Re: cvs: php4 /ext/standard file.c

> Your API has other problems, that some would consider to be worse (data > being lost without any ability to know that it was lost). I'm not proposing an API. I actually acknowledged earlier that it might be better to signal a failure in the case of a failed write in a single place. > For some odd reason, most major packages I've seen to date use one > implementation or another of full_write(). I guess they didn't bother to > educate themselves either? As I already said earlier, "feel free to prove me incorrect". My reading of the standard indicates that writing such an API would require ignoring SIGXFSZ. In the case of PHP, I don't think it is feasible to temper with signals though, and that is why I think it is impossible to do in the given scope. - Sascha