pg_config_h.in not up-to-date

Started by Antonin Houskaalmost 5 years ago7 messages

ah@cybertec.at

almost 5 years ago

1 attachment(s)

When I run "autoreconf" on the master branch, git generates the diff
below. Shouldn't it just be applied? I suppose someone changed configure.ac
and forgot to update the generated file.

--
Antonin Houska
Web: https://www.cybertec-postgresql.com

Attachments:

pg_config_h.in.difftext/x-diffDownload

diff --git a/src/include/pg_config.h.in b/src/include/pg_config.h.in
index 55cab4d2bf..806eabac81 100644
--- a/src/include/pg_config.h.in
+++ b/src/include/pg_config.h.in
@@ -899,7 +899,7 @@
 /* Define to select named POSIX semaphores. */
 #undef USE_NAMED_POSIX_SEMAPHORES
 
-/* Define to build with OpenSSL support. (--with-ssl=openssl) */
+/* Define to 1 if you have OpenSSL support. */
 #undef USE_OPENSSL
 
 /* Define to 1 to build with PAM support. (--with-pam) */

Tom Lane

tgl@sss.pgh.pa.us

almost 5 years ago

In reply to: Antonin Houska (#1)

Re: pg_config_h.in not up-to-date

Antonin Houska <ah@cybertec.at> writes:

When I run "autoreconf" on the master branch, git generates the diff
below. Shouldn't it just be applied? I suppose someone changed configure.ac
and forgot to update the generated file.

Yeah, looks like fe61df7f8 is at fault. Michael?

regards, tom lane

Michael Paquier

michael@paquier.xyz

almost 5 years ago

In reply to: Tom Lane (#2)

1 attachment(s)

Re: pg_config_h.in not up-to-date

On Fri, Feb 19, 2021 at 01:42:38AM -0500, Tom Lane wrote:

Antonin Houska <ah@cybertec.at> writes:

When I run "autoreconf" on the master branch, git generates the diff
below. Shouldn't it just be applied? I suppose someone changed configure.ac
and forgot to update the generated file.

Yeah, looks like fe61df7f8 is at fault. Michael?

Indeed, thanks. It looks like a "git add" that was fat-fingered. I
would like to make things more consistent with the attached.
Thoughts?
--
Michael

Attachments:

openssl-autoconf.patchtext/x-diff; charset=us-asciiDownload

diff --git a/src/include/pg_config.h.in b/src/include/pg_config.h.in
index 55cab4d2bf..3c1c515d3c 100644
--- a/src/include/pg_config.h.in
+++ b/src/include/pg_config.h.in
@@ -899,7 +899,7 @@
 /* Define to select named POSIX semaphores. */
 #undef USE_NAMED_POSIX_SEMAPHORES
 
-/* Define to build with OpenSSL support. (--with-ssl=openssl) */
+/* Define to 1 to build with OpenSSL support. (--with-ssl=openssl). */
 #undef USE_OPENSSL
 
 /* Define to 1 to build with PAM support. (--with-pam) */
diff --git a/configure.ac b/configure.ac
index 07da84d401..9a7d836ead 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1234,7 +1234,7 @@ if test "$with_ssl" = openssl ; then
   # thread-safety. In 1.1.0, it's no longer required, and CRYPTO_lock()
   # function was removed.
   AC_CHECK_FUNCS([CRYPTO_lock])
-  AC_DEFINE([USE_OPENSSL], 1, [Define to 1 if you have OpenSSL support.])
+  AC_DEFINE([USE_OPENSSL], 1, [Define to 1 to build with OpenSSL support. (--with-ssl=openssl).])
 elif test "$with_ssl" != no ; then
   AC_MSG_ERROR([--with-ssl must specify openssl])
 fi

Tom Lane

tgl@sss.pgh.pa.us

almost 5 years ago

In reply to: Michael Paquier (#3)

Re: pg_config_h.in not up-to-date

Michael Paquier <michael@paquier.xyz> writes:

Indeed, thanks. It looks like a "git add" that was fat-fingered. I
would like to make things more consistent with the attached.

+1, but I think the first period in this comment is redundant:

+ AC_DEFINE([USE_OPENSSL], 1, [Define to 1 to build with OpenSSL support. (--with-ssl=openssl).])

regards, tom lane

Michael Paquier

michael@paquier.xyz

almost 5 years ago

In reply to: Tom Lane (#4)

Re: pg_config_h.in not up-to-date

On Fri, Feb 19, 2021 at 02:21:21AM -0500, Tom Lane wrote:

Michael Paquier <michael@paquier.xyz> writes:

Indeed, thanks. It looks like a "git add" that was fat-fingered. I
would like to make things more consistent with the attached.

+1, but I think the first period in this comment is redundant:

+ AC_DEFINE([USE_OPENSSL], 1, [Define to 1 to build with OpenSSL support. (--with-ssl=openssl).])

I guess that you mean the second period here to be more consistent
with the others?  That would mean the following diff:
+  AC_DEFINE([USE_OPENSSL], 1, [Define to 1 to build with OpenSSL support. (--with-ssl=openssl)])
--
Michael

Tom Lane

tgl@sss.pgh.pa.us

almost 5 years ago

In reply to: Michael Paquier (#5)

Re: pg_config_h.in not up-to-date

Michael Paquier <michael@paquier.xyz> writes:

On Fri, Feb 19, 2021 at 02:21:21AM -0500, Tom Lane wrote:

+1, but I think the first period in this comment is redundant:
+  AC_DEFINE([USE_OPENSSL], 1, [Define to 1 to build with OpenSSL support. (--with-ssl=openssl).])

I guess that you mean the second period here to be more consistent
with the others? That would mean the following diff:
+ AC_DEFINE([USE_OPENSSL], 1, [Define to 1 to build with OpenSSL support. (--with-ssl=openssl)])

Hm. It should be consistent with the rest, for sure. Personally I'd put
the only period at the end, but I suppose we should stick with the
prevailing style if there is one.

regards, tom lane

Michael Paquier

michael@paquier.xyz

almost 5 years ago

In reply to: Tom Lane (#6)

Re: pg_config_h.in not up-to-date

On Fri, Feb 19, 2021 at 09:57:22AM -0500, Tom Lane wrote:

Hm. It should be consistent with the rest, for sure. Personally I'd put
the only period at the end, but I suppose we should stick with the
prevailing style if there is one.

Thanks. I have just used the same style as XML, LDAP and LLVM then.
Thanks Antonin for the report.
--
Michael