This repository was archived by the owner on Aug 13, 2026. It is now read-only.
Repository navigation
This repository was archived by the owner on Aug 13, 2026. It is now read-only.
Duplicated charset when storing a RiakObject with a content type with charset [JIRA: CLIENTS-602] #549
Description
Activity
- changed the title
[-]Duplicated charset when storing a RiakObject with a content type with charset[/-][+]Duplicated charset when storing a RiakObject with a content type with charset [JIRA: CLIENTS-602][/+]on Sep 23, 2015 Few issues here actually:
- When we manually set a
charseton the RiakObject, it actually goes onto thecontent-type[0]. - When we convert the RiakObject to the ProtocolBuffer wire format, it will grab the
charsetoff of thecontent-typebuffer, but it won't remove it [1]. This is where we get the duplicateUTF-8. - When we convert from the ProtocolBuffer wire format back to a RiakObject, the charset gets put back onto the
content-type[2], but de-dupes it in the process [3].
Also, according to the IETF,
charsetisn't a real optional parameter to theapplication/jsoncontent-type:Note: No "charset" parameter is defined for this registration.
Adding one really has no effect on compliant recipients.http://www.iana.org/assignments/media-types/application/json
With that, we should stop putting the charset on the content-type automatically. If the user includes it in both fields, it will be duplicated by Webmachine (As seen in the reproducer). If they put each value in each field respectively, it should come out as expected in Webmachine.
- When we manually set a
If I store a RiakObject with a content type of
application/json;charset=UTF-8, the charset gets duplicated on the stored value when read by http, ending up with this content type:application/json;charset=UTF-8; charset=UTF-8I did a small sample to show the issue here:
https://github.com/gfbett/RiakIssueSample