\012\013\014\015\016\017\020\021\022\023\024\025\026\027\030\031\032\033\034\035\036\037" > char ** > prepare_spawn (char **argv) > { > diff --git a/modules/spawn-pipe b/modules/spawn-pipe > index 2c62586..7845afa 100644 > --- a/modules/spawn-pipe > +++ b/modules/spawn-pipe > @@ -4,6 +4,8 @@ Creation of subprocesses, communicating via pipes. > Files: > lib/spawn-pipe.h > lib/spawn-pipe.c > +lib/os2-spawn.h > +lib/os2-spawn.c > m4/spawn-pipe.m4 > > Depends-on: > @@ -35,6 +37,12 @@ windows-spawn > > configure.ac: > gl_SPAWN_PIPE > +AC_REQUIRE([AC_CANONICAL_HOST]) > +case "$host_os" in > + os2) s/os2/os2*/ Thanks! -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under OS/2 Warp 4 for Korean with FixPak #15 In VirtualBox v6.1.10 on Intel Core i7-3615QM 2.30GHz with 8GB RAM Korean OS/2 User Community : http://www.os2.kr/

[PATCH] config.sub: fix i*86-pc-os2-emx is not recognized as a valid OS

OS part is os2-emx not emx. * build-aux/config.sub [i*86-pc-os2-emx]: Set os to os2-emx. --- build-aux/config.sub | 4 1 file changed, 4 insertions(+) diff --git a/build-aux/config.sub b/build-aux/config.sub index 9bc49a7e9..9fb81d598 100755 --- a/build-aux/config.sub +++ b/build-aux/config

[PATCH] pthread_sigmask: fix implicit declaration of 'pthread_sigmask'

This fixes the following warning: - CC lib/pthread_sigmask.o lib/pthread_sigmask.c: In function 'rpl_pthread_sigmask': lib/pthread_sigmask.c:52:9: warning: implicit declaration of function 'pthread_sigmask'; did you mean 'rpl_pthread_sigmask'? [-Wimplicit-function-declaration]

Re: [PATCH] binary-io: do not treat set_binary_mode() on stdin/out/err as an error on OS/2

7-02/msg00058.html>, namely: > * set_binary_mode on a tty was changed in 2017 to return an error code and > now again succeeds. > * Inline and remove the __gl_setmode_check function. I found that the > indirection through this function makes the code harder to understand. > >

Re: [PATCH] binary-io: do not treat set_binary_mode() on stdin/out/err as an error on OS/2

Bruno Haible wrote: > KO Myung-Hun wrote: >> setmode(O_BINARY) on tty works well instead >> of returning -1 as previous set_binary_mode() does. > > "works well", but isn't this the case which produces staircase-shaped > output? > setmode(O_BINARY)

Re: [PATCH] binary-io: do not treat set_binary_mode() on stdin/out/err as an error on OS/2

Hi/2. Bruno Haible wrote: > KO Myung-Hun wrote: >> * lib/binary-io.c (__gl_setmode_check) [__EMX__]: Remove __EMX__ guard. >> * lib/binary-io.h (__gl_setmode_check) [__EMX__]: Remove __EMX__ guard. >> (set_binary_mode) [__EMX__]: Override mode with O_TEXT if tty. >

Re: [PATCH] binary-io: do not treat set_binary_mode() on stdin/out/err as an error on OS/2

Hi/2. Paul Eggert wrote: > On 5/24/19 4:31 AM, KO Myung-Hun wrote: >> Setting stdin/out/err to binary mode is allowed on OS/2. But it's not >> useful, because it generates stair-output hard to read. >> >> Instead, let's set them to text mode all the time. &g

[PATCH] binary-io: do not treat set_binary_mode() on stdin/out/err as an error on OS/2

Setting stdin/out/err to binary mode is allowed on OS/2. But it's not useful, because it generates stair-output hard to read. Instead, let's set them to text mode all the time. This fixes that tee always crashes at startup if stdin or stdout are not piped. * lib/binary-io.c (__gl_setmode_check)

Re: [PATCH 2/2] mknod: mknod() needs unistd.h on OS/2 kLIBC

Bruno Haible wrote: > Thanks. I installed both of your patches, the second one with some > comment tweaks. > > Apologies for the mistake that resulted in the syntax error in fcntl.c. > No problem. Thanks! -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under OS/2 Warp 4

Re: [PATCH 2/2] mknod: mknod() needs unistd.h on OS/2 kLIBC

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Eric Blake wrote: > On 1/16/19 10:35 PM, KO Myung-Hun wrote: >> On OS/2 kLIBC, mknod() is declared in unistd.h. This fixes >> 'implicit declaration of function' compilation warning. >> >> * lib/mknod.c: Inclu

[PATCH 0/2] OS/2 patches

Hi/2. These are the patches for OS/2. Review, please... [PATCH 1/2] fcntl: Remove mis-placed ';' [PATCH 2/2] mknod: mknod() needs unistd.h on OS/2 kLIBC

[PATCH 1/2] fcntl: Remove mis-placed ';'

This fixes the compilation error. * lib/fcntl.c (klibc_fcntl): Remove mis-placed ';'. --- lib/fcntl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/fcntl.c b/lib/fcntl.c index f602fad62..51f62ef78 100644 --- a/lib/fcntl.c +++ b/lib/fcntl.c @@ -545,7 +545,7 @@ rpl_fcntl_D

[PATCH 2/2] mknod: mknod() needs unistd.h on OS/2 kLIBC

On OS/2 kLIBC, mknod() is declared in unistd.h. This fixes 'implicit declaration of function' compilation warning. * lib/mknod.c: Include unistd.h. --- lib/mknod.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/lib/mknod.c b/lib/mknod.c index dbbfe29c4..1d4b490ca 100644 --- a/lib/mknod.c +

Re: [PATCH] wcwidth: check a macro version of wcwidth () as well

Hi/2. Bruno Haible wrote: > Hi, > > KO Myung-Hun wrote: >>> With your patch, does the following command sequence pass its tests? >>> >>> $ ./gnulib-tool --create-testdir --dir=testdir1 --single-configure >>> --with-c++-tests wcwidth >>

Re: test-init.sh failure on OS/2

Bruno Haible wrote: > KO Myung-Hun wrote: >>> Does the init.sh test failure go away with this patch? >>> >> >> 'shopt' related failure disappeared. > > OK, I've pushed the patch. > >> However, 'invalid path dir' is

Re: test-init.sh failure on OS/2

Hi/2. Bruno Haible wrote: > KO Myung-Hun wrote: >> I attach test-suite.log. > > Does the init.sh test failure go away with this patch? > 'shopt' related failure disappeared. However, 'invalid path dir' is still. I looked into that. And ':' was a

Re: [PATCH] wcwidth: check a macro version of wcwidth () as well

Hi/2. Bruno Haible wrote: > Hi, > > KO Myung-Hun wrote: >> Check the functionality of a macro version wcwidth () as well as a >> real function. >> >> This is better than undefining wcwidth on OS/2 kLIBC without functional >> check. > > With your pa

[PATCH] wcwidth: check a macro version of wcwidth () as well

Check the functionality of a macro version wcwidth () as well as a real function. This is better than undefining wcwidth on OS/2 kLIBC without functional check. * lib/wchar.in.h: Revert commit caee51. * m4/wcwidth.m4 (gl_cv_func_wcwidth_macro): Check if wcwidth () is a macro. --- lib/wchar.in.h

[PATCH 3/4] alphasort, scandir: Port to OS/2 kLIBC

On OS/2 kLIBC, scandir() declaration is different from POSIX. As a result, alphasort() declaration is different, too. * lib/alphasort.c (alphasort): Implement according to OS/2 kLIBC declaration. * lib/scandir.c (scandir): Add declaration for OS/2 kLIBC. --- lib/alphasort.c | 10 ++ lib/s

[PATCH 1/4] sys_socket: typedef sa_family_t correctly on OS/2 kLIBC

On OS/2 kLIBC, sa_family_t is unsigned char unless TCPV40HDRS is defined. * lib/sys_socket.in.h (sa_family_t): Typedef to unsigned char on OS/2 kLIBC unless TCPV40HDRS is defined. --- lib/sys_socket.in.h | 5 + 1 file changed, 5 insertions(+) diff --git a/lib/sys_socket.in.h b/lib/sys_socket

[PATCH 4/4] scandir: Fix _D_ALLOC_NAMLEN() on OS/2 kLIBC

On OS/2 kLIBC, d_name is not the last field of struct dirent. So copying struct dirent according to the size calculated based on d_name blows the fields after d_name up. The correct way is to allocate the whole size of struct dirent. * lib/scandir.c (_D_ALLOC_NAMLEN): Consider the fields after d_

[PATCH 0/4] OS/2 patches

Hi/2. These are OS/2 patches. Review, please... [PATCH 1/4] sys_socket: typedef sa_family_t correctly on OS/2 kLIBC [PATCH 2/4] relocatable: Fix that /@unixroot prefix is not working on [PATCH 3/4] alphasort, scandir: Port to OS/2 kLIBC [PATCH 4/4] scandir: Fix _D_ALLOC_NAMLEN() on OS/2 kLIBC

[PATCH 2/4] relocatable: Fix that /@unixroot prefix is not working on OS/2 kLIBC

OS/2 kLIBC has a feature to rewrite some path components. For example, '/@unixroot' is replaced with a value of $UNIXROOT if it is. So prepending a drive letter to the path starting with '/' makes the path starting with '/@unixroot' to 'x:/@unixroot' which is unexpected. This will breaks the beha

[PATCH] sys_socket: typedef sa_family_t correctly on OS/2 kLIBC

On OS/2 kLIBC, sa_family_t is unsigned char unless TCPV40HDRS is defined. * lib/sys_socket.in.h (sa_family_t): Typedef to unsigned char on OS/2 kLIBC unless TCPV40HDRS is defined. --- lib/sys_socket.in.h | 5 + 1 file changed, 5 insertions(+) diff --git a/lib/sys_socket.in.h b/lib/sys_socket

Re: dependencies with incompatible licenses: glob

ulib/2011-09/msg00231.html > > Can we change the license of the modules 'closedir', 'opendir', 'readdir' > to 'LGPLv2+'? > > Agreements are needed: > - For closedir from Bruno, KO Myung-Hun. > - For opendir from Bruno, Eric, KO Myung-

Re: [PATCH v7 11/11] gnulib-tool: fall back into copy if symbolic link is not supported

Paul Eggert wrote: > KO Myung-Hun wrote: >> But, you said that it was enough to warn only once. > > True, and the current implementation doesn't have that nicety. However, > it's not essential to support it, and not all that important as > gnulib-tool will be

Re: [PATCH v7 11/11] gnulib-tool: fall back into copy if symbolic link is not supported

Paul Eggert wrote: > KO Myung-Hun wrote: >> Some time ago, you reviewed my patch, and advised to do so because 'ln >> -s' may succeed on some file systems, or may fail on another file >> systems. > > Thanks for reminding me; I had forgotten about that cons

[PATCH v2] openat_proc_name: fix that last '/' is overwritten on OS/2 kLIBC

* lib/openat-proc.c (openat_proc_name): Increase dirlen by 1 after copying a directory. --- lib/openat-proc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/openat-proc.c b/lib/openat-proc.c index 6fba6c2..1712340 100644 --- a/lib/openat-proc.c +++ b/lib/openat-proc.c @@ -

Re: bug#22443: new snapshot available: grep-2.22.30-e07b

orting the gnulib-defined dirfd() function. > - if test -z "${host_os##os2*}"; then > -if test $HAVE_OPENDIR = 1; then > - REPLACE_OPENDIR=1 > -fi > - fi > + case $host_os,$HAVE_OPENDIR in > +os2,1) > + REPLACE_OPENDIR=1;; > + esac > ]) Ooops... typos at the beginning... s/OPENDIR/CLOSEDIR/g And s/os2/os2*/, please... -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under OS/2 Warp 4 for Korean with FixPak #15 In VirtualBox v4.1.32 on Intel Core i7-3615QM 2.30GHz with 8GB RAM Korean OS/2 User Community : http://www.ecomstation.co.kr

Re: [PATCH v7 08/11] openat_proc_name: port to OS/2 kLIBC

Ping ? -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under OS/2 Warp 4 for Korean with FixPak #15 In VirtualBox v4.1.32 on Intel Core i7-3615QM 2.30GHz with 8GB RAM Korean OS/2 User Community : http://www.ecomstation.co.kr

Re: [PATCH v7 11/11] gnulib-tool: fall back into copy if symbolic link is not supported

o gnulib patches > pretty much as-is. At last, all my OS/2 patches have been applied in almost two years. Thanks a lot. ^^ -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under OS/2 Warp 4 for Korean with FixPak #15 In VirtualBox v4.1.32 on Intel Core i7-3615QM 2.30GHz with 8GB RAM Korean OS/2 User Community : http://www.ecomstation.co.kr

Re: [PATCH v7 08/11] openat_proc_name: port to OS/2 kLIBC

Hi/2. Not work here. I attach the fix. Paul Eggert wrote: > Thanks, I installed the attached somewhat-different patch. It also > incorporates the ChangeLog fixes from your earlier patches that I forgot > to install earlier. Please give it a try. -- KO Myung-Hun Using Mozilla SeaMon

Re: [PATCH v7 00/11] OS/2 patches

Paul Eggert wrote: > Thanks, I've installed patches 02, 03, 04, 05, 06, 07, and 09 in that > series, and have responded with comments for the other patches. Thanks a lot. I've replied updated patches to those comments. -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under

Re: [PATCH v7 11/11] gnulib-tool: fall back into copy if symbolic link is not supported

Paul Eggert wrote: > On 01/13/2016 06:23 PM, KO Myung-Hun wrote: >> + *) # SRC is relative to the directory of DEST. >> +cp_src="`echo "$2" | sed -e 's,\(^.*\)/[^/]*$,\1,'`/$1" ;; > > Can we use parameter expansion instead? Someth

Re: [PATCH v7 10/11] utimes: detect utimes() correctly on OS/2 kLIBC

Paul Eggert wrote: > On 01/13/2016 06:23 PM, KO Myung-Hun wrote: >> +@item >> +On OS/2, this funciton cannot set timestamp earlier than 1980 year in >> local >> +time. >> +@item >> +On OS/2, this function cannot set timestamp in odd seconds. > > M

Re: [PATCH v7 08/11] openat_proc_name: port to OS/2 kLIBC

Hi/2. Paul Eggert wrote: > On 01/13/2016 06:23 PM, KO Myung-Hun wrote: > >> +if (__libc_Back_ioFHToPath (fd, path, sizeof (path))) > > Don't parenthesize the argument of sizeof here. > >> +sprintf (result, "%s/%s", path, file); > >

Re: [PATCH v7 01/11] stdint: check _INTPTR_T_DECLARED before defining intptr_t and uintptr_t

tptr_t gl_intptr_t > #define uintptr_t gl_uintptr_t > +#endif > > /* 7.18.1.5. Greatest-width integer types */ > Updated. -- KO Myung-Hun Using Mozilla SeaMonkey 2.7.2 Under OS/2 Warp 4 for Korean with FixPak #15 In VirtualBox v4.1.32 on Intel Core i7-3615QM 2.30GHz

[PATCH v7 09/11] opendir, closedir, dirfd, fdopendir: port to OS/2 kLIBC

* lib/closedir.c (closedir): Unregister fd if closedir() succeeds. * lib/dirent.in.h (_gl_register_dirp_fd, _gl_unregister_dirp_fd): Declare on kLIBC. * lib/dirfd.c (struct dirp_fd_list): New. Structures to keep track of fd associated with dirp. (_gl_register_dirp_fd): New. Register fd associated w

[PATCH v7 10/11] utimes: detect utimes() correctly on OS/2 kLIBC

utimes() of OS/2 kLIBC has some limitations. 1. OS/2 itself supports a file date since 1980 year in local time. 2. OS/2 itself supports only even seconds for a file time. 3. utimes() of OS/2 kLIBC does not work on an opened file. * m4/utimes.m4: Detect utimes() correctly on OS/2 kLIBC. * doc/posi

[PATCH v7 11/11] gnulib-tool: fall back into copy if symbolic link is not supported

And warn about it only once. * gnulib-tool (first_fall_back_to_cp_warn): New. Indicator for only once warning about falling back to cp -p. (func_ln_s): New. Fall back into cp -p if ln -s fails. And warn about this only once. (func_ln): Replace ln -s with func_ln_s. --- gnulib-tool | 35 ++

[PATCH v7 08/11] openat_proc_name: port to OS/2 kLIBC

OS/2 kLIBC provides a function to retrive a path from a fd. Use it instead of /proc/self/fd. * lib/openat-proc.c (openat_proc_name): Port to OS/2 kLIBC with __libc_Back_ioFHToPath(). --- lib/openat-proc.c | 29 + 1 file changed, 29 insertions(+) diff --git a/lib/opena

[PATCH v7 00/11] OS/2 patches

Hi/2. These are v7(totally) OS/2 patches rebased in according to year change. Review, please...

[PATCH v7 03/11] pipe-filter-aux: undefine HAVE_SELECT on KLIBC

On OS/2 kLIBC, select() works only on sockets. * lib/pipe-filter-aux.h (HAVE_SELECT): Undefine on OS/2 kLIBC. --- lib/pipe-filter-aux.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/lib/pipe-filter-aux.h b/lib/pipe-filter-aux.h index f4249f4..50ad62c 100644 --- a/lib/pi

[PATCH v7 06/11] pipe_filter_ii_execute: port to OS/2 kLIBC

Pipes on kLIBC does not support O_NONBLOCK like Win32. * lib/pipe-filter-ii.c (start_wrapper, _beginthreadex, CloseHandle, WaiForSingleObject, WaitForMultipleObjects): New on OS/2 kLIBC. Reuse Win32 codes on OS/2 kLIBC. * lib/spawn-pipe.c: Reuse Win32 codes on OS/2 kLIBC. * lib/w32spawn.h: Do not

[PATCH v7 07/11] dup, dup2, fcntl: support a directory fd on OS/2 kLIBC

On OS/2 kLIBC, dup(), dup2() and fcntl() do not work on a directory fd. * lib/dup.c (dup_nothrow): New. * lib/dup2.c (klibc_dup2dirfd): New. dup2() for a directory fd. (klibc_dup2): New. * lib/fcntl.c (klibc_fcntl): New. * m4/dup.m4 (gl_FUNC_DUP): Check if dup() works on a directory fd. * m4/dup2.

[PATCH v7 02/11] binary-io: don't put fd in binary mode if it is a console on EMX

* lib/binary-io.h (SET_BINARY): Don't put fd in binary mode if it is a console on EMX. --- lib/binary-io.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/binary-io.h b/lib/binary-io.h index d5c8233..9f17c0d 100644 --- a/lib/binary-io.h +++ b/lib/binary-io.h @@ -60,7 +60,7

[PATCH v7 05/11] wchar: fix 'conflicting types' error for `__wcwidth' on OS/2 kLIBC

On OS/2 kLIBC, wcwidth is a macro to static inline function. Implementation of wcwidth in wcwidth.c causes 'conflicting types' error. * lib/wchar.in.h: Undefine wcwidth on OS/2 kLIBC. --- lib/wchar.in.h | 6 +- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/lib/wchar.in.h b/lib/

[PATCH v7 04/11] w32spawn: clear SHELL_SPECIAL_CHARS and SHELL_SPACE_CHAR on OS/2 kLIBC

spawn() on OS/2 kLIBC is not silly like one on Windows * libc/w32spawn.h (SHELL_SPECIAL_CHARS, SHELL_SPACE_CHAR): Set both to empty string on OS/2 kLIBC. --- lib/w32spawn.h | 9 +++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/lib/w32spawn.h b/lib/w32spawn.h index 7487950..

[PATCH v7 01/11] stdint: check _INTPTR_T_DECLARED before defining intptr_t and uintptr_t

OS/2 kLIBC's stdint.h defines _INTPTR_T_DECLARED and needs its own definitions of intptr_t and uintptr_t (which use int and unsigned) to avoid clashes with declarations of system functions like sbrk. * lib/stdint.in.h (intptr_t, uintptr_t): Check _INTPTR_T_DECLARED before defining them. --- lib/s

Re: [PATCH 11/11] gnulib-tool: fall back into copy if symbolic link is not supported

Updated. Process a relative path of SRC correctly. KO Myung-Hun wrote: > And warn about it only once. > > * gnulib-tool (first_fall_back_to_cp_warn): New. Indicator for only > once warning about falling back to cp -p. > (func_ln_s): New. Fall back into cp -p if ln -s fails.

[PATCH 10/11] utimes: detect utimes() correctly on OS/2 kLIBC

(-) diff --git a/ChangeLog b/ChangeLog index 1de5860..157d50f 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,18 @@ 2015-11-07 KO Myung-Hun +utimes: detect utimes() correctly on OS/2 kLIBC +utimes() of OS/2 kLIBC has some limitations. + +1. OS/2 itself supports a file date since 1980

[PATCH 07/11] dup, dup2, fcntl: support a directory fd on OS/2 kLIBC

@@ 2015-11-07 KO Myung-Hun +dup, dup2, fcntl: support a directory fd on OS/2 kLIBC +On OS/2 kLIBC, dup(), dup2() and fcntl() do not work on a directory +fd. + +* lib/dup.c (dup_nothrow): New. +* lib/dup2.c (klibc_dup2dirfd): New. dup2() for a directory fd. +(klibc_dup2

[PATCH 09/11] opendir, closedir, dirfd, fdopendir: port to OS/2 kLIBC

ff --git a/ChangeLog b/ChangeLog index c55fce2..1de5860 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,30 @@ 2015-11-07 KO Myung-Hun +opendir, closedir, dirfd, fdopendir: port to OS/2 kLIBC +* lib/closedir.c (closedir): Unregister fd if closedir() succeeds. +* lib/d

[PATCH 06/11] pipe_filter_ii_execute: port to OS/2 kLIBC

/ChangeLog index 602ce34..205ed29 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,16 @@ 2015-11-07 KO Myung-Hun +pipe_filter_ii_execute: port to OS/2 kLIBC +Pipes on kLIBC does not support O_NONBLOCK like Win32. + +* lib/pipe-filter-ii.c (start_wrapper, _beginthreadex, CloseHandle

[PATCH 11/11] gnulib-tool: fall back into copy if symbolic link is not supported

+++ gnulib-tool | 28 2 files changed, 35 insertions(+), 4 deletions(-) diff --git a/ChangeLog b/ChangeLog index 157d50f..b2804bd 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,16 @@ 2015-11-07 KO Myung-Hun +gnulib-tool: fall back into copy if symbolic link

[PATCH 08/11] openat_proc_name: port to OS/2 kLIBC

insertions(+) diff --git a/ChangeLog b/ChangeLog index 15f50bc..c55fce2 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,14 @@ 2015-11-07 KO Myung-Hun +openat_proc_name: port to OS/2 kLIBC +OS/2 kLIBC provides a function to retrive a path from a fd. Use it +instead of /proc/self

[PATCH 01/11] stdint: check _INTPTR_T_DECLARED before defining intptr_t and uintptr_t

them. --- ChangeLog | 10 ++ lib/stdint.in.h | 13 + 2 files changed, 19 insertions(+), 4 deletions(-) diff --git a/ChangeLog b/ChangeLog index b0385b7..f26a5b4 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,13 @@ +2015-11-07 KO Myung-Hun + +stdint: check _INTPTR_T_DEC

[PATCH 05/11] wchar: fix 'conflicting types' error for `__wcwidth' on OS/2 kLIBC

1 deletion(-) diff --git a/ChangeLog b/ChangeLog index f3f5503..602ce34 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,14 @@ 2015-11-07 KO Myung-Hun +wchar: fix 'conflicting types' error for `__wcwidth' on OS/2 kLIBC +On OS/2 kLIBC, wcwidth is a macro to static inline func

[PATCH 04/11] w32spawn: clear SHELL_SPECIAL_CHARS and SHELL_SPACE_CHAR on OS/2 kLIBC

/ChangeLog index 6fc07ad..f3f5503 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,13 @@ 2015-11-07 KO Myung-Hun +w32spawn: clear SHELL_SPECIAL_CHARS and SHELL_SPACE_CHAR on OS/2 kLIBC +spawn() on OS/2 kLIBC is not silly like one on Windows + +* libc/w32spawn.h (SHELL_SPECIAL_CHARS

  1   2   3   >