Yeah, I recognized that too. These are EOL changes, caused by mixed EOLs
introduced earlier. So I decided to leave them in the patch.
________________________________
Von: Henrique Mendonça (JIRA)
Gesendet: 16.02.2014 10:44
An: [email protected]
Betreff: [jira] [Commented] (THRIFT-2365) C# decodes too many binary bytes from
JSON
[
https://issues.apache.org/jira/browse/THRIFT-2365?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13902671#comment-13902671
]
Henrique Mendonça commented on THRIFT-2365:
-------------------------------------------
+1, nice test case!
ps.: wonder why you got all those non-modified lines in the patch, are using
git diff directly?
> C# decodes too many binary bytes from JSON
> ------------------------------------------
>
> Key: THRIFT-2365
> URL: https://issues.apache.org/jira/browse/THRIFT-2365
> Project: Thrift
> Issue Type: Bug
> Components: C# - Library
> Reporter: Jens Geyer
> Assignee: Jens Geyer
> Labels: json
> Fix For: 0.9.2
>
> Attachments:
> THRIFT-2365_Csharp_decodes_too_many_binary_bytes_from_JSON.patch
>
>
> If the length of a certain binary data block is not a multiple of 3 (e.g.
> 16384 bytes), the JSON decoder returns a byte[] array with the length of the
> next multiple of 3 (e.g. 16386 bytes), because possible [base64 fill
> bytes|http://tools.ietf.org/html/rfc4648#section-4] contained in the input
> data are not handled correctly during the decoding process.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)