On Sun, 6 Mar 2005 16:08:47 +0800 "Jin Tao" <[EMAIL PROTECTED]> wrote:
> 1. Macro BM_MIN_LENGTH and BM_BLOCK_SIZE are both defined value 3. So > this loop is actually run only once every time, why we use a loop > structure here? Because I've decided to leave the original structure for educational purposes. The loop is eliminated by a compiler. Moreover the value of BM_MIN_LENGTH will change when I implement regularity classifier mentioned in TODO at the top of the source file. > 2. And bm_shift[idx] is initialized as 1 in cli_bm_init() function. The initial maximal shift is BM_MIN_LENGTH - BM_BLOCK_SIZE + 1 (i.e. 1 in our case) and it's the value of a maximal possible shift. > Hence, the value of root->bm_shift[idx] must be larger than > BM_MIN_LENGTH - BM_BLOCK_SIZE - i, which is negative. This value will never be negative because BM_MIN_LENGTH >= BM_BLOCK_SIZE and the initial i = BM_MIN_LENGTH - BM_BLOCK_SIZE and it's _decreasing_. BM_MIN_LENGTH - BM_BLOCK_SIZE - i is an amount of shifting required to get to the actual substring. > Thanks a lot for your help! You're welcome, -- oo ..... Tomasz Kojm <[EMAIL PROTECTED]> (\/)\......... http://www.ClamAV.net/gpg/tkojm.gpg \..........._ 0DCA5A08407D5288279DB43454822DC8985A444B //\ /\ Sun Mar 6 15:49:39 CET 2005
pgpRs96YUts5t.pgp
Description: PGP signature
_______________________________________________ http://lurker.clamav.net/list/clamav-devel.html
