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);
}
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);
}
// Parse just the link target
// For reference link definition, this is the bit after "[id]: thisbit"
// For inline link, this is the bit in the parens: [link text](thisbit)
internal static LinkDefinition ParseLinkTarget(StringScanner p, string id, bool ExtraMode)
{
// Skip whitespace
p.SkipWhitespace();
// End of string?
if (p.eol)
return null;
// Create the link definition
var r = new LinkDefinition(id);
// Is the url enclosed in angle brackets
if (p.SkipChar('<'))
{
// Extract the url
p.Mark();
// Find end of the url
while (p.current != '>')
{
if (p.eof)
return null;
p.SkipEscapableChar(ExtraMode);
}
string url = p.Extract();
if (!p.SkipChar('>'))
return null;
// Unescape it
r.url = Utils.UnescapeString(url.Trim(), ExtraMode);
// Skip whitespace
p.SkipWhitespace();
}
else
{
// Find end of the url
p.Mark();
int paren_depth = 1;
while (!p.eol)
{
char ch=p.current;
if (char.IsWhiteSpace(ch))
break;
if (id == null)
{
if (ch == '(')
paren_depth++;
else if (ch == ')')
{
paren_depth--;
if (paren_depth==0)
break;
}
}
p.SkipEscapableChar(ExtraMode);
}
r.url = Utils.UnescapeString(p.Extract().Trim(), ExtraMode);
}
p.SkipLinespace();
// End of inline target
if (p.DoesMatch(')'))
return r;
bool bOnNewLine = p.eol;
int posLineEnd = p.position;
if (p.eol)
{
p.SkipEol();
p.SkipLinespace();
}
// Work out what the title is delimited with
char delim;
switch (p.current)
{
case '\'':
case '\"':
delim = p.current;
break;
case '(':
delim = ')';
break;
default:
if (bOnNewLine)
{
p.position = posLineEnd;
return r;
}
else
return null;
}
// Skip the opening title delimiter
p.SkipForward(1);
// Find the end of the title
p.Mark();
while (true)
{
if (p.eol)
return null;
if (p.current == delim)
{
if (delim != ')')
{
int savepos = p.position;
// Check for embedded quotes in title
// Skip the quote and any trailing whitespace
p.SkipForward(1);
p.SkipLinespace();
// Next we expect either the end of the line for a link definition
// or the close bracket for an inline link
if ((id == null && p.current != ')') ||
(id != null && !p.eol))
{
continue;
}
p.position = savepos;
}
// End of title
break;
}
p.SkipEscapableChar(ExtraMode);
}
// Store the title
r.title = Utils.UnescapeString(p.Extract(), ExtraMode);
// Skip closing quote
p.SkipForward(1);
// Done!
return r;
}
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;
}
// Parse just the link target
// For reference link definition, this is the bit after "[id]: thisbit"
// For inline link, this is the bit in the parens: [link text](thisbit)
internal static LinkDefinition ParseLinkTarget(StringScanner p, string id, bool ExtraMode)
{
// Skip whitespace
p.SkipWhitespace();
// End of string?
if (p.eol)
{
return(null);
}
// Create the link definition
var r = new LinkDefinition(id);
// Is the url enclosed in angle brackets
if (p.SkipChar('<'))
{
// Extract the url
p.Mark();
// Find end of the url
while (p.current != '>')
{
if (p.eof)
{
return(null);
}
p.SkipEscapableChar(ExtraMode);
}
string url = p.Extract();
if (!p.SkipChar('>'))
{
return(null);
}
// Unescape it
r.url = Utils.UnescapeString(url.Trim(), ExtraMode);
// Skip whitespace
p.SkipWhitespace();
}
else
{
// Find end of the url
p.Mark();
int paren_depth = 1;
while (!p.eol)
{
char ch = p.current;
if (char.IsWhiteSpace(ch))
{
break;
}
if (id == null)
{
if (ch == '(')
{
paren_depth++;
}
else if (ch == ')')
{
paren_depth--;
if (paren_depth == 0)
{
break;
}
}
}
p.SkipEscapableChar(ExtraMode);
}
r.url = Utils.UnescapeString(p.Extract().Trim(), ExtraMode);
}
p.SkipLinespace();
// End of inline target
if (p.DoesMatch(')'))
{
return(r);
}
bool bOnNewLine = p.eol;
int posLineEnd = p.position;
if (p.eol)
{
p.SkipEol();
p.SkipLinespace();
}
// Work out what the title is delimited with
char delim;
switch (p.current)
{
case '\'':
case '\"':
delim = p.current;
break;
case '(':
delim = ')';
break;
default:
if (bOnNewLine)
{
p.position = posLineEnd;
return(r);
}
else
{
return(null);
}
}
// Skip the opening title delimiter
p.SkipForward(1);
// Find the end of the title
p.Mark();
while (true)
{
if (p.eol)
{
return(null);
}
if (p.current == delim)
{
if (delim != ')')
{
int savepos = p.position;
// Check for embedded quotes in title
// Skip the quote and any trailing whitespace
p.SkipForward(1);
p.SkipLinespace();
// Next we expect either the end of the line for a link definition
// or the close bracket for an inline link
if ((id == null && p.current != ')') ||
(id != null && !p.eol))
{
continue;
}
p.position = savepos;
}
// End of title
break;
}
p.SkipEscapableChar(ExtraMode);
}
// Store the title
r.title = Utils.UnescapeString(p.Extract(), ExtraMode);
// Skip closing quote
p.SkipForward(1);
// Done!
return(r);
}
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);
}