[PATCH 2/3] ref-filter: teach the API to correctly handle CRLF

["Philippe Blain via GitGitGadget" <gitgitgadget@gmail.com>]

From: Philippe Blain <levraiphilippeblain@gmail.com>

The ref-filter API does not correctly handle commit or tag messages that
use CRLF as the line terminator. Such messages can be created with the
`--verbatim` option of `git commit` and `git tag`, or by using `git
commit-tree` directly.

This impacts the output of `git branch -v`, and `git branch`, `git
tag` and `git for-each-ref` when used with a `--format` argument
containing the atoms `%(contents:subject)` or `%(contents:body)`.

Fix this bug in ref-filter by improving the logic in `copy_subject` and
`find_subpos`.

Add tests for `branch`, `tag` and `for-each-ref` using
lib-crlf-messages.sh. The 'make commits' test at the beginning of
t3203-branch-output.sh needs to be modified since it did not use
`test_tick` and thus the commit hashes were not reproducible. For
simplicity, use `test_commit` as the content and name of the files
created in this setup test are irrelevant to the rest of the test
script.

Signed-off-by: Philippe Blain <levraiphilippeblain@gmail.com>
---
 ref-filter.c             | 19 +++++++++++++++++--
 t/t3203-branch-output.sh | 26 +++++++++++++++++++++-----
 t/t6300-for-each-ref.sh  |  5 +++++
 t/t7004-tag.sh           |  7 +++++++
 4 files changed, 50 insertions(+), 7 deletions(-)

diff --git a/ref-filter.c b/ref-filter.c
index 79bb5206783..537cc4de42c 100644
--- a/ref-filter.c
+++ b/ref-filter.c
@@ -1050,10 +1050,18 @@ static char *copy_subject(const char *buf, unsigned long len)
 {
 	char *r = xmemdupz(buf, len);
 	int i;
+	struct strbuf sb = STRBUF_INIT;
 
-	for (i = 0; i < len; i++)
+	strbuf_attach(&sb, r, len, len + 1);
+	for (i = 0; i < sb.len; i++) {
 		if (r[i] == '\n')
 			r[i] = ' ';
+		if (r[i] == '\r') {
+			strbuf_remove(&sb, i, 1);
+			i -= 1;
+		}
+	}
+	strbuf_detach(&sb, NULL);
 
 	return r;
 }
@@ -1184,15 +1192,22 @@ static void find_subpos(const char *buf,
 		eol = strchrnul(buf, '\n');
 		if (*eol)
 			eol++;
+		/*  protect against messages that might contain \r\n */
+		if (*eol == '\r')
+			eol++;
 		buf = eol;
 	}
 	*sublen = buf - *sub;
 	/* drop trailing newline, if present */
 	if (*sublen && (*sub)[*sublen - 1] == '\n')
 		*sublen -= 1;
+	/*  protect against commit messages that might contain \r\n */
+	else if (*sublen && (*sub)[*sublen - 1] == '\r')
+		*sublen -= 3; /* drop '\r\n\r' */
 
 	/* skip any empty lines */
-	while (*buf == '\n')
+	/* and protect against commit messages that might contain \r\n */
+	while (*buf == '\n' || *buf == '\r')
 		buf++;
 	*body = buf;
 	*bodylen = strlen(buf);
diff --git a/t/t3203-branch-output.sh b/t/t3203-branch-output.sh
index 71818b90f00..1235089619c 100755
--- a/t/t3203-branch-output.sh
+++ b/t/t3203-branch-output.sh
@@ -3,13 +3,11 @@
 test_description='git branch display tests'
 . ./test-lib.sh
 . "$TEST_DIRECTORY"/lib-terminal.sh
+. "$TEST_DIRECTORY"/lib-crlf-messages.sh
 
 test_expect_success 'make commits' '
-	echo content >file &&
-	git add file &&
-	git commit -m one &&
-	echo content >>file &&
-	git commit -a -m two
+	test_commit one &&
+	test_commit two
 '
 
 test_expect_success 'make branches' '
@@ -95,6 +93,24 @@ test_expect_success 'git branch --ignore-case --list -v pattern shows branch sum
 	awk "{print \$NF}" <tmp >actual &&
 	test_cmp expect actual
 '
+test_create_crlf_refs
+
+test_expect_success 'git branch -v works with CRLF input' '
+	cat >expect <<-EOF &&
+	  branch-one                     139b20d two
+	  branch-two                     d79ce16 one
+	  crlf                           2113b0e Subject first line
+	  crlf-empty-lines-after-subject 0a9530d Subject first line
+	  crlf-two-line-subject          f9ded1f Subject first line Subject second line
+	* master                         139b20d two
+	EOF
+	git branch -v >actual &&
+	test_cmp expect actual
+'
+
+test_crlf_subject_body_and_contents branch --list crlf*
+
+test_cleanup_crlf_refs
 
 test_expect_success 'git branch -v pattern does not show branch summaries' '
 	test_must_fail git branch -v branch*
diff --git a/t/t6300-for-each-ref.sh b/t/t6300-for-each-ref.sh
index 9c910ce7467..467924de3df 100755
--- a/t/t6300-for-each-ref.sh
+++ b/t/t6300-for-each-ref.sh
@@ -8,6 +8,7 @@ test_description='for-each-ref test'
 . ./test-lib.sh
 . "$TEST_DIRECTORY"/lib-gpg.sh
 . "$TEST_DIRECTORY"/lib-terminal.sh
+. "$TEST_DIRECTORY"/lib-crlf-messages.sh
 
 # Mon Jul 3 23:18:43 2006 +0000
 datestamp=1151968723
@@ -888,4 +889,8 @@ test_expect_success 'for-each-ref --ignore-case ignores case' '
 	test_cmp expect actual
 '
 
+test_create_crlf_refs
+
+test_crlf_subject_body_and_contents for-each-ref refs/heads/crlf*
+
 test_done
diff --git a/t/t7004-tag.sh b/t/t7004-tag.sh
index 6db92bd3ba6..25f465f34ea 100755
--- a/t/t7004-tag.sh
+++ b/t/t7004-tag.sh
@@ -10,6 +10,7 @@ Tests for operations with tags.'
 . ./test-lib.sh
 . "$TEST_DIRECTORY"/lib-gpg.sh
 . "$TEST_DIRECTORY"/lib-terminal.sh
+. "$TEST_DIRECTORY"/lib-crlf-messages.sh
 
 # creating and listing lightweight tags:
 
@@ -1969,6 +1970,12 @@ test_expect_success '--format should list tags as per format given' '
 	test_cmp expect actual
 '
 
+test_create_crlf_refs
+
+test_crlf_subject_body_and_contents tag --list tag-crlf*
+
+test_cleanup_crlf_refs
+
 test_expect_success "set up color tests" '
 	echo "<RED>v1.0<RESET>" >expect.color &&
 	echo "v1.0" >expect.bare &&
-- 
gitgitgadget