[
https://issues.apache.org/jira/browse/HTTPCORE-736?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17702630#comment-17702630
]
Julian Reschke commented on HTTPCORE-736:
-----------------------------------------
I believe you are confusing validity of message formats with the question
whether a specific message defines semantics for the payload.
The uniform message format just implies that in HTTP/1.1, if a request message
has a content-length (or uses T-E: chunked), that's what it is. That is, you
either need to handle the request body (or close the connection if being
unfriendly is ok :)). What happens next is defined by the semantics of the
specific message, and I'm not aware of any changes here.
In RFC 9110 we clarified that no, GET, HEAD and DELETE do _not_ define any
semantics for the request body (see
<https://greenbytes.de/tech/webdav/rfc9110.html#rfc.section.B.3.p.8>). OPTIONS
{*}does{*}, and this is actually used in WebDAV (RFCs 4918), please also see
<[https://greenbytes.de/tech/webdav/rfc2616.html#rfc.section.9.2.p.3]>.
> HttpClient GET method adds Content-Length: 0 header
> ---------------------------------------------------
>
> Key: HTTPCORE-736
> URL: https://issues.apache.org/jira/browse/HTTPCORE-736
> Project: HttpComponents HttpCore
> Issue Type: Bug
> Affects Versions: 4.4.16, 5.2.1
> Reporter: Jin Xu
> Priority: Critical
> Time Spent: 50m
> Remaining Estimate: 0h
>
> HttpClient GET method adds Content-Length: 0 header
> This would cuase some server/firewall deny this kind of http requests.
> See fixes of the same bug in other frameworks:
> java: https://bugs.openjdk.org/browse/JDK-8283544
> dart: https://github.com/dart-lang/shelf/pull/180
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@hc.apache.org
For additional commands, e-mail: dev-h...@hc.apache.org
