/// <summary>Loads texts and metadata (revision ID, timestamp, last comment,
/// last contributor, minor edit mark) for pages in this PageList.
/// Non-existent pages will be automatically removed from the PageList.
/// Please, don't use this function when going to edit big amount of pages on
/// popular public wikis, as it compromises edit conflict detection. In that case,
/// each page's text should be loaded individually right before its processing
/// and saving.</summary>
public void LoadWithMetadata()
{
if (IsEmpty())
throw new WikiBotException(Bot.Msg("The PageList is empty. Nothing to load."));
Console.WriteLine(Bot.Msg("Loading {0} pages..."), pages.Count);
string res = site.indexPath + "?title=Special:Export&action=submit";
string postData = "curonly=True&pages=";
foreach (Page page in pages)
postData += HttpUtility.UrlEncode(page.title) + "\r\n";
string src = site.PostDataAndGetResult(res, postData);
XmlReader reader = XmlReader.Create(new StringReader(src));
PageList pl = new PageList(site);
while (reader.ReadToFollowing("page")) {
Page p = new Page(site);
p.ParsePageXml(reader.ReadOuterXml());
pl.Add(p);
}
reader.Close();
if (pages.Count > 0) {
Clear();
pages = pl.pages;
return;
}
else { // FALLBACK, use alternative parsing way, XPath
Console.WriteLine(
Bot.Msg("XML parsing failed, switching to alternative parser..."), pages.Count);
src = Bot.RemoveXMLRootAttributes(src);
StringReader strReader = new StringReader(src);
XPathDocument doc = new XPathDocument(strReader);
strReader.Close();
XPathNavigator nav = doc.CreateNavigator();
foreach (Page page in pages) {
if (page.title.Contains("'")) { // There's no good way to escape "'" in XPath
page.LoadWithMetadata();
continue;
}
string query = "//page[title='" + page.title + "']/";
try {
page.text =
nav.SelectSingleNode(query + "revision/text").InnerXml;
}
catch (System.NullReferenceException) {
continue;
}
page.text = HttpUtility.HtmlDecode(page.text);
page.pageId = nav.SelectSingleNode(query + "id").InnerXml;
try {
page.lastUser = nav.SelectSingleNode(query +
"revision/contributor/username").InnerXml;
page.lastUserId = nav.SelectSingleNode(query +
"revision/contributor/id").InnerXml;
}
catch (System.NullReferenceException) {
page.lastUser = nav.SelectSingleNode(query +
"revision/contributor/ip").InnerXml;
}
page.lastUser = HttpUtility.HtmlDecode(page.lastUser);
page.revision = nav.SelectSingleNode(query + "revision/id").InnerXml;
page.lastMinorEdit = (nav.SelectSingleNode(query +
"revision/minor") == null) ? false : true;
try {
page.comment = nav.SelectSingleNode(query + "revision/comment").InnerXml;
page.comment = HttpUtility.HtmlDecode(page.comment);
}
catch (System.NullReferenceException) {;}
page.timestamp =
nav.SelectSingleNode(query + "revision/timestamp").ValueAsDateTime;
}
if (string.IsNullOrEmpty(pages[0].text)) { // FALLBACK 2, load pages one-by-one
foreach (Page page in pages)
page.LoadWithMetadata();
}
}
}
/// <summary>Finds all internal wikilinks in page text, excluding interwiki
/// links, links to sister projects, categories, embedded images and links in
/// image descriptions.</summary>
/// <returns>Returns the PageList object, in which page titles are the wikilinks,
/// found in text.</returns>
public PageList GetLinks()
{
MatchCollection matches = Site.wikiLinkRE.Matches(text);
StringCollection exclLinks = new StringCollection();
exclLinks.AddRange(GetInterWikiLinks());
exclLinks.AddRange(GetSisterWikiLinks(true));
string str;
int fragmentPosition;
PageList pl = new PageList(site);
for(int i = 0; i < matches.Count; i++) {
str = matches[i].Groups[1].Value;
if (str.StartsWith(site.namespaces["6"] + ":", true, site.langCulture) ||
str.StartsWith(Site.wikiNSpaces["6"] + ":", true, site.langCulture) ||
str.StartsWith(site.namespaces["14"] + ":", true, site.langCulture) ||
str.StartsWith(Site.wikiNSpaces["14"] + ":", true, site.langCulture))
continue;
str = str.TrimStart(':');
if (exclLinks.Contains(str))
continue;
fragmentPosition = str.IndexOf("#");
if (fragmentPosition != -1)
str = str.Substring(0, fragmentPosition);
pl.Add(new Page(site, str));
}
return pl;
}
/// <summary>Finds all wikilinks in page text, excluding interwiki
/// links, categories, embedded images and links in
/// image descriptions.</summary>
/// <returns>Returns the PageList object, in which page titles are the wikilinks,
/// found in text.</returns>
public PageList GetLinks()
{
MatchCollection matches = site.regexes["wikiLink"].Matches(text);
var exclLinks = GetSisterwikiLinks();
exclLinks.AddRange(GetInterLanguageLinks());
string str;
int fragmentPosition;
PageList pl = new PageList(site);
for(int i = 0; i < matches.Count; i++) {
str = matches[i].Groups["title"].Value;
if (str.StartsWith(site.GetNsPrefix(6), true, site.langCulture) || // image
str.StartsWith(site.GetEnglishNsPrefix(6), true, site.langCulture) ||
str.StartsWith(site.GetNsPrefix(14), true, site.langCulture) || // category
str.StartsWith(site.GetEnglishNsPrefix(14), true, site.langCulture))
continue;
str = str.TrimStart(':');
if (exclLinks.Contains(str))
continue;
fragmentPosition = str.IndexOf("#");
if (fragmentPosition != -1)
str = str.Substring(0, fragmentPosition);
pl.Add(new Page(site, str));
}
return pl;
}
