[PATCH 1/2] ident.c: New function valid_ident for checking ident string formatting

[drafnel@gmail.com]

From: Brandon Casey <drafnel@gmail.com>

This function provides a centralized method for checking ident
strings which can be shared by the functions creating or parsing
author, committer, or tagger fields.

Signed-off-by: Brandon Casey <drafnel@gmail.com>
---
 cache.h |    1 +
 ident.c |   69 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 70 insertions(+), 0 deletions(-)

diff --git a/cache.h b/cache.h
index d5d5dad..92f21e1 100644
--- a/cache.h
+++ b/cache.h
@@ -580,6 +580,7 @@ extern const char *git_author_info(int);
 extern const char *git_committer_info(int);
 extern const char *fmt_ident(const char *name, const char *email, const char *date_str, int);
 extern const char *fmt_name(const char *name, const char *email);
+extern int valid_ident(const char *buf, size_t size, int flag);
 
 struct checkout {
 	const char *base_dir;
diff --git a/ident.c b/ident.c
index ed44a53..1dc9b22 100644
--- a/ident.c
+++ b/ident.c
@@ -167,6 +167,75 @@ static int copy(char *buf, size_t size, int offset, const char *src)
 	return offset;
 }
 
+int valid_ident(const char *buf, size_t size, int flag)
+{
+        const char *s, *lb, *rb;
+	int name_addr_only = (flag & IDENT_NO_DATE);
+
+        /* check for leading crud */
+	if (crud(*buf))
+		return 0;
+
+	/* Find start of email and check for disallowed chars in name */
+	if (!(lb = strstr(buf, " <")) || strpbrk(buf, "<>\n") < lb)
+		return 0;
+
+	/* Check for empty name and trailing crud in name */
+	if (lb == buf || crud(lb[-1]))
+		return 0;
+
+	/* Find end of email and check for disallowed chars in email */
+	if (!(rb = strchr(lb+2, '>')) || strpbrk(lb+2, "><\n ") < rb)
+		return 0;
+
+	/* Check for leading or trailing crud in email */
+	if (rb > lb+2 && (crud(lb[2]) || crud(rb[-1])))
+		return 0;
+
+	s = rb+1;
+	if (!name_addr_only) {
+		char tz[5];
+		size_t len;
+
+		/* a space separates email from timestamp */
+		if (*s != ' ')
+			return 0;
+		s++;
+
+		/*
+		 * timestamp, 1 or more digits followed by space.
+		 * fast-import at least assumes the timestamp and
+		 * timezone fields will be 23 bytes or less. That
+		 * leaves 17 bytes for the timestamp.
+		 */
+		if (!(len = strspn(s, "0123456789")) || len > 17)
+			return 0;
+		s += len;
+
+		/* a space separates timestamp from timezone */
+		if (*s != ' ')
+			return 0;
+		s++;
+
+		/*
+                 * timezone, 5 digits [+-]hhmm, max. 1400
+                 * A buffer is used here to hold the timezone characters
+                 * since our string may not be nul terminated immediately
+                 * after the timezone field. Though it must adhere to the
+                 * size parameter supplied by the caller.
+                 */
+		if (!((s[0] == '+' || s[0] == '-') &&
+		      strlcpy(tz, s+1, 5) >= 4 &&
+		      strspn(tz, "0123456789") == 4 &&
+		      atoi(tz) <= 1400))
+			return 0;
+
+		s += 5;
+	}
+
+	return (s - buf == size);
+}
+
 static const char au_env[] = "GIT_AUTHOR_NAME";
 static const char co_env[] = "GIT_COMMITTER_NAME";
 static const char *env_hint =
-- 
1.5.5.67.g9a49