If that was because of me.. I think i do not get an answer because i refused to sign those agreements that now everyone wants to have before merging code etc. But the patch was most-possible trivial and in fact built upon other peoples work -- merely copy+paste thus --, and so i did not. I even posted the changeset "as Matt Caswell", but they simply did not use it as such. (I had signed the change to Apache license.) Anyway i do not mind as long as at some time there will be an interface that is as transparent as the _fetch series (and digestbyname etc before) or loadprivatekey, as i do not know what all that is, it is what the user gave me; i will do the operation, but it is nonsense that i cannot do so with those algorithms which need the entire data to walk thereover multiple times, or which have a fixed message digest. If such algorithms exist, and they do, then i need to have the ability to choose the correct code path transparently, through some interface. Or have reproducible error codes available that i can use to at least detect the other code path might be necessary. I see nothing such. But hey -- there is brute force available! I will use that!! --steffen | |Der Kragenbaer, The moon bear, |der holt sich munter he cheerfully and one by one |einen nach dem anderen runter wa.ks himself off |(By Robert Gernhardt)
