Re^3: use bytes vs packed data

The second string I'm packing here came from the Net::Async::FastCGI::Request stdin data, and was UTF8-flagged.

From what you say, it sounds like rabbit wants bytes. Did you encode the data? Why are you decoding an HTTP request in the first place?

Comment on Re^3: use bytes vs packed data

Replies are listed 'Best First'.
Re^4: use bytes vs packed data by RDOlson (Initiate) on May 05, 2011 at 21:55 UTC
I'm not explicitly decoding anything in my code -- I just call `Net::Async::FastCGI::Request->read_stdin`. However, reading through that I see that THAT code is explicitly calling a UTF8 decode (it has a default encoding set to UTF8). So it looks like the correct solution is to explicitly set the default encoding to undef on the FastCGI object. Thanks all for the clues to get here ... this was not at all obvious to me.	[reply] [d/l]

Replies are listed 'Best First'.

Re^4: use bytes vs packed data
by RDOlson (Initiate) on May 05, 2011 at 21:55 UTC

I'm not explicitly decoding anything in my code -- I just call Net::Async::FastCGI::Request->read_stdin. However, reading through that I see that THAT code is explicitly calling a UTF8 decode (it has a default encoding set to UTF8).

So it looks like the correct solution is to explicitly set the default encoding to undef on the FastCGI object. Thanks all for the clues to get here ... this was not at all obvious to me.

[reply]
[d/l]