Closed GoogleCodeExporter closed 9 years ago
This was fixed with 1.2.0 release. Most likely fix came with BlazeDS 4.0 AMF
serialization libraries.
Original comment by Piotr.Walczyszyn
on 2 Aug 2010 at 2:58
Looks like someone implemented a message delimiter that is attached to each
message to help distinguish between messages. This approach appears to work
fine although it is not as robust as a message length process like I
recommended. The delimiter is currently set to 4 bytes with values 99, 99, 99,
99. If there is a possibility that the amf data contains 4 bytes with the same
values, the message will be delimited unnecessarily which would cause messages
to be lost.
Just a suggestion. Nice job on releasing 1.2!
Original comment by nros...@gmail.com
on 2 Aug 2010 at 3:43
Your suggestion is valid and I think we will implement it this way in next
release of flerry. You can always join the project and give us a help with this
one ;)
Piotr
Original comment by Piotr.Walczyszyn
on 3 Aug 2010 at 9:02
Original issue reported on code.google.com by
nros...@gmail.com
on 3 Jul 2010 at 3:42Attachments: