client. Failure to release the GIL in the latter case can result in a
daemon process deadlocking, thus freezing the process, when a large request
content is received which is not consumed before a response is generated.
Set timeout on socket used to communicate to daemon process, for the period
when request content being sent to the daemon process. The timeout value
is taken from standard Apache timeout set by Apache Timeout directive.
This ensures that socket doesn't get stuck forever when deadlock ensues
from large request content being sent to handler that returns a response
without consuming the request content.

was of incorrect type and error was beign raised. Raised version string on
1.X branch to 1.3-TRUNK at same time.

causing status of sub process created by popen() to not be caught.

Apache 2.2, but require the Apache 2.2 behaviour. Thus duplicate this function
from Apache 2.2 and use that copy for all versions of Apache. Doing so fixes
problem whereby wrong application group (interpreter) can be selected when
relying on default calculation for application group and repeating slashes
appear in URL after the mount point of the application.

removed in between the time that Apache matched request to it and the WSGI
script file was loaded and the request passed to it. Code was also incorrectly
logging via request object in a few cases before fake request object in
daemon process had been constructed properly.

empty yeilded string to conform to WSGI specification. Fix crash when calling
start_response() with exc_info and a non empty string has already been sent.

mode support.