In message <[email protected]> on Mon, 04 May 2009 12:46:40 +0100, Ben Laurie <[email protected]> said:
ben> Richard Levitte wrote: ben> > In message <[email protected]> on Sun, 03 May 2009 20:13:20 +0100, Ben Laurie <[email protected]> said: ben> > ben> > ben> Richard Levitte wrote: ben> > ben> > OpenSSL CVS Repository ben> > ben> > http://cvs.openssl.org/ ben> > ben> > ____________________________________________________________________________ ben> > ben> > ben> > ben> > Server: cvs.openssl.org Name: Richard Levitte ben> > ben> > Root: /v/openssl/cvs Email: [email protected] ben> > ben> > Module: openssl Date: 26-Apr-2009 14:16:08 ben> > ben> > Branch: HEAD Handle: 2009042613160800 ben> > ben> > ben> > ben> > Modified files: ben> > ben> > openssl/apps dgst.c ben> > ben> > ben> > ben> > Log: ben> > ben> > Cast to avoid signedness confusion ben> > ben> ben> > ben> Wouldn't it be better to make len signed, if it actually is? ben> > ben> > I agree, but it's passed by address to some function taking a size_t*, ben> > which means there's gonna be a cast somewhere anyway. I simply took ben> > the easiest route. ben> ben> Then one has to wonder why it is being tested for < 0. Because it may get a negative value at one point. You can have a look for yourself, it can get its value in one of two ways: 1. Assigned through passed reference to EVP_DigestSignFinal(), which takes a size_t* as last argument. 2. Assigned the returned value from BIO_gets(), which may return a negative value and therefore returns a int. len is the used further on in a call to BIO_write(). ben> > Really, we need to harmonize our uses of size_t, int, long and ben> > so on... It's a big job. ben> ben> Yeah. I started on it once. It was a mess :-) I know, did it as well... Cheers, Richard -- Richard Levitte [email protected] http://richard.levitte.org/ "Life is a tremendous celebration - and I'm invited!" -- from a friend's blog, translated from Swedish ______________________________________________________________________ OpenSSL Project http://www.openssl.org Development Mailing List [email protected] Automated List Manager [email protected]
