I recently found a problem with the reject.dat mechanism for excluding directories during crawls. The docs mention putting lines like: http://www.serv1.com/cgi-bin/* in reject.dat to exclude all such link from a traversal. I found it was not working. In particular I was trying to exclude all user dirs from my traversal with this line in reject.dat: http://www.server2.com/~* Anyway I poked around in the code and made the following patch which seems to have fixed it up. Brian McQueen NASA/Ames NAS Division %%% Created Thu Aug 9 14:47:03 PDT 2001 by target lynx.patch. %%% diff -bru orig/lynx2-8-4/src/LYTraversal.c lynx2-8-4/src/LYTraversal.c --- orig/lynx2-8-4/src/LYTraversal.c Thu Dec 21 18:44:11 2000 +++ lynx2-8-4/src/LYTraversal.c Thu Aug 9 14:46:16 2001 @@ -153,9 +153,9 @@ while (LYSafeGets(&buffer, ifp) != NULL && !result) { len = strlen(buffer); if (len > 0) { /* if not an empty line */ - if (buffer[len-1] == '*') { + if (buffer[len-2] == '*') { /* if last char is * and the rest of the chars match */ - if ((len == 1) || (strncmp(line,buffer,len - 1) == 0)) { + if ((len == 1) || (strncmp(line,buffer,len - 2) == 0)) { result = TRUE; } } else { ; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to [EMAIL PROTECTED]
