Apple

Someone pointed out the fpurge() function. Calling fpurge() on the reader before the 2^nd read fixes the problem.

Norm Green

From: Norm Green [mailto:email@hidden]
Sent: Friday, January 04, 2008 1:37 PM
To: 'email@hidden'
Subject: fread/fwrite/fflush bug

I’ve found what I consider to be a stream IO bug. The code below can be compiled with g++ to easily illustrate the problem. Here’s what’s happening:

Open a file for writing (using fopen(…, “w”) ) and write some data to the file
Flush the writer using fflush()
Open a second stream to the file for reading only using fopen(…., “r”)
Read some bytes using fseek() & fread() to confirm the reader can see the data written in step 1
Have the writer update 1 byte using fwrite()
Flush the writer using fflush()
Using the read stream, read the byte written by the writer in step 5 using fseek() / fread()
You will get the wrong answer here. The read should return the byte written in step 5 but instead returns the old value written in step 1.

Is this is a bug in Darwin? Are there any work arounds?

Here’s my test case code. BTW, it passes on Solaris.

Norm Green