ZF-5567: Investigate Zend_Json_Encoder/Decoder::encodeUnicodeString failure on UTF-8 Chars > 65535

Issue Type:
Bug
Created:
2009-01-16T08:31:52.000+0000
Last Updated:
2012-05-03T13:00:35.000+0000
Status:
Reopened
Fix version(s):
    Reporter:
    Benjamin Eberlei (beberlei)
    Assignee:
    Benjamin Eberlei (beberlei)
    Tags:
    • Zend_Json
    • zf-crteam-padraic
    • zf-crteam-priority
    Related issues:
    Attachments:

    Description

    It seems that the algorithms for Unicode Encoding and/or Decoding fail on characters 65535.

    Comments

    Posted by Pádraic Brady (padraic) on 2009-12-23T05:40:05.000+0000

    Moving Priority to Major from N/A (may have buried it in filters). May have been recently duplicated.

    Posted by Pádraic Brady (padraic) on 2009-12-23T05:41:20.000+0000

    Benjamin, any chance you could expand on the description perhaps with a test case or example?

    Posted by Pádraic Brady (padraic) on 2011-08-20T12:21:09.000+0000

    Test with higher Unicode (e.g. Gothic range) but could locate no issues. Translation to/from array/strings using a mix of native and Zend_Json functions showed no translation difficulty. Given lack of reporter tests, I'm closing this as a non-issue.

    Posted by Pádraic Brady (padraic) on 2011-08-20T12:34:34.000+0000

    Of course, would help if my last test was written correctly... Reopening. Results in garbage text on conversion.

    Posted by BitConstructor Co. Ltd. (bitconstructor) on 2012-05-03T13:00:35.000+0000

    encodeUnicodeString misses the default case in the switch on line 458. This may break the conversion for not handled cases. I got the problem with the following string: คอนเสิร์ต ไผ่ พงศธร

    The issue seems to occur less if I introduce the default case and copy the char as it is