patch 8.0.0245: zh_CN.cp936.po has a conversion error

Problem:    The generated zh_CN.cp936.po message file is not encoded properly.
Solution:   Instead of using zh_CN.po as input, use zh_CN.UTF-8.po.

Author: brammool

src/po/Makefile

@@ -232,12 +232,13 @@ sk.cp1250.po: sk.po
 	iconv -f iso-8859-2 -t cp1250 sk.po | \
 		sed -e 's/charset=ISO-8859-2/charset=cp1250/' -e 's/# Original translations/# Generated from sk.po, DO NOT EDIT/' > sk.cp1250.po
 
-# Convert zh_CN.po to create zh_CN.cp936.po.
-# set 'charset' to gbk to avoid that msfmt generates a warning
-zh_CN.cp936.po: zh_CN.po
+# Convert zh_CN.UTF-8.po to create zh_CN.cp936.po.
+# Set 'charset' to gbk to avoid that msfmt generates a warning.
+# This used to convert from zh_CN.po, but that results in a conversion error.
+zh_CN.cp936.po: zh_CN.UTF-8.po
 	rm -f zh_CN.cp936.po
-	iconv -f gb2312 -t cp936 zh_CN.po | \
-		sed -e 's/charset=gb2312/charset=gbk/' -e 's/# Original translations/# Generated from zh_CN.po, DO NOT EDIT/' > zh_CN.cp936.po
+	iconv -f UTF-8 -t cp936 zh_CN.UTF-8.po | \
+		sed -e 's/charset=utf-8/charset=gbk/' -e 's/# Original translations/# Generated from zh_CN.po, DO NOT EDIT/' > zh_CN.cp936.po
 
 # Convert ko.UTF-8.po to create ko.po.
 ko.po: ko.UTF-8.po

src/version.c

@@ -764,6 +764,8 @@ static char *(features[]) =
 
 static int included_patches[] =
 {   /* Add new patch number below this line */
+/**/
+    245,
 /**/
     244,
 /**/