Brian Gernhardt [off-list ref] writes:
quoted hunk
It used to be that APPLE_COMMON_CRYPTO did nothing when BLK_SHA1 was
set. But APPLE_COMMON_CRYPTO is now used for more than just SHA1 (see
3ef2bca) so make sure that the appropriate libraries are always set.
Signed-off-by: Brian Gernhardt <redacted>
---
Makefile | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/Makefile b/Makefile
index 82f2e22..7051956 100644
--- a/Makefile
+++ b/Makefile
@@ -1182,6 +1182,9 @@ ifdef NEEDS_SSL_WITH_CRYPTO
else
LIB_4_CRYPTO = $(OPENSSL_LINK) -lcrypto
endif
+ifdef APPLE_COMMON_CRYPTO
+ LIB_4_CRYPTO += -framework Security -framework CoreFoundation
+endif
endif
ifdef NEEDS_LIBICONV
ifdef ICONVDIR
@@ -1413,7 +1416,6 @@ ifdef PPC_SHA1
LIB_H += ppc/sha1.h
else
ifdef APPLE_COMMON_CRYPTO
- LIB_4_CRYPTO += -framework Security -framework CoreFoundation
COMPAT_CFLAGS += -DCOMMON_DIGEST_FOR_OPENSSL
SHA1_HEADER = <CommonCrypto/CommonDigest.h>
else
Hmph.
So the people previously tested this must have built imap-send
without blk-sha1, which not just linked with these libs but also
included the <CommonCrypto/CommonDigest.h> header file and defined
the -DCOMMON_DIGEST_FOR_OPENSSL preprocessor macro. Building with
blk-sha1 would not have worked for them.
Now we always link with these libraries, even when building with
blk-sha1. Do the COMPAT_CFLAGS and SHA1_HEADER pieces only needed
when using the SHA1 digest implementation from CommonCrypto and
nothing imap-send uses?
On Mon, Aug 5, 2013 at 10:52 AM, Junio C Hamano [off-list ref] wrote:
Brian Gernhardt [off-list ref] writes:
quoted
It used to be that APPLE_COMMON_CRYPTO did nothing when BLK_SHA1 was
set. But APPLE_COMMON_CRYPTO is now used for more than just SHA1 (see
3ef2bca) so make sure that the appropriate libraries are always set.
Signed-off-by: Brian Gernhardt <redacted>
---
Makefile | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/Makefile b/Makefile
index 82f2e22..7051956 100644
--- a/Makefile
+++ b/Makefile
@@ -1182,6 +1182,9 @@ ifdef NEEDS_SSL_WITH_CRYPTO
else
LIB_4_CRYPTO = $(OPENSSL_LINK) -lcrypto
endif
+ifdef APPLE_COMMON_CRYPTO
+ LIB_4_CRYPTO += -framework Security -framework CoreFoundation
+endif
endif
ifdef NEEDS_LIBICONV
ifdef ICONVDIR@@ -1413,7 +1416,6 @@ ifdef PPC_SHA1
LIB_H += ppc/sha1.h
else
ifdef APPLE_COMMON_CRYPTO
- LIB_4_CRYPTO += -framework Security -framework CoreFoundation
COMPAT_CFLAGS += -DCOMMON_DIGEST_FOR_OPENSSL
SHA1_HEADER = <CommonCrypto/CommonDigest.h>
else
Hmph.
So the people previously tested this must have built imap-send
without blk-sha1, which not just linked with these libs but also
included the <CommonCrypto/CommonDigest.h> header file and defined
the -DCOMMON_DIGEST_FOR_OPENSSL preprocessor macro. Building with
blk-sha1 would not have worked for them.
Now we always link with these libraries, even when building with
blk-sha1. Do the COMPAT_CFLAGS and SHA1_HEADER pieces only needed
when using the SHA1 digest implementation from CommonCrypto and
nothing imap-send uses?
LIB_4_CRYPTO is used by imap-send only, and these libraries are needed
for the base64 git_CC_EVP_(Encode|Decode), so unconditionally adding
these libraries there is correct.
COMPAT_CFLAGS and SHA1_HEADER enable the common crypto SHA1 only.
BLK_SHA1 provides its own SHA1 so they're not needed there.
I tested the tip of da/darwin (pu) w/ and w/out BLK_SHA1.
Tested-by: David Aguilar <redacted>
--
David