For the FileServer Example:
The FileServerHandler Sends a String Msg to the client requesting the 
client to send a file name to retrieve.
The msg flow through the channel pipeline is as follows:
1. FileServerHandler 
2. StringEncoder

However, the outbound channel pipeine msg flow as specified in the 
FileServer is:
1. FileServerHandler
2. chunkedWriteHandler
3. StringEncoder

*Question 1*: *How & why does the String msg generated from the 
FileServerHandler skip over the chunkedWriteHandler to the StringEncoder? 
What causes the ChunkedWriteHandler to be skipped over? *Is this because 
ChunkedWriteHandler is expecting a ChunkInput and so the handler rejects 
any msg that is not of type ChunkedInput and passes it on to the next 
outbound channelHandler which will be the StringEncoder which is expecting 
a String. if so does the ChunkedWriteHandler automatically ignores the 
String msg and uses the (context.fireEventType(ev) ) method to push the msg 
to the StringEncoder

When the fileServerHandler sends the contents of the file, it uses 
chunkWriteHandler if zero-copy is not allowed on the native machine and 
therefore the msg flow through the channel pipeline is as follows:
1. FileServerHandler
2. chunkedWriteHandler


*Question 2:** How does the data stream generated by chunkedWriteHandler 
bypass the StringEncoder? What causes the StringEncoder to be skipped over?* I 
know the StringEncoder is expecting a String to decode to write out to the 
socket, but how does data stream generated by chunkedWriteHandler bypass 
the StringEncoder

-- 
You received this message because you are subscribed to the Google Groups 
"Netty discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/netty/7de1daf7-3a9c-4f17-b2aa-ac435115bae0%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to