private void CreateCorrectionEndTagBefore(int pos)
{
string endTagName = tag.TagName;
int endTagBegin = pos;
int endTagEnd = endTagBegin + endTagName.Length + 2;
endTag = new EndTag(new TagData(endTagBegin, endTagEnd, endTagName, currLine));
}
public void TagExtraction()
{
string testHTML = "<AREA \n coords=0,0,52,52 href=\"http://www.yahoo.com/r/c1\" shape=RECT>";
CreateParser(testHTML);
Tag tag = Tag.Find(parser.Reader, testHTML, 0);
Assert.IsNotNull(tag);
}
private void DoForceCorrectionCheckOn(Tag possibleEndTagCauser)
{
if (IsEndTagMissing(possibleEndTagCauser))
{
CreateCorrectionEndTagBefore(possibleEndTagCauser);
endTagFound = true;
}
}
protected override Tag CreateTag(TagData tagData, Tag tag, string url)
{
System.Collections.Hashtable table = tag.Attributes;
string metaTagName = (string)table["NAME"];
string metaTagContents = (string)table["CONTENT"];
string httpEquiv = (string)table["HTTP-EQUIV"];
return(new MetaTag(tagData, httpEquiv, metaTagName, metaTagContents));
}
public void ExtractLinkInvertedCommasBug()
{
string tagContents = "a href=r/anorth/top.html";
Tag tag = new Tag(new TagData(0, 0, tagContents, ""));
string url = "c:\\cvs\\html\\binaries\\yahoo.htm";
LinkScanner scanner = new LinkScanner("-l");
Assert.AreEqual("r/anorth/top.html", scanner.ExtractLink(tag, url), "Extracted Link");
}
public virtual string ReplaceFaultyTagWithEndTag(Tag tag, string currentLine)
{
string newLine = currentLine.Substring(0, tag.ElementBegin);
newLine += "</" + tag.TagName + ">";
newLine += currentLine.Substring(tag.ElementEnd + 1, currentLine.Length - (tag.ElementEnd + 1));
return(newLine);
}
public void InsertEndTagBeforeTag()
{
string currentLine = "<a href=s/7509><b>Yahoo! Movies</b></a>";
Tag tag = new Tag(new TagData(0, 14, "a href=s/7509", currentLine));
LinkScanner linkScanner = new LinkScanner();
string newLine = linkScanner.InsertEndTagBeforeNode(tag, currentLine);
Assert.AreEqual("</A><a href=s/7509><b>Yahoo! Movies</b></a>", newLine, "Expected insertion");
}
private void AssertTagNameMatches(Tag nextExpectedTag, Tag nextActualTag, string displayMessage)
{
string expectedTagName = nextExpectedTag.TagName;
string actualTagName = nextActualTag.TagName;
displayMessage = "The tag names did not match: Expected " + expectedTagName + " \nbut was " + actualTagName +
displayMessage;
AssertStringEquals(displayMessage, expectedTagName, actualTagName);
}
private void CreateCorrectionEndTagBefore(Tag possibleEndTagCauser)
{
string endTagName = tag.TagName;
int endTagBegin = possibleEndTagCauser.ElementBegin;
int endTagEnd = endTagBegin + endTagName.Length + 2;
possibleEndTagCauser.TagBegin = endTagEnd + 1;
reader.AddNextParsedNode(possibleEndTagCauser);
endTag = new EndTag(new TagData(endTagBegin, endTagEnd, endTagName, currLine));
}
public virtual string ExtractFormMethod(Tag tag)
{
string method = tag["METHOD"];
if (method == null)
{
method = FormTag.GET;
}
return(method.ToUpper());
}
public void AssertTagEquals(Node expected, Node actual, string displayMessage)
{
if (expected is Tag)
{
Tag expectedTag = (Tag)expected;
Tag actualTag = (Tag)actual;
AssertTagNameMatches(expectedTag, actualTag, displayMessage);
AssertAttributesMatch(expectedTag, actualTag, displayMessage);
}
}
private int CheckBeginParsingState(int i, int state, char ch, Tag tag)
{
if (ch == '<' && (state == TAG_BEFORE_PARSING_STATE || state == TAG_ILLEGAL_STATE))
{
// Transition from State 0 to State 1 - Record data till > is encountered
tag.TagBegin = i;
state = TAG_BEGIN_PARSING_STATE;
}
return(state);
}
private int CheckIllegalState(int i, int state, char ch, Tag tag)
{
if (ch == '/' && i > 0 && tag.TagLine[i - 1] == '<' &&
state != TAG_IGNORE_DATA_STATE &&
state != TAG_IGNORE_BEGIN_TAG_STATE)
{
state = TAG_ILLEGAL_STATE;
}
return(state);
}
private void DoEmptyXmlTagCheckOn(Node currentNode)
{
if (currentNode is Tag)
{
Tag possibleEndTag = (Tag)currentNode;
if (IsXmlEndTag(tag))
{
endTag = possibleEndTag;
endTagFound = true;
}
}
}
public void ReplaceFaultyTagWithEndTag()
{
string currentLine =
"<p>Site Comments?<br><a href=\"mailto:[email protected]?subject=Site Comments\">Mail Us<a></p>";
Tag tag = new Tag(new TagData(85, 87, "a", currentLine));
LinkScanner linkScanner = new LinkScanner();
string newLine = linkScanner.ReplaceFaultyTagWithEndTag(tag, currentLine);
Assert.AreEqual(
"<p>Site Comments?<br><a href=\"mailto:[email protected]?subject=Site Comments\">Mail Us</A></p>",
newLine, "Expected replacement");
}
public void LinkDataContents()
{
CreateParser(
"<a href=\"http://transfer.go.com/cgi/atransfer.pl?goto=http://www.signs.movies.com&name=114332&srvc=nws&context=283&guid=4AD5723D-C802-4310-A388-0B24E1A79689\" target=\"_new\"><img src=\"http://ad.abcnews.com/ad/sponsors/buena_vista_pictures/bvpi-ban0003.gif\" width=468 height=60 border=\"0\" alt=\"See Signs in Theaters 8-2 - Starring Mel Gibson\" align=><font face=\"verdana,arial,helvetica\" SIZE=\"1\"><b></b></font></a>",
"http://transfer.go.com");
// Register the image scanner
LinkScanner linkScanner = new LinkScanner("-l");
parser.AddScanner(linkScanner);
parser.AddScanner(linkScanner.CreateImageScanner("-i"));
ParseAndAssertNodeCount(1);
Assert.IsTrue(node[0] is LinkTag, "Node 0 should be a link tag");
LinkTag linkTag = (LinkTag)node[0];
Assert.AreEqual(
"http://transfer.go.com/cgi/atransfer.pl?goto=http://www.signs.movies.com&name=114332&srvc=nws&context=283&guid=4AD5723D-C802-4310-A388-0B24E1A79689",
linkTag.Link, "Link URL");
Assert.AreEqual("", linkTag.LinkText, "Link Text");
Node[] containedNodes = new AbstractNode[10];
int i = 0;
foreach (Node nestedNode in linkTag)
{
containedNodes[i++] = nestedNode;
}
Assert.AreEqual(5, i, "There should be 5 contained nodes in the link tag");
Assert.IsTrue(containedNodes[0] is ImageTag, "First contained node should be an image tag");
ImageTag imageTag = (ImageTag)containedNodes[0];
Assert.AreEqual("http://ad.abcnews.com/ad/sponsors/buena_vista_pictures/bvpi-ban0003.gif", imageTag.ImageURL,
"Image Location");
Assert.AreEqual("60", imageTag["HEIGHT"], "Image Height");
Assert.AreEqual("468", imageTag["WIDTH"], "Image Width");
Assert.AreEqual("0", imageTag["BORDER"], "Image Border");
Assert.AreEqual("See Signs in Theaters 8-2 - Starring Mel Gibson", imageTag["ALT"], "Image Alt");
Assert.IsTrue(containedNodes[1] is Tag, "Second contained node should be Tag");
Tag tag1 = (Tag)containedNodes[1];
Assert.AreEqual("font face=\"verdana,arial,helvetica\" SIZE=\"1\"", tag1.Text, "Tag Contents");
Assert.IsTrue(containedNodes[2] is Tag, "Third contained node should be Tag");
Tag tag2 = (Tag)containedNodes[2];
Assert.AreEqual("b", tag2.Text, "Tag Contents");
Assert.IsTrue(containedNodes[3] is EndTag, "Fourth contained node should be HTMLEndTag");
EndTag endTag1 = (EndTag)containedNodes[3];
Assert.AreEqual("b", endTag1.Text, "Fourth Tag contents");
Assert.IsTrue(containedNodes[4] is EndTag, "Fifth contained node should be HTMLEndTag");
EndTag endTag2 = (EndTag)containedNodes[4];
Assert.AreEqual("font", endTag2.Text, "Fifth Tag contents");
}
protected override Tag CreateTag(TagData tagData, Tag tag, string url)
{
string baseUrl = tag["HREF"];
string absoluteBaseUrl = "";
if (baseUrl != null && baseUrl.Length > 0)
{
absoluteBaseUrl = LinkProcessor.RemoveLastSlash(baseUrl.Trim());
processor.BaseUrl = absoluteBaseUrl;
}
return(new BaseHrefTag(tagData, absoluteBaseUrl));
}
private void CheckIfAppendable(Bool encounteredQuery, int state, char ch, Tag tag)
{
if (state == TAG_IGNORE_DATA_STATE || state == TAG_BEGIN_PARSING_STATE ||
state == TAG_IGNORE_BEGIN_TAG_STATE)
{
if (ch == '?')
{
encounteredQuery.Boolean = true;
}
tag.Append(ch);
}
}
public CompositeTagData(Tag startTag, Tag endTag, NodeList children)
{
this.startTag = startTag;
this.endTag = endTag;
this.children = new NodeList();
if (children != null)
{
foreach (Node child in children)
{
this.children.Add(child);
}
}
}
public CompositeTagScannerHelper(CompositeTagScanner scanner, Tag tag, string url, NodeReader reader,
string currLine, bool balance_quotes)
{
this.scanner = scanner;
this.tag = tag;
this.url = url;
this.reader = reader;
this.currLine = currLine;
this.endTag = null;
this.nodeList = new NodeList();
this.endTagFound = false;
this.balance_quotes = balance_quotes;
}
public bool IsTagToBeEndedFor(Tag tag)
{
bool isEndTag = tag is EndTag;
string tagName = tag.TagName;
if ((isEndTag && endTagEnderSet.Contains(tagName)) || (!isEndTag && tagEnderSet.Contains(tagName)))
{
return(true);
}
else
{
return(false);
}
}
public static bool IsXMLTagFound(Node node, string tagName)
{
bool xmlTagFound = false;
if (node is Tag)
{
Tag tag = (Tag)node;
if (tag.Text.ToUpper().IndexOf(tagName) == 0)
{
xmlTagFound = true;
}
}
return(xmlTagFound);
}
private void DoChildAndEndTagCheckOn(Node currentNode)
{
if (currentNode is EndTag)
{
EndTag possibleEndTag = (EndTag)currentNode;
if (IsExpectedEndTag(possibleEndTag))
{
endTagFound = true;
endTag = possibleEndTag;
return;
}
}
nodeList.Add(currentNode);
scanner.ChildNodeEncountered(currentNode);
}
private void FixIfXmlEndTag(Parser parser, Node node)
{
if (node is Tag)
{
Tag tag = (Tag)node;
if (tag.EmptyXmlTag)
{
// Add end tag
string currLine = parser.Reader.CurrentLine;
int pos = parser.Reader.LastReadPosition;
currLine = currLine.Substring(0, (pos + 1) - (0)) + "</" + tag.TagName + ">" +
currLine.Substring(pos + 1, (currLine.Length) - (pos + 1));
parser.Reader.ChangeLine(currLine);
}
}
}
public virtual void CorrectTag(Tag tag)
{
string tempText = tag.Text;
StringBuilder absorbedText = new StringBuilder();
foreach (char c in tempText)
{
if (c != '"')
{
absorbedText.Append(c);
}
}
// Go into the next stage.
StringBuilder result = InsertInvertedCommasCorrectly(absorbedText);
tag.Text = result.ToString();
}
/// <summary> Extract the location of the image, given the string to be parsed, and the url
/// of the html page in which this tag exists.
/// </summary>
/// <param name="tag">String to be parsed
/// </param>
/// <param name="url">URL of web page being parsed
///
/// </param>
public virtual string ExtractImageLocn(Tag tag, string url)
{
string relativeLink = null;
try
{
table = tag.Attributes;
relativeLink = (string)table["SRC"];
if (relativeLink != null)
{
relativeLink = ParserUtils.RemoveChars(relativeLink, '\n');
relativeLink = ParserUtils.RemoveChars(relativeLink, '\r');
}
if (relativeLink == null || relativeLink.Length == 0)
{
// try fix
string tagText = tag.Text.ToUpper();
int indexSrc = tagText.IndexOf("SRC");
if (indexSrc != -1)
{
// There is a missing equals.
tag.Text = tag.Text.Substring(0, (indexSrc + 3) - (0)) + "=" +
tag.Text.Substring(indexSrc + 3, (tag.Text.Length) - (indexSrc + 3));
table = tag.RedoParseAttributes();
relativeLink = (string)table["SRC"];
}
}
if (relativeLink == null)
{
return("");
}
else
{
return(processor.Extract(relativeLink, url));
}
}
catch (System.Exception e)
{
throw new ParserException(
"HTMLImageScanner.ExtractImageLocn() : Error in extracting image location, relativeLink = " +
relativeLink + ", url = " + url, e);
}
}
public void CompositeTagWithTagChild()
{
CreateParser("<Custom>" + "<Hello>" + "</Custom>");
CustomTag customTag = ParseCustomTag(1);
int x = customTag.ChildCount;
Assert.AreEqual(1, customTag.ChildCount, "child count");
Assert.IsFalse(customTag.EmptyXmlTag, "custom tag should not be xml end tag");
Assert.AreEqual(0, customTag.StartTag.ElementBegin, "starting loc");
Assert.AreEqual(7, customTag.StartTag.ElementEnd, "ending loc");
Assert.AreEqual(0, customTag.ElementBegin, "custom tag starting loc");
Assert.AreEqual(23, customTag.ElementEnd, "custom tag ending loc");
Node child = customTag[0];
AssertType("child", typeof(Tag), child);
Tag tag = (Tag)child;
AssertStringEquals("child html", "<HELLO>", child.ToHtml());
}
public void FreshMeatBug()
{
CreateParser("<a>Revision</a>", "http://www.yahoo.com");
// Register the image scanner
parser.AddScanner(new LinkScanner("-l"));
ParseAndAssertNodeCount(3);
Assert.IsTrue(node[0] is Tag, "Node 0 should be a tag");
Tag tag = (Tag)node[0];
Assert.AreEqual("a", tag.Text, "Tag Contents");
Assert.IsTrue(node[1] is StringNode, "Node 1 should be a string node");
StringNode stringNode = (StringNode)node[1];
Assert.AreEqual("Revision", stringNode.Text, "StringNode Contents");
Assert.IsTrue(node[2] is EndTag, "Node 2 should be a string node");
EndTag endTag = (EndTag)node[2];
Assert.AreEqual("a", endTag.Text, "End Tag Contents");
}
public static string ToString(Tag tag)
{
string tagName = tag[Tag.TAGNAME];
System.Collections.Hashtable attrs = tag.Attributes;
System.Text.StringBuilder lString = new System.Text.StringBuilder(tagName);
lString.Append(" TAG\n");
lString.Append("--------\n");
foreach (string key in attrs.Keys)
{
string value = (string)attrs[key];
if (!key.ToUpper().Equals(Tag.TAGNAME.ToUpper()) && value.Length > 0)
{
lString.Append(key).Append(" : ").Append(value).Append("\n");
}
}
return(lString.ToString());
}
public virtual int IncrementCounter(int i, NodeReader reader, int state, Tag tag)
{
string nextLine = null;
if ((state == TAG_BEGIN_PARSING_STATE || state == TAG_IGNORE_DATA_STATE ||
state == TAG_IGNORE_BEGIN_TAG_STATE) && i == tag.TagLine.Length - 1)
{
// The while loop below is a bug fix contributed by
// Annette Doyle - see testcase HTMLImageScannerTest.testImageTagOnMultipleLines()
// Further modified by Somik Raha, to remove bug - HTMLTagTest.testBrokenTag
int numLinesAdvanced = 0;
do
{
nextLine = reader.GetNextLine();
numLinesAdvanced++;
} while (nextLine != null && nextLine.Length == 0);
if (nextLine == null)
{
// This means we have a broken tag. Fill in an end tag symbol here.
nextLine = ">";
}
else
{
// This means this is just a new line, hence add the new line character
tag.Append(Parser.LineSeparator);
}
// Ensure blank lines are included in tag's 'tagLines'
while (--numLinesAdvanced > 0)
{
tag.TagLine = "";
}
// We need to continue parsing to the next line
tag.TagLine = nextLine;
i = -1;
}
return(++i);
}
