Jim Jagielski wrote: > > On Sep 1, 2007, at 3:35 AM, William A. Rowe, Jr. wrote: > >> Steven Nairn wrote: >>> >>> The "testfile" failure on win32 is in test_gets_buffered() because >>> read_with_timeout() (in file_io/win32/readwrite.c) does not correctly >>> report end of file when the file is opened with the APR_FOPEN_XTHREAD >>> flag. >>> >>> The fix is to check for ERROR_HANDLE_EOF in read_with_timeout() >>> (attached patch). >> >> Very nice catch, this is applied to 0.9, 1.2 and trunk now. >> >> Thank you again Steven! >> > > Bill, are you planning to retag 0.9 with this fix? I know > I need to hold off on httpd 2.2 until the new 1.2 tag, but > if you are planning to also retag 0.9, I'd prefer to > wait until you do so for httpd 2.0...
As Mladen pointed out, we want to finish reading at EOF (e.g., we want those last bytes). I'm trying to figure out -where- they are. The IO Completion status calls aren't telling me anything just yet. I'd hope to have this wrapped around noon... still fighting. I know you are anxiously waiting to tag (although I'm frankly really glad you didn't - apparently this is one holiday weekend when folks found allot of enthusiasm for hacking at things long orphaned ;-) Bill
