No. It is a simple bug. The payload is defined to be a byte stream. It should
never be manipulated. The implementation MUST send the bytes exactly as the api
caller provided.
But I think there is something else at play - the implementation should not be
reading the bytes at all - which is the
The implementation has to read them because websockets use an odd xor masking, and the implementation has optimized this with longs, which have this endian problem, (yes it’s a bug). Maybe some vector api can replace the handmade procedure in the future without caring for the declared endianess.Gru
Yes. More garbage engineering. The effective chance of this is 0 so setting
mask to 1 isn’t really doing anything. Sometimes I’m really embarrassed by my
colleagues - the fact that this made it into the spec is just insane.
> On Mar 13, 2025, at 10:24 PM, Bernd wrote:
>
>
> The implementati
On Thu, 13 Mar 2025 13:42:40 GMT, Volkan Yazici wrote:
>> Fixes endian handling `jdk.internal.net.http.websocket.Frame.Masker`.
>>
>> ### Implementation notes
>>
>> I deleted the `Frame` clone in tests, and rewired the test code depending on
>> it to the actual `Frame`. To enable this, I relax
> Fixes endian handling `jdk.internal.net.http.websocket.Frame.Masker`.
>
> ### Implementation notes
>
> I deleted the `Frame` clone in tests, and rewired the test code depending on
> it to the actual `Frame`. To enable this, I relaxed the visibility of the
> actual `Frame`. I guess the `Frame`
On Thu, 13 Mar 2025 18:10:53 GMT, Daniel Fuchs wrote:
>> Volkan Yazici has updated the pull request incrementally with one additional
>> commit since the last revision:
>>
>> Fix copyright years
>
> src/java.net.http/share/classes/jdk/internal/net/http/websocket/Frame.java
> line 105:
>
>>
On Thu, 13 Mar 2025 13:42:40 GMT, Volkan Yazici wrote:
>> Fixes endian handling `jdk.internal.net.http.websocket.Frame.Masker`.
>>
>> ### Implementation notes
>>
>> I deleted the `Frame` clone in tests, and rewired the test code depending on
>> it to the actual `Frame`. To enable this, I relax
On Fri, 7 Mar 2025 17:54:29 GMT, Daniel Fuchs wrote:
>> This is a trivial doc cleanup. HttpResponse and module-info have a few links
>> that should use {@linkplain } instead of {@link } or . Also I noticed
>> one link target mismatch, and some clarification missing in the send/receive
>> buffe
On Thu, 13 Mar 2025 13:42:40 GMT, Volkan Yazici wrote:
>> Fixes endian handling `jdk.internal.net.http.websocket.Frame.Masker`.
>>
>> ### Implementation notes
>>
>> I deleted the `Frame` clone in tests, and rewired the test code depending on
>> it to the actual `Frame`. To enable this, I relax
On Thu, 13 Mar 2025 01:21:54 GMT, Archie Cobbs wrote:
> This PR fixes minor Javadoc typos in a few `java.base` classes. Extra curly
> braces have been sneaking in.
LGTM
-
PR Review: https://git.openjdk.org/jdk/pull/24022#pullrequestreview-2681298904
10 matches
Mail list logo
