On 24/11/2010, at 13:52, Jorge wrote: > On 24/11/2010, at 13:24, Marc Lehmann wrote: >> On Wed, Nov 24, 2010 at 10:46:27AM +0100, Jorge <[email protected]> >> wrote: >>> You're right, I know, there's no problem with O_APPEND, nevertheless a fix >>> for write()s to non-O_APPEND fds would be a Good Thing™ for Mac users, >>> don't you think so ? >> >> yes, thats why apple would need to get involved, as they are the only ones >> who can fix it. >> >> another thing that comes to mind is read protection - if write has these >> issues, I bet read is similarly broken. > > No, it isn't, as far as I can see.
Damn it, you're right, yes it's broken too : https://gist.github.com/713677 $ curl 'https://gist.github.com/raw/713677/206ed0924e9a0cf5a1844029d6dae0221a34bdfe/test-read-a-lot.js' > rbug.js $ node rbug.js -> ERROR: READ MORE THAN ONCE -> 0000 ERROR: READ MORE THAN ONCE -> 0000 ERROR: READ MORE THAN ONCE -> 0000 ERROR: READ MORE THAN ONCE -> 3894 ERROR: READ MORE THAN ONCE -> 5295 ERROR: READ MORE THAN ONCE -> 5896 ERROR: READ MORE THAN ONCE -> 5911 ERROR: READ MORE THAN ONCE -> 7739 ERROR: READ MORE THAN ONCE -> 7781 ERROR: READ MORE THAN ONCE -> 7793 ERROR: READ MORE THAN ONCE -> 7808 ERROR: READ MORE THAN ONCE -> 7823 ERROR: READ MORE THAN ONCE -> 7832 ERROR: READ MORE THAN ONCE -> 7848 ERROR: READ MORE THAN ONCE -> 8204 -- Jorge. _______________________________________________ libev mailing list [email protected] http://lists.schmorp.de/cgi-bin/mailman/listinfo/libev
