/// <summary>
/// Looks up page in database
/// </summary>
/// <param name="ns">Namespace to look page up in</param>
/// <param name="title">Title of page to look up</param>
/// <param name="followRedirection">Whether or not to look up destination of redirect pages</param>
/// <param name="recoverSections">Whether or not to recover the section structure of the page</param>
/// <param name="readLinks">Read link information</param>
/// <param name="readTFTable">Read term frequency table</param>
/// <returns>Page instance</returns>
public Page LookupPage(WikiDB.Namespace ns, string title, bool followRedirection,
bool recoverSections, bool readLinks, bool readTFTable)
{
string url = GetURLFromPageTitle(title);
// check page cache
if (_pageCache.ContainsKey(url))
{
// see if page is dirty
Page cached = (Page)_pageCache[url];
bool dirty = (recoverSections && !cached.SectionsRecovered) ||
(readLinks && cached.WikiLinks.Count == 0) ||
(readTFTable && cached.TermFrequencies.Count == 0);
if (!dirty)
{
return(cached);
}
else
{
_pageCache.Remove(url);
}
}
if (!CheckConnection(true))
{
throw new Exception("Could not establish connection with Wikipedia database");
}
Page p = null;
url = MySQLEscape(url);
int nsVal = NamespaceValue(ns);
// get text and redirect page
string selectCols = "page_id, page_text, redirects_to";
if (recoverSections)
{
selectCols += ", section_layout";
}
string query = "SELECT " + selectCols + " FROM page " +
"WHERE page_namespace=" + nsVal + " AND page_title=\"" + url + "\"";
IDataReader reader = SubmitQuery(query);
if (reader.Read())
{
int id = int.Parse(reader["page_id"].ToString());
string text = Encoding.UTF8.GetString((byte[])reader["page_text"]);
string redirectsTo = reader["redirects_to"].ToString();
// split into lines
List <string> lines = new List <string>(text.Split(new char[] { '\n' }, StringSplitOptions.RemoveEmptyEntries));
if (recoverSections)
{
string sectionLayout = reader["section_layout"].ToString();
p = new Page(title, ns, id, lines, sectionLayout);
}
else
{
p = new Page(title, ns, id);
p.Lines = lines;
// add a single section to the page
Section s = new Section("full page section", lines, null, p, 0, lines.Count - 1);
p.Sections.Add(s);
}
reader.Close();
// check for page redirection
Page redirectPage = null;
if (redirectsTo != "" && followRedirection)
{
redirectPage = LookupPage(ns, redirectsTo, followRedirection, recoverSections, readLinks, readTFTable);
}
p.RedirectsTo = redirectPage;
}
else
{
reader.Close();
return(null);
}
reader.Close();
// get links
if (readLinks)
{
query = "SELECT link_list FROM pagelinks " +
"WHERE page_namespace=" + nsVal + " AND page_title=\"" + url + "\"";
reader = SubmitQuery(query);
if (reader.Read())
{
string linkList = Encoding.UTF8.GetString((byte[])reader["link_list"]);
string[] splitList = linkList.Split(new char[] { '\n' }, StringSplitOptions.RemoveEmptyEntries);
foreach (string link in splitList)
{
string[] entry = link.Split(' ');
if (entry.Length != 2)
{
throw new Exception("Invalid link entry");
}
string destPage = entry[0];
string sourceSectionID = recoverSections ? entry[1] : "1";
Section s = p.GetSection(sourceSectionID);
WikiLink wl = new WikiLink("[[" + destPage + "]]", destPage, "", destPage, s);
s.AddLink(wl);
}
}
reader.Close();
}
// get TFs
if (readTFTable)
{
query = "SELECT freq_list FROM termfreqs WHERE page_namespace=" + nsVal + " AND page_title=\"" + url + "\"";
reader = SubmitQuery(query);
if (reader.Read())
{
string freqList = Encoding.UTF8.GetString((byte[])reader["freq_list"]);
string[] splitList = freqList.Split(new char[] { '\n' }, StringSplitOptions.RemoveEmptyEntries);
// get freqs
for (int i = 0; i < splitList.Length; ++i)
{
string[] entry = splitList[i].Split(' ');
if (entry.Length != 2)
{
throw new Exception("Invalid frequency entry");
}
string word = entry[0];
float freq = float.Parse(entry[1]);
if (p.TermFrequencies.ContainsKey(word))
{
throw new Exception("Duplicate TF entry");
}
p.TermFrequencies[word] = freq;
}
}
reader.Close();
}
// add page to cache
CachePage(p);
return(p);
}
