void
list_emplace_back (linked_list_t list, void* val)
{
node_t n;
n = node_create ();
n->data = val;
list_insert_back (list, n);
}
void* heap_wrap_alloc_c(struct heap_wrap* h, int size) {
int __size = size + sizeof(struct listlink);
void* buff = alloc(h->__parent_alloc, h->__parent, size);
dbg_assert(buff != NULL);
/* link the allocated block in to list */
{
struct listlink* n_link = (struct listlink*)buff;
list_insert_back(&h->allocated, n_link);
}
/* return the 'real' address */
return (char*)buff + sizeof(struct listlink);
}
void* heap_wrap_alloc_v(struct heap_wrap* h, int size, const char* file ,int line) {
int __size = size + sizeof(struct listlink);
//void* buff = alloc(h->__alloc, h->__parent, __size);
void* buff = h->__parent_alloc(h->__parent, __size, file, line);
dbg_assert(buff != NULL);
/* link the allocated block in to list */
{
struct listlink* n_link = (struct listlink*)buff;
list_insert_back(&h->allocated, n_link);
}
/* return the 'real' address */
buff = (void*)((char*)buff + sizeof(struct listlink));
return buff;
}
struct list* list_remove(struct list* l, int x, int y)
{
struct list* l_or = l;
struct list* l_new = list_new();
while(l != NULL)
{
if(l->x != x || l->y != y) {
l_new = list_insert_back(l_new, l->x, l->y, l->type, l->data);
}
l = l->next;
}
l_or = list_delete(l_or);
return l_new;
}
/*=============================================================================
Function route_packet
Purpose: routes a packet...that is it decrements the ttl and places it on the
appropriate outgoing queue if packet is valid, a route is found,
the queue has room, and the packet has not timed out
Parameters:
*buf (IN) - the buffer containing the packet to route
len (IN) - the length of the buffer
Returns: nothing
=============================================================================*/
void route_packet(char *buf, int len) {
struct packet_entry *pe;
pthread_t *new_st; /* new send thread */
struct ip_hdr *ip = (struct ip_hdr *)buf;
int idx, qsize;
printf("Routing received packet:\n");
#ifdef DEBUG
printf("recv'ed (%d): ",len);
print_packet(buf,len);
#endif
if (len<sizeof(struct ip_hdr)) {
printf("Invalid packet, packet discarded.\n");
return;
}
if (ip->ttl<=1) {
printf("Discarding timed out packet.\n");
return;
}
ip->ttl-=1;
printf("ttl decremented to %d decimal, 0x%02x hex.\n",ip->ttl,ip->ttl);
if ((idx = iface_match(ip->destadd))==-1)
printf("No matching route found, packet discarded.\n");
else {
pthread_mutex_lock(ifaces[idx].qmut);
qsize = ifaces[idx].sendq->size;
if (qsize<MAX_QUEUE) {
printf("Sending packet on iface with port %d.\n",ifaces[idx].lport);
pe = packet_entry_init(buf,len);
list_insert_back(ifaces[idx].sendq,pe);
} else {
printf("Send queue full for iface with port %d, packet discarded.\n",
ifaces[idx].lport);
}
pthread_mutex_unlock(ifaces[idx].qmut);
if (!qsize) { /* queue was empty, thread must be started */
new_st = (pthread_t *)malloc(sizeof(pthread_t));
if (pthread_create(new_st,NULL,send_packet,&ifaces[idx])!=0)
error("pthread_create");
}
}
}
/*=============================================================================
Function process_hrefs
Purpose: processes the hrefs on a page adding them to the end of the queue
Parameters:
*line (IN) - line to process for hrefs
*server_context (IN) - server part of URL containing line
*full_context (IN) - server and path part of URL containing line
context_depth (IN) - the depth in the spider of the page containing line
Returns: adds found URLs from href tags to the end of the queue
=============================================================================*/
void process_hrefs(char *line, char *server_context, char *full_context,
int context_depth) {
int len;
char *p = line;
char url[MAX_URL];
while ((p = strcasestr(p,"a href="))) {
p += strlen("a href=");
p += strspn(p," '\""); /* skip anything before the URL */
len = strcspn(p," '\">"); /* find the termination of the URL */
if (*p=='#' || strncasecmp(p,"mailto:",strlen("mailto:"))==0) return;
url[0] = 0; /* prepare for strncat */
/* three cases for URL, full, relative to server, relative to context */
if (strncasecmp(p,"http://",strlen("http://"))==0 && len+1<MAX_URL) {
/* full URL and length OK */
strncat(url,p,len);
} else if (*p=='/' && strlen(server_context)+len+1<MAX_URL) {
/* relative to server URL and length OK */
strncat(url,server_context,strlen(server_context));
strncat(url,p,len);
} else if (strlen(full_context)+len+1<MAX_URL) {
/* relative to context URL and length OK */
strncat(url,full_context,strlen(full_context));
strncat(url,p,len);
} else {
/* URL too long for our MAX_URL */
return;
}
/* try to put URL in the trie with allow duplicates = false */
if (trie_insert(url_trie,url,0)) {
list_insert_back(url_entries,url_entry_init(url,context_depth+1));
#ifdef DEBUG
printf("process_hrefs: inserting new URL: %s\n",url);
#endif
} else {
#ifdef DEBUG
printf("process_hrefs: rejecting duplicate URL: %s\n",url);
#endif
}
}
}
static void heap_buddy_expand_memory(struct heap_buddy* pheap, int expand_size) {
struct block_c_pool* n_blk_pool = (struct block_c_pool*)
alloc(pheap->__parent_alloc, pheap->__parent, sizeof(struct block_c_pool));
void* n_mem_begin =
alloc(pheap->__parent_alloc, pheap->__parent, expand_size);
dbg_assert(n_mem_begin != NULL);
{
/* initialize the new memory */
struct block_c *sent_first = NULL;
struct block_c *sent_last = NULL;
struct block_c *init_block = NULL;
void* n_mem_end = (void*)((char*)n_mem_begin + expand_size);
init_block = block_com_make_sentinels(
n_mem_begin, n_mem_end, &sent_first, &sent_last);
block_com_set_free(init_block, true);
n_blk_pool->memory = n_mem_begin;
n_blk_pool->size = expand_size;
n_blk_pool->bc_first = init_block;
n_blk_pool->bc_front_sent = sent_first;
n_blk_pool->bc_end_sent = sent_last;
}
/* link the new memory into memory list */
list_insert_back(&pheap->memlist, &n_blk_pool->link);
/* insert the new block into the heap */
{
struct heap_buddy_block *n_block =
container_of(n_blk_pool->bc_first, struct heap_buddy_block, common);
int sz = block_com_data_size(n_blk_pool->bc_first);
blink_push(&pheap->buddy[mlog2(sz)], &n_block->link);
}
}
/*=============================================================================
Function main
Purpose: main entry point of program, processes input and starts search
Parameters:
argc (IN) - count of command line arguments
*argv (IN) - array of pointers to command line arguments
Returns: nothing
=============================================================================*/
int main(int argc, char *argv[]) {
int c;
char *p;
struct url_entry *ue; /* used for processing url_entries */
/* first process command line optional arguments */
while(--argc>0 && (*++argv)[0]=='-')
while((c = *++argv[0]))
switch(c) {
case 'i':
case_independent = 1;
break;
case 'l':
only_show_url = 1;
break;
case 'v':
show_non_matching = 1;
break;
case 'd':
if (*++argv[0]!='=') {
fprintf(stderr,"Illegal depth value, argument will be ignored\n");
break;
}
search_depth = strtol(++argv[0],&p,0);
if (search_depth<0 || errno==ERANGE || p==argv[0]) {
fprintf(stderr,"Illegal depth value, argument will be ignored\n");
search_depth = DEFAULT_SEARCH_DEPTH;
argv[0] = p-1; /* above ++ will point to first unconverted */
break;
}
argv[0] = p-1; /* so above ++ will point to first unconverted */
break;
case 't':
if (*++argv[0]!='=') {
fprintf(stderr,"Illegal connect timeout value, "
"argument will be ignored\n");
break;
}
connect_timeout = strtol(++argv[0],&p,0);
if (connect_timeout<0 || errno==ERANGE || p==argv[0]) {
fprintf(stderr,"Illegal connect timeout value, "
"argument will be ignored\n");
connect_timeout = DEFAULT_CONNECT_TIMEOUT;
argv[0] = p-1; /* above ++ will point to first unconverted */
break;
}
argv[0] = p-1; /* so above ++ will point to first unconverted */
break;
default:
fprintf(stderr,"Unknown option %c ignored\n",c);
}
/* now process pattern URL+ if they are present */
if (argc<2) {
fprintf(stderr, "Usage: websearch [OPTION...] pattern URL+\n"
"valid options:\n"
" -i\tspecifies case-independent matching\n"
" -l\tspecifies show matching URLs, not lines\n"
" -v\tspecifies show non-matching lines or URLs\n"
" -d=N\tspecifies search depth, default = 0\n"
" -t=N\tspecifies timeout in secs, default = 0 (no timeout)\n");
exit(1);
}
pattern = argv[0]; /* store pattern */
url_entries = list_init(); /* create url entries list */
url_trie = trie_init(); /* create url trie */
/* add all URLs to the URL entries with depth 0 */
while (*++argv) {
ue = url_entry_init(argv[0],0);
list_insert_back(url_entries,ue);
}
#ifdef DEBUG
printf("Case independence: %d\n",case_independent);
printf("Only show url: %d\n",only_show_url);
printf("Show non mathcing: %d\n",show_non_matching);
printf("Search depth: %d\n",search_depth);
printf("Connect timeout: %d\n",connect_timeout);
printf("Pattern: %s\n",pattern);
printf("URL Count: %d\n",url_entries->size);
#endif
/* keep processing an entry at a time until the list is empty */
while(url_entries->size>0) {
ue = list_remove_front(url_entries);
process_url_entry(ue);
url_entry_free(ue);
}
return 0;
}