Re: apr_file_open() and non blocking IO

2010-02-28 Thread Dan Poirier
On Sat, 27 Feb 2010 16:19 +0200, Graham Leggett minf...@sharp.fm wrote: I am trying to open a fifo on disk with the O_NONBLOCK flag set, and I notice that apr's apr_file_open() cannot do this. As a result, if nobody else is already reading from the fifo, apr_file_open blocks, and this is

Re: apr_file_open() and non blocking IO

2010-02-28 Thread Graham Leggett
On 28 Feb 2010, at 2:20 PM, Dan Poirier wrote: I am trying to open a fifo on disk with the O_NONBLOCK flag set, and I notice that apr's apr_file_open() cannot do this. As a result, if nobody else is already reading from the fifo, apr_file_open blocks, and this is the behaviour I want to

apr_file_open() and non blocking IO

2010-02-27 Thread Graham Leggett
Hi all, I am trying to open a fifo on disk with the O_NONBLOCK flag set, and I notice that apr's apr_file_open() cannot do this. As a result, if nobody else is already reading from the fifo, apr_file_open blocks, and this is the behaviour I want to avoid. Does anyone know of any reason

Re: apr_file_open() and non blocking IO

2010-02-27 Thread William A. Rowe Jr.
On 2/27/2010 8:19 AM, Graham Leggett wrote: Hi all, I am trying to open a fifo on disk with the O_NONBLOCK flag set, and I notice that apr's apr_file_open() cannot do this. As a result, if nobody else is already reading from the fifo, apr_file_open blocks, and this is the behaviour I want

Re: apr_file_open() and non blocking IO

2010-02-27 Thread Graham Leggett
On 27 Feb 2010, at 11:08 PM, William A. Rowe Jr. wrote: Does anyone know of any reason why this wouldn't work? Yes - principal of least surprise. You cannot deprecate an API in this way, and the patch deprecating that flag would be rejected anyways. When you modify a library, please try