OpenSSL::Buffering#read_nonblock
# File tmp/rubies/ruby-4.0.0/ext/openssl/lib/openssl/buffering.rb, line 203
def read_nonblock(maxlen, buf=nil, exception: true)
if maxlen == 0
if buf
buf.clear
return buf
else
return ""
end
end
if @rbuffer.empty?
return sysread_nonblock(maxlen, buf, exception: exception)
end
ret = consume_rbuff(maxlen)
if buf
buf.replace(ret)
ret = buf
end
ret
end
Reads at most maxlen bytes in the non-blocking manner.
When no data can be read without blocking it raises OpenSSL::SSL::SSLError extended by IO::WaitReadable or IO::WaitWritable.
IO::WaitReadable means SSL needs to read internally so read_nonblock should be called again when the underlying IO is readable.
IO::WaitWritable means SSL needs to write internally so read_nonblock should be called again after the underlying IO is writable.
OpenSSL::Buffering#read_nonblock needs two rescue clause as follows:
# emulates blocking read (readpartial). begin result = ssl.read_nonblock(maxlen) rescue IO::WaitReadable IO.select([io]) retry rescue IO::WaitWritable IO.select(nil, [io]) retry end
Note that one reason that read_nonblock writes to the underlying IO is when the peer requests a new TLS/SSL handshake. See openssl the FAQ for more details. www.openssl.org/support/faq.html
By specifying a keyword argument exception to false, you can indicate that read_nonblock should not raise an IO::Wait*able exception, but return the symbol :wait_writable or :wait_readable instead. At EOF, it will return nil instead of raising EOFError.