[ 
https://issues.apache.org/jira/browse/HTTPCORE-431?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15461555#comment-15461555
 ] 

ASF GitHub Bot commented on HTTPCORE-431:
-----------------------------------------

Github user pauldraper commented on the issue:

    https://github.com/apache/httpcore/pull/30
  
    > This code makes no distinction between BE and LE encodings described in 
rfc4627.
    > rfc7159 explicitly forbids byte order marks, while this patch depends on 
BOMs.
    
    RFC 7159 disallows UTF-16BE, UTF-16LE, UTF-32BE and UTF-32LE.
    
    > JSON text SHALL be encoded in UTF-8, UTF-16, or UTF-32. 
    
    This code works for these three encodings.
    
    I agree, however, that this code should work for RFC 4627 which permitted 
the BE/LE encodings.
    
    > IMHO, workarounds for improper servers should not be injected in core 
functionality.
    
    I don't see how this is a "workaround" for improper servers. This project 
follows RFC 2616 when decoding entities. I suggest that it also follow RFC 
4627/7159.
    
    > Probably, separate function like EntityUtils.safeJsonToString(...) should 
be introduced, so anybody using this function will be informed than slight 
performance penalty will apply.
    
    To be clear, the performance penalty you're thinking of is a string 
comparison of the MIME type?


> Correct character encoding default for application/json
> -------------------------------------------------------
>
>                 Key: HTTPCORE-431
>                 URL: https://issues.apache.org/jira/browse/HTTPCORE-431
>             Project: HttpComponents HttpCore
>          Issue Type: Improvement
>          Components: HttpCore
>            Reporter: Paul Draper
>            Priority: Minor
>
> RFC 4627 and RFC 7159 require that parsers auto-detect the encoding of 
> application/json. UTF-8 is the suggested default, though it's possible to 
> detect the encoding of valid JSON unambiguously.
> Currently, EntityUtils.toString defaults to ISO-8859-1 (the default for text 
> formats without a charset parameter).
> Given how common application/json is, it would be great for 
> EntityUtils.toString to use the correct encoding.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@hc.apache.org
For additional commands, e-mail: dev-h...@hc.apache.org

Reply via email to