Google AppEngine - How to set default Charset / file-encoding (to UTF-8) for Google AppEngine

问题

Situation

I'm using AppEngine Cloud Endpoints (Java) as backendserver for my Android-application. In a Cron-Job the server regularly downloads new data (String), filters out relevant information and sends it via Firebase Cloud Messaging to different topics (by posting HTTP-Request to https://fcm.googleapis.com/fcm/send).

The (damn simple) Problem

The downloaded text is in German, containing letters like ä,ö,ü,ß and (to not have the famous ?s istead of these characters) I want to change the encoding to UTF-8.

US-ASCII encoding is the default when deployed to AppEngine (why isn't it UTF-8 by default...). When tested locally, the default encoding of my computer is used which is UTF-8 and everything is working just fine.

So all I want is Charset.getDefaultCharset() to return UTF-8.
How do I achieve this? Thanks in advance, all help appreciated.

What I've tried

• adding this to appengine-web.xml:

  <env-variables> <env-var name="DEFAULT_ENCODING" value="UTF-8" /> </env-variables>

  When testing locally I got IncorrectEnvironmentVariableException. When deployed encoding was still ASCII.

• adding this to <system-properties>-tag in appengine-web.xml:

  <property name="file.encoding" value="UTF-8" /> <property name="DEFAULT_ENCODING" value="UTF-8" />

  changed nothing, not only when deployed but also when I tried to use this to set the encoding to US-ASCII on local developement server, still UTF-8 was returned by Charset.getDefaultCharset()

• even tried to use reflection:

  System.setProperty("file.encoding","UTF-8"); Field charset = Charset.class.getDeclaredField("defaultCharset"); charset.setAccessible(true); charset.set(null,null);

  (caused IllegalAccessException)

• tried setting encoding as content-type-property with the HTTP-post (didn't work):

  connection.setRequestProperty("Content-Type", application/json; charset=UTF-8);

回答1:

Short Answer

If you have problems with encoding (i.e. '?'s) and it seems like changing default-charset solves the problem then what you really have to do is to find the code where the default-charset is used for encoding and pass a custom charset to be used instead.

(Like @AndreiVolgin commented: You should never rely on the system default-charset.)

More detailed Answer

A charset specifies how chars are represented as numbers (i.e. bytes). The internal charset of all Strings in Java is in fact UTF-16 so the only time custom charsets become important is when converting a String to a byte-Array (String.getBytes()) or vice versa (new String(byte[]).
This is where the default-charset is used if you don't pass a charset like String.getBytes(Charset) or new String(byte[], Charset).

You get to see '?'s when different charsets are used for encoding (String.getBytes()) and decoding (new String(byte[]).

So to get rid of the '?'s replace String.getBytes() by String.getBytes(Charset) (or maybe replace new String(byte[]) by new String(byte[], Charset) but don't know if thats a thing).

---

For me the encoding problem occured when sending some text via the internet because the String needs to be converted to byte[] to send it using HttpURLConnection.getOutputStream().write(byte[]).

To be precise: In class com.google.android.gcm.server.Sender, method post(java.lang.String, java.lang.String, java.lang.String), line 471 was byte[] bytes = body.getBytes();.

I changed it (overwrote it) to byte[] bytes = body.getBytes("UTF-8"); - solved it for me.

来源：https://stackoverflow.com/questions/41204936/google-appengine-how-to-set-default-charset-file-encoding-to-utf-8-for-goo