Keep the charset off types that only spell xml

Matching xml anywhere in the type also caught the Office formats, whose
names hold openxmlformats, so a .docx went out with a UTF-8 charset.

2 files changed, 29 insertions, 2 deletions

diff --git a/source/ui-shared.c b/source/ui-shared.cindex c23e2e1..e5afb2e 100644--- a/source/ui-shared.c+++ b/source/ui-shared.c@@ -1426,11 +1426,13 @@ void cgit_print_age(time_t t, int tz, time_t max_relative)  /*  * Whether a content type carries text a charset applies to, which is every- * text type and the XML ones, the atom feed among them.+ * text type and the XML types of RFC 7303, the atom feed among them.  */ static int type_has_charset(const char *mimetype) {-	return starts_with(mimetype, "text/") || strstr(mimetype, "xml");+	return starts_with(mimetype, "text/") ||+		!strcmp(mimetype, "application/xml") ||+		ends_with(mimetype, "+xml"); }  void cgit_print_http_headers(void)
diff --git a/source/ui-shared.c b/source/ui-shared.c
index c23e2e1..e5afb2e 100644
--- a/source/ui-shared.c
+++ b/source/ui-shared.c
@@ -1426,11 +1426,13 @@ void cgit_print_age(time_t t, int tz, time_t max_relative)
 
 /*
  * Whether a content type carries text a charset applies to, which is every
- * text type and the XML ones, the atom feed among them.
+ * text type and the XML types of RFC 7303, the atom feed among them.
  */
 static int type_has_charset(const char *mimetype)
 {
-	return starts_with(mimetype, "text/") || strstr(mimetype, "xml");
+	return starts_with(mimetype, "text/") ||
+		!strcmp(mimetype, "application/xml") ||
+		ends_with(mimetype, "+xml");
 }
 
 void cgit_print_http_headers(void)
diff --git a/tests/t0105-plain.sh b/tests/t0105-plain.shindex 4041d98..27e8daa 100755--- a/tests/t0105-plain.sh+++ b/tests/t0105-plain.sh@@ -46,4 +46,29 @@ test_expect_success 'a blob whose bytes look binary is served as octet-stream' ' 	sed "/^$/q" tmp | grep "^Content-Type: application/octet-stream$" ' +# RFC 7303 names the XML types by an xml subtype or a +xml suffix, and those+# take a charset like text does. A type that only spells xml inside a word,+# as the Office formats do, is binary and gets none.+test_expect_success 'only text and xml types carry a charset' '+	(+		cd repos/foo &&+		printf "PK" >doc.docx &&+		printf "<svg/>" >pic.svg &&+		git add doc.docx pic.svg &&+		git commit -m types+	) &&+	docx=application/vnd.openxmlformats-officedocument. &&+	docx=${docx}wordprocessingml.document &&+	{+		echo "enable-html-serving=1" &&+		echo "mimetype.docx=$docx" &&+		echo "mimetype.svg=image/svg+xml" &&+		cat cgitrc+	} >typesrc &&+	CGIT_CONFIG="$PWD/typesrc" QUERY_STRING="url=foo/plain/doc.docx" cgit >tmp &&+	sed "/^$/q" tmp | grep "^Content-Type: $docx$" &&+	CGIT_CONFIG="$PWD/typesrc" QUERY_STRING="url=foo/plain/pic.svg" cgit >tmp &&+	sed "/^$/q" tmp | grep "^Content-Type: image/svg+xml; charset=UTF-8$"+'+ test_done
diff --git a/tests/t0105-plain.sh b/tests/t0105-plain.sh
index 4041d98..27e8daa 100755
--- a/tests/t0105-plain.sh
+++ b/tests/t0105-plain.sh
@@ -46,4 +46,29 @@ test_expect_success 'a blob whose bytes look binary is served as octet-stream' '
 	sed "/^$/q" tmp | grep "^Content-Type: application/octet-stream$"
 '
 
+# RFC 7303 names the XML types by an xml subtype or a +xml suffix, and those
+# take a charset like text does. A type that only spells xml inside a word,
+# as the Office formats do, is binary and gets none.
+test_expect_success 'only text and xml types carry a charset' '
+	(
+		cd repos/foo &&
+		printf "PK" >doc.docx &&
+		printf "<svg/>" >pic.svg &&
+		git add doc.docx pic.svg &&
+		git commit -m types
+	) &&
+	docx=application/vnd.openxmlformats-officedocument. &&
+	docx=${docx}wordprocessingml.document &&
+	{
+		echo "enable-html-serving=1" &&
+		echo "mimetype.docx=$docx" &&
+		echo "mimetype.svg=image/svg+xml" &&
+		cat cgitrc
+	} >typesrc &&
+	CGIT_CONFIG="$PWD/typesrc" QUERY_STRING="url=foo/plain/doc.docx" cgit >tmp &&
+	sed "/^$/q" tmp | grep "^Content-Type: $docx$" &&
+	CGIT_CONFIG="$PWD/typesrc" QUERY_STRING="url=foo/plain/pic.svg" cgit >tmp &&
+	sed "/^$/q" tmp | grep "^Content-Type: image/svg+xml; charset=UTF-8$"
+'
+
 test_done