K kclone(K a)//Deep copy -- eliminate where possible
{
if(!a) R 0;
I t=a->t,n=a->n;
K z= 7==t?Kv():newK(t,n);
if (4==ABS(t)) DO(n, kS(z)[i]=kS(a)[i]) //memcpy everywhere is better
else if(3==ABS(t)) DO(n, kC(z)[i]=kC(a)[i])
else if(2==ABS(t)) DO(n, kF(z)[i]=kF(a)[i])
else if(1==ABS(t)) DO(n, kI(z)[i]=kI(a)[i])
else if(0== t ) DO(n, kK(z)[i]=kclone(kK(a)[i]))
else if(5== t ) DO(n, kK(z)[i]=kclone(kK(a)[i]))
else if(7== t )
{
I k=0;
z->t=a->t;
I vt=z->n = a->n;
K kv;
V*v;
SW(vt)
{
CS(1, k=((K)kV(a)[CODE])->n-1;
M(z,kv=newK(-4,k+1))
v=(V*)kK(kv);
//v[k]=0;//superfluous reminder
DO(k, V w=kW(a)[i];
if(VA(w))v[i]=w; //TODO: is this ok for NAMES? see similar code in capture()
else
{
K r=kclone(*(K*)w); //oom
V q=newE(LS,r); //oom
kap((K*) kV(z)+LOCALS,&q);//oom
cd(q);//kap does ci
q=EVP(q); //oom free z etc. kap needs checking
v[i]=q;
}
)
)
CS(2, M(z,kv=newK(-4,3))
v=(V*)kK(kv);
memcpy(v,kW(a),3*sizeof(V));
)
/*
* Read the graph from a file and store it in a Graph named G
*/
Graph readGraph(char * filename){
FILE *fp;
Graph G;
char linha[BUFFSIZE];
int head, tail, i;
// Opening the file with a 'read' flag
if ((fp = fopen(filename, "r")) == NULL){
printf("File not found\n");
exit(1);
}
// Creating graph with no nodes
G = (Graph) malloc(sizeof(struct graph));
G->V = 0;
G->E = 0;
G->n_flow = 0;
for(i = 0; i < NETSIZE; i++){
G->flow[i] = 0;
G->list[i].pred = -1;
G->list[i].n = NULL;
G->list[i + NETSIZE].pred = -1;
G->list[i + NETSIZE].n = NULL;
}
// Node addition from file input
while(fgets(linha, BUFFSIZE, fp) != NULL){
if(sscanf(linha, "%d %d", &tail, &head) == 2)
graphInsertE(G, newE(tail, head));
else
printf("Invalid line format\n");
}
// Closing file
fclose(fp);
return G;
}
string WebCrawler::crawl(string & webpage, string & stopWordsFile)
{
URL startURL(webpage,"");
string currentPageContents;
try
{
currentPageContents = downloader->download(startURL);
}
catch(CS240Exception &e)
{
string newString = e.GetMessage();
newString += " Invalid Start URL";
CS240Exception newE(newString);
throw newE;
}
cout << "Crawling ... " << endl;
string currentPageDescription = parser->findDescription(currentPageContents);
currentPage = new Page(currentPageDescription, startURL);
queue->AddPage(currentPage);
history->AddPage(currentPage);
while (queue->HasNext())
{
currentPage = queue->GetNext();
URL currentURL(currentPage->GetURL());
currentPageContents = downloader->download(currentURL);
cout <<" Crawling: "<<currentURL.GetResolvedURL() << endl;
//Note: findLinks adds all the links in currentPageContents to the queue, if they
// are not already in the history. If a page is added to the queue, it is also
// added to the history.
URL * newURLstoCrawl = parser->findLinks(currentPageContents,currentPage);
int numOfURLs = parser->countLinks(currentPageContents);
processNewLinks(newURLstoCrawl, numOfURLs, startURL);
delete [] newURLstoCrawl;
//Note: findWords, finds words and adds them to the index.
parser->findWords(currentPageContents, index, currentPage);
}
//Note: the xmlPrinter will not print words that are in the stop word file.
cout << "Generating Output File..." << endl;
return xmlPrinter->print(index, stopWordsFile, history);
}
