private void BindData()
{
DataSet ds = new d.LinkData().GetOutboundLinksP(getId(), pager.PageSize, pager.CurrentPage, dgrPages.SortExpression);
dgrPages.DataSource = ds.Tables[0];
dgrPages.DataBind();
pager.AdjustAfterBinding(Convert.ToInt32(ds.Tables[1].Rows[0][0]));
}
private void load()
{
DataTable docs = new d.DocData().GetLinkCounts();
inboundLinks = new Hashtable();
outboundLinks = new Hashtable();
int id;
foreach (DataRow dr in docs.Rows)
{
id = Convert.ToInt32(dr[0]);
inboundLinks.Add(id, new int[Convert.ToInt32(dr[1])]);
outboundLinks.Add(id, new int[Convert.ToInt32(dr[2])]);
}
d.LinkData linkData = new d.LinkData();
DataTable dt = linkData.GetLinksSortByTo();
int currId = -1;
int cursor = 0;
int toid;
int fromid;
int[] currLinks = null;
foreach (DataRow dr in dt.Rows)
{
toid = Convert.ToInt32(dr[0]);
fromid = Convert.ToInt32(dr[1]);
if (currId < toid)
{
cursor = 0;
currId = toid;
currLinks = (int[])inboundLinks[toid];
}
currLinks[cursor++] = fromid;
}
dt = linkData.GetLinksSortByFrom();
currId = -1;
cursor = 0;
currLinks = null;
foreach (DataRow dr in dt.Rows)
{
fromid = Convert.ToInt32(dr[0]);
toid = Convert.ToInt32(dr[1]);
if (currId < fromid)
{
cursor = 0;
currId = fromid;
currLinks = (int[])outboundLinks[fromid];
}
currLinks[cursor++] = toid;
}
}
public void ConvertLinkTable()
{
DataTable dt = new d.LinkData().GetAll();
string fromid;
string toid;
string text;
foreach (DataRow dr in dt.Rows)
{
StringBuilder sb = new StringBuilder();
fromid = dr[1].ToString();
toid = dr[2].ToString();
text = dr[3].ToString();
sb.AppendFormat("{0} ", fromid);
sb.AppendFormat("{0} ", toid);
sb.AppendFormat("[[{0}]]", text);
Console.WriteLine(sb.ToString());
}
}
public void Run()
{
UrlHelper urlHelper = new UrlHelper();
Regex regex = makeRegex();
MatchCollection mc;
Regex rSpace = new Regex(@"\s");
d.DocData pd = new d.DocData();
d.LinkData ld = new d.LinkData();
Uri baseUri;
Uri childUri;
FileInfo fi;
d.Doc p;
int pageId;
StreamReader sr;
string html;
string linkToProcess;
int linkId;
string linkText;
string path = Helper.DOCS_PATH;
string[] files = Directory.GetFiles(path);
for (int i = 545; i < files.Length; i++) //545 already done
{
Console.WriteLine("processing file #" + i);
fi = new FileInfo(files[i]);
pageId = Convert.ToInt32(fi.Name.Substring(0, fi.Name.IndexOf(".")));
p = new d.Doc(pageId);
baseUri = new Uri(p.Url);
sr = new StreamReader(fi.OpenRead());
html = sr.ReadToEnd();
mc = regex.Matches(html);
Console.WriteLine("found " + mc.Count + " links");
foreach (Match m in mc)
{
try
{
childUri = new Uri(baseUri, urlHelper.MakeLink(m.Groups[3].ToString()));
linkToProcess = urlHelper.NormalizeUrl(childUri.AbsoluteUri);
linkText = m.Groups[4].ToString();
linkId = pd.GetIdByUrl(linkToProcess);
if (linkId > 0 && linkText != "") //found page!
{
linkText = rSpace.Replace(linkText, " ");
linkText = linkText.Trim();
linkText = linkText.Replace(" ", " ");
linkText = linkText.Replace(" ", " ");
linkText = linkText.Replace(" ", " ");
linkText = linkText.Replace(" ", " ");
linkText = linkText.Replace(" ", " ");
linkText = linkText.Replace(" ", " ");
linkText = linkText.Replace(" ", " ");
linkText = linkText.Replace(" ", " ");
linkText = linkText.Replace(" ", " ");
ld.UpdateText(pageId, linkId, linkText);
}
}
catch (Exception) {}
}
//if (i % 100 == 0)
// Console.WriteLine("processing file #" + i);
}
}
public void AddAnchorText()
{
d.StopList stopList = new d.StopList();
ParseHelper parseHelper = new ParseHelper();
char[] delims = parseHelper.GetDelims();
PorterStemmer stemmer = new PorterStemmer();
d.LinkData ld = new d.LinkData();
d.TermDocData tdd = new d.TermDocData();
DataTable linksTable;
int docId;
StringBuilder sb;
string[] terms;
string term;
Hashtable currTerms;
DataTable dt = new d.DocData().GetIds();
for (int i = 0; i < dt.Rows.Count; i++)
{
if (i % 10 == 0)
{
Console.WriteLine(i);
}
//accumulate all link text for this doc into StringBuilder
sb = new StringBuilder();
docId = (int)dt.Rows[i][0];
linksTable = ld.GetRecordsByToId(docId);
foreach (DataRow dr in linksTable.Rows)
{
sb.AppendFormat("{0} ", dr[0].ToString());
}
//accum terms + counts into currTerms hashtable
currTerms = new Hashtable();
terms = sb.ToString().Split(delims);
for (int j = 0; j < terms.Length; j++)
{
term = stemmer.stemTerm(terms[j].ToLower().Trim());
if (term != "home" && term.Length > 0 && term.Length < 25 && !stopList.Contains(term) && parseHelper.IsAsciiLetters(term))
{
if (!currTerms.Contains(term))
{
currTerms.Add(term, 1);
}
else
{
currTerms[term] = (int)currTerms[term] + 1;
}
}
}
//write terms and counts to database
IDictionaryEnumerator en = currTerms.GetEnumerator();
string currTerm;
int currCount;
while (en.MoveNext())
{
currTerm = en.Key.ToString();
currCount = (int)currTerms[currTerm];
tdd.UpdateAnchorTextCount(currTerm, docId, currCount);
}
}
}