http: fix charset detection of extract_content_type()

extract_content_type() could not extract a charset parameter if the
parameter is not the first one and there is a whitespace and a following
semicolon just before the parameter. For example:

    text/plain; format=fixed ;charset=utf-8

And it also could not handle correctly some other cases, such as:

    text/plain; charset=utf-8; format=fixed
    text/plain; some-param="a long value with ;semicolons;"; charset=utf-8

Thanks-to: Jeff King <[email protected]>
Signed-off-by: Yi EungJun <[email protected]>
Signed-off-by: Junio C Hamano <[email protected]>

Author: Yi EungJun

http.c

@@ -927,7 +927,7 @@ static int extract_param(const char *raw, const char *name,
 		return -1;
 	raw++;
 
-	while (*raw && !isspace(*raw))
+	while (*raw && !isspace(*raw) && *raw != ';')
 		strbuf_addch(out, *raw++);
 	return 0;
 }
@@ -971,7 +971,7 @@ static void extract_content_type(struct strbuf *raw, struct strbuf *type,
 
 	strbuf_reset(charset);
 	while (*p) {
-		while (isspace(*p))
+		while (isspace(*p) || *p == ';')
 			p++;
 		if (!extract_param(p, "charset", charset))
 			return;

t/lib-httpd/error.sh

@@ -19,6 +19,10 @@ case "$PATH_INFO" in
 	printf "text/plain; charset=utf-16"
 	charset=utf-16
 	;;
+*odd-spacing*)
+	printf "text/plain; foo=bar ;charset=utf-16; other=nonsense"
+	charset=utf-16
+	;;
 esac
 printf "\n"
 

t/t5550-http-fetch-dumb.sh

@@ -191,5 +191,10 @@ test_expect_success 'http error messages are reencoded' '
 	grep "this is the error message" stderr
 '
 
+test_expect_success 'reencoding is robust to whitespace oddities' '
+	test_must_fail git clone "$HTTPD_URL/error/odd-spacing" 2>stderr &&
+	grep "this is the error message" stderr
+'
+
 stop_httpd
 test_done