(Java) app engine in web application use out.println("测试") and print messy code ,(后台打印中文信息乱码)

[Haitao Zhang]

i was a new user for engine web application

After deploy my web application project to app engine,use out.println("测试") print  messy code . 

1.i test project in window eclispe , and it can print correct chinese , but when deploy to app engine, the log print messy code(like ??????--???)

2.another question was when i fetch info from some china website, is also print messy code 

i search  that question whereever ,but nobody question like that, it let me feel helpness,

can someone tell me how to fix that situation?

[Vinny P]

On Tue, Sep 24, 2013 at 3:57 AM, Haitao Zhang <seati...@gmail.com> wrote:

i was a new user for engine web application

After deploy my web application project to app engine,use out.println("测试") print  messy code . 

1.i test project in window eclispe , and it can print correct chinese , but when deploy to app engine, the log print messy code(like ??????--???)

The actual logs are still there, but the charset/encoding is different from what App Engine is expecting so you get those question marks and characters that look like boxes. See here for an example: 

http://stackoverflow.com/questions/5445137/utf-8-encoded-html-pages-show-questions-marks-instead-of-characters

If you use Chrome, try this page and see if it helps: http://productforums.google.com/forum/#!topic/chrome/EfrrDYQ1-_g

On Tue, Sep 24, 2013 at 3:57 AM, Haitao Zhang <seati...@gmail.com> wrote:

2.another question was when i fetch info from some china website, is also print messy code 

Is the urlfetch itself corrupted or the printing into logs damaged? Make sure you're encoding/decoding in UTF-8.

 

-----------------

-Vinny P

Technology & Media Advisor

Chicago, IL

App Engine Code Samples: http://www.learntogoogleit.com

 

[Haitao Zhang]

: ) thanks, that give me a way to find the answer. so i change the charset in app engine many many times. but still unsolve. so i change the server ,hahah 

i found some idea in chinese , so list here:

//-------------------------------------------------------------

最终的class文件都是以unicode编码的，我们前面所做的工作就是把各种不同的编码转换为unicode编码，比如从GBK转换为unicode,从UTF-8转换为unicode。因为只有采用正确的编码来转码才能保证不出现乱码。Jvm在运行时其内部都是采用unicode编码的，其实在输出时，又会做一次编码的转换。让我们分两种情况来讨论。

1.java中采用Sysout.out.println输出。

比如：Sysout.out.println(“我们”)。经过正确的解码后”我们”是unicode保存在内存中的，但是在向标准输出(控制台)输出时，jvm又做了一次转码，它会采用操作系统默认编码(中文操作系统是GBK)，将内存中的unicode编码转换为GBK编码，然后输出到控制台。因为我们操作系统是中文系统，所以往终端显示设备上打印字符时使用的也是GBK编码。因为终端的编码无法手动改变，所以这个过程对我们来说是透明的，只要编译时能正确转码，最终的输出都将是正确的，不会出现乱码。在Eclipse中可以设置控制台的字符编码，具体位置在Run Configuration对话框的Common标签里,我们可以试着设置为UTF-8,此时的输出就是乱码了。因为输出时是采用GBK编码的，而显示却是使用UTF-8，编码不同，所以出现乱码。

//-------------------------------------------------------------

i used that method to print the app engine system params

System.getProperties().list(System.out); 

and i found app engine charset was "ANSI_X3.4-1968"

so i set system-properties params to change the charset in appengine-web.xml ,like :

<system-properties>

<property name="java.util.logging.config.file" value="WEB-INF/logging.properties" />

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

<property name="default.client.encoding" value="gbk" />

<property name="sun.jnu.encoding" value="gbk" />

<property name="sun.io.unicode.encoding" value="gbk" />

<property name="user.language" value="en_US.gbk" />

</system-properties>

it can change the charset in app engine,but after i test many many times it was still failure. 

anyway i think that should be the correct way to solve that question