Push the NATIVE_CRLF Makefile variable to C and added a test for native.

From: Pat Thoyts <[email protected]>

Commit 95f31e9a correctly points out that the NATIVE_CRLF setting is incorrectly set on Mingw git. However, the Makefile variable is not propagated to the C preprocessor and results in no change. This patch pushes the definition to the C code and adds a test to validate that when core.eol as native is crlf, we actually normalize text files to this line ending convention when core.autocrlf is false.

---
This mini series mainly updates git.git with patches from msysgit:
Patch 1 is taken "as is",
Patch 2 is taken from msysgit "nearly as is": the broken && chain had been fixed.

Changes since V2:
- Patch 1 shoud have the right Author line
(I used git-send-email for V2, where the Author was lost:
web.de does not allow to send mails with "From: Pat Thoyts" in the mail header, so it will change it to "From: tboegi". And git-send-email does not add another
"From: " line into the subject of the email, so I did it by hand)

Patch 2+3 had been squeezed together
(This may make the integration in msysgit a liitle bit harder)

 Makefile              |  3 +++
 t/t0026-eol-config.sh | 18 ++++++++++++++++++
 2 files changed, 21 insertions(+)

index 9f984a9..7d7db02 100644
@@ -1479,6 +1479,9 @@ ifdef NO_REGEX
 	COMPAT_CFLAGS += -Icompat/regex
 	COMPAT_OBJS += compat/regex/regex.o
 endif
+ifdef NATIVE_CRLF
+	BASIC_CFLAGS += -DNATIVE_CRLF
+endif
 
 ifdef USE_NED_ALLOCATOR
        COMPAT_CFLAGS += -Icompat/nedmalloc

index 4807b0f..43a580a 100755
@@ -80,4 +80,22 @@ test_expect_success 'autocrlf=true overrides unset eol' '
 	test -z "$onediff" && test -z "$twodiff"
 '
 
+test_expect_success NATIVE_CRLF 'eol native is crlf' '
+
+	rm -rf native_eol && mkdir native_eol &&
+	( cd native_eol &&
+	printf "*.txt text\n" > .gitattributes
+	printf "one\r\ntwo\r\nthree\r\n" > filedos.txt
+	printf "one\ntwo\nthree\n" > fileunix.txt
+	git init &&
+	git config core.autocrlf false &&
+	git config core.eol native &&
+	git add filedos.txt fileunix.txt &&
+	git commit -m "first" &&
+	rm file*.txt &&
+	git reset --hard HEAD &&
+	has_cr filedos.txt && has_cr fileunix.txt
+	)
+'
+
 test_done

2.1.0.rc2.210.g636bceb

Shouldn't Patch 1 also be made "nearly as is" to fix the same?