2008. 01. 17, csütörtök keltezéssel 12.14-kor Ken Kixmoeller -- reply to ken@xxxxxxxxxxxxxx ezt írta: > (forgot to copy the list) > > On Jan 16, 2008, at 5:08 PM, Richard Lynch wrote: > > > > Is it possible that 4% of the time, you have spaces on the start/end > > of the string, which get trimmed before encryption? > > > > In this case, no. In trying to simplify the situation to narrow the > possibilities of error, I am generating "random" character strings of > only alphanumeric (or numeric-only) characters. Each is exactly 16 > characters. > > > > > And if rijndael is one of the algorithms which requires a fixed-size > > input, that also would be "bad" to trim it. > > > > No documentation that I was able to find suggests that requirement. > > > > > > Actually, I'd suggest that the encryption function has no business > > trimming the text anyway. > > > > Philosophically I agree with you, but mCrypt has this nasty habit of > appending bunches of nulls to the decrypted string. So philosophical > purity gives way to practical application. yeah, I just ran into the same thing yesterday evening with mcrypt and rijndael_256. encrypting went fine, decrypted string had a lot of nulls at the end. so I too had to use trim() on it. greets Zoltán Németh > > Good ideas, as usual. Thank you. > > Ken > > -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
