On Mon 04 Dec 2017 07:26:02 PM CET, Max Reitz wrote: >>> +static void curl_refresh_filename(BlockDriverState *bs) >>> +{ >>> + BDRVCURLState *s = bs->opaque; >>> + >>> + if (!s->sslverify || s->cookie || >>> + s->username || s->password || s->proxyusername || s->proxypassword) >>> + { >> >> Is !s->sslverify negative because that setting is true by default? > > Yes, exactly. If it's false, you'd need to override it (and you can't > do that through a plain filename).
I think this is not the only case in this series, but I'm not very comfortable with the idea that this condition and the default value of the setting are implicity dependent on each other. If you change one and forget to change the other things will break. I understand that the default value is never supposed to change so in practice I don't see this breaking, but is it perhaps worth adding tests for all these cases? Berto