public void Tests()
{
var p = new StringScanner();
p.Reset("This is a string with something [bracketed]");
Assert.IsTrue(p.Bof);
Assert.IsFalse(p.Eof);
Assert.IsTrue(p.SkipString("This"));
Assert.IsFalse(p.Bof);
Assert.IsFalse(p.Eof);
Assert.IsFalse(p.SkipString("huh?"));
Assert.IsTrue(p.SkipLinespace());
Assert.IsTrue(p.SkipChar('i'));
Assert.IsTrue(p.SkipChar('s'));
Assert.IsTrue(p.SkipWhitespace());
Assert.IsTrue(p.DoesMatchAny(new char[] { 'r', 'a', 't'} ));
Assert.IsFalse(p.Find("Not here"));
Assert.IsFalse(p.Find("WITH"));
Assert.IsFalse(p.FindI("Not here"));
Assert.IsTrue(p.FindI("WITH"));
Assert.IsTrue(p.Find('['));
p.SkipForward(1);
p.Mark();
Assert.IsTrue(p.Find(']'));
Assert.AreEqual("bracketed", p.Extract());
Assert.IsTrue(p.SkipChar(']'));
Assert.IsTrue(p.Eof);
}
/// <summary>
/// Strips the special attributes specified at the location of start. Special attributes are a php markdown extension
/// and specified between {}. Attributes are separated with spaces. It recognizes the following attributes:
/// id, which start with a '#'. Only the first one is used
/// css classes, which start with a'.'. All specified are used
/// name=value pairs, which will end up as attributes on the element.
/// </summary>
/// <param name="str">The string we're scanning.</param>
/// <param name="start">The start (current) position.</param>
/// <param name="end">The end position. Is only valid if returned list contains at least one value.</param>
/// <returns>list of special attributes found, or null if none found or error in string format.</returns>
public static List <string> StripSpecialAttributes(string str, int start, out int end)
{
end = start;
var scanner = new StringScanner(str, start);
if (!scanner.DoesMatch('{'))
{
// not a start of a special attribute block
return(null);
}
// first find the end of the block,
scanner.SkipForward(1);
var startOfAttributes = scanner.Position;
// first find the next EOL, as the closing } has to be on this line.
scanner.SkipToEol();
var nextEolPos = scanner.Position;
scanner.Position = startOfAttributes;
scanner.Mark();
if (!scanner.Find('}') || scanner.Position >= nextEolPos)
{
// not enclosed properly.
return(null);
}
var attributesString = scanner.Extract();
if (string.IsNullOrWhiteSpace(attributesString))
{
return(null);
}
// Position is on enclosing '}' (due to the Find('}'), so we have to skip 1 character
end = scanner.Position + 1;
return(attributesString.Split(' ').Where(s => !string.IsNullOrWhiteSpace(s)).ToList());
}
internal static LinkDefinition ParseLinkDefinitionInternal(StringScanner p, bool extraMode)
{
// Skip leading white space
p.SkipWhitespace();
// Must start with an opening square bracket
if (!p.SkipChar('['))
return null;
// Extract the id
p.Mark();
if (!p.Find(']'))
return null;
var id = p.Extract();
if (id.Length == 0)
return null;
if (!p.SkipString("]:"))
return null;
// Parse the url and title
var link = ParseLinkTarget(p, id, extraMode);
// and trailing whitespace
p.SkipLinespace();
// Trailing crap, not a valid link reference...
if (!p.EndOfLine)
return null;
return link;
}
internal static LinkDefinition ParseLinkDefinitionInternal(StringScanner p, bool ExtraMode)
{
// Skip leading white space
p.SkipWhitespace();
// Must start with an opening square bracket
if (!p.SkipChar('['))
{
return(null);
}
// Extract the id
p.Mark();
if (!p.Find(']'))
{
return(null);
}
string id = p.Extract();
if (id.Length == 0)
{
return(null);
}
if (!p.SkipString("]:"))
{
return(null);
}
// Parse the url and title
var link = ParseLinkTarget(p, id, ExtraMode);
// and trailing whitespace
p.SkipLinespace();
// Trailing crap, not a valid link reference...
if (!p.eol)
{
return(null);
}
return(link);
}
private static HtmlTag ParseHelper(StringScanner p)
{
// Does it look like a tag?
if (p.Current != '<')
return null;
// Skip '<'
p.SkipForward(1);
// Is it a comment?
if (p.SkipString("!--"))
{
p.Mark();
if (p.Find("-->"))
{
var t = new HtmlTag("!");
t.m_attributes.Add("content", p.Extract());
t.m_closed = true;
p.SkipForward(3);
return t;
}
}
// Is it a closing tag eg: </div>
bool bClosing = p.SkipChar('/');
// Get the tag name
string tagName=null;
if (!p.SkipIdentifier(ref tagName))
return null;
// Probably a tag, create the HtmlTag object now
HtmlTag tag = new HtmlTag(tagName);
tag.m_closing = bClosing;
// If it's a closing tag, no attributes
if (bClosing)
{
if (p.Current != '>')
return null;
p.SkipForward(1);
return tag;
}
while (!p.Eof)
{
// Skip whitespace
p.SkipWhitespace();
// Check for closed tag eg: <hr />
if (p.SkipString("/>"))
{
tag.m_closed=true;
return tag;
}
// End of tag?
if (p.SkipChar('>'))
{
return tag;
}
// attribute name
string attributeName = null;
if (!p.SkipIdentifier(ref attributeName))
return null;
// Skip whitespace
p.SkipWhitespace();
// Skip equal sign
if (p.SkipChar('='))
{
// Skip whitespace
p.SkipWhitespace();
// Optional quotes
if (p.SkipChar('\"'))
{
// Scan the value
p.Mark();
if (!p.Find('\"'))
return null;
// Store the value
tag.m_attributes.Add(attributeName, p.Extract());
// Skip closing quote
p.SkipForward(1);
}
else
{
// Scan the value
p.Mark();
while (!p.Eof && !char.IsWhiteSpace(p.Current) && p.Current != '>' && p.Current != '/')
p.SkipForward(1);
if (!p.Eof)
{
// Store the value
tag.m_attributes.Add(attributeName, p.Extract());
}
}
}
else
{
tag.m_attributes.Add(attributeName, "");
}
}
return null;
}
private static HtmlTag ParseHelper(StringScanner p)
{
// Does it look like a tag?
if (p.current != '<')
{
return(null);
}
// Skip '<'
p.SkipForward(1);
// Is it a comment?
if (p.SkipString("!--"))
{
p.Mark();
if (p.Find("-->"))
{
var t = new HtmlTag("!");
t.m_attributes.Add("content", p.Extract());
t.m_closed = true;
p.SkipForward(3);
return(t);
}
}
// Is it a closing tag eg: </div>
bool bClosing = p.SkipChar('/');
// Get the tag name
string tagName = null;
if (!p.SkipIdentifier(ref tagName))
{
return(null);
}
// Probably a tag, create the HtmlTag object now
HtmlTag tag = new HtmlTag(tagName);
tag.m_closing = bClosing;
// If it's a closing tag, no attributes
if (bClosing)
{
if (p.current != '>')
{
return(null);
}
p.SkipForward(1);
return(tag);
}
while (!p.eof)
{
// Skip whitespace
p.SkipWhitespace();
// Check for closed tag eg: <hr />
if (p.SkipString("/>"))
{
tag.m_closed = true;
return(tag);
}
// End of tag?
if (p.SkipChar('>'))
{
return(tag);
}
// attribute name
string attributeName = null;
if (!p.SkipIdentifier(ref attributeName))
{
return(null);
}
// Skip whitespace
p.SkipWhitespace();
// Skip equal sign
if (p.SkipChar('='))
{
// Skip whitespace
p.SkipWhitespace();
// Optional quotes
if (p.SkipChar('\"'))
{
// Scan the value
p.Mark();
if (!p.Find('\"'))
{
return(null);
}
// Store the value
tag.m_attributes.Add(attributeName, p.Extract());
// Skip closing quote
p.SkipForward(1);
}
else
{
// Scan the value
p.Mark();
while (!p.eof && !char.IsWhiteSpace(p.current) && p.current != '>' && p.current != '/')
{
p.SkipForward(1);
}
if (!p.eof)
{
// Store the value
tag.m_attributes.Add(attributeName, p.Extract());
}
}
}
else
{
tag.m_attributes.Add(attributeName, "");
}
}
return(null);
}