private IToken TryGetVariable()
{
// Consume first Character which must be a ?
ConsumeCharacter();
// Consume other valid Characters
char next = Peek();
while (Char.IsLetterOrDigit(next) || UnicodeSpecsHelper.IsLetterOrDigit(next) || next == '-' || next == '_')
{
ConsumeCharacter();
next = Peek();
}
// Validate
String value = Value;
if (IsValidVarName(value))
{
return(new VariableToken(value, CurrentLine, StartPosition, EndPosition));
}
else
{
throw Error("The value '" + value + "' is not valid a Variable Name");
}
}
private IToken TryGetDataType()
{
// Expect to either see a start of a Uri or QName
char next = Peek();
if (next == '<')
{
// Uri specified DataType
IToken temp = TryGetUri();
LastTokenType = Token.DATATYPE;
return(new DataTypeToken("<" + temp.Value + ">", temp.StartLine, temp.StartPosition, temp.EndPosition));
}
else if (Char.IsLetter(next) || UnicodeSpecsHelper.IsLetter(next) || next == '_')
{
// QName specified Data Type
IToken temp = TryGetQName();
if (temp.TokenType == Token.QNAME)
{
LastTokenType = Token.DATATYPE;
return(new DataTypeToken(temp.Value, temp.StartLine, temp.StartPosition, temp.EndPosition));
}
else
{
throw Error("Unexpected Token '" + temp.GetType().ToString() + "' was produced when a QName for a Data Type was expected!");
}
}
else
{
// Invalid Start Character
throw Error("Unexpected Character (Code " + (int)next + " " + next + "\nExpected a < to start a URI or a valid start character for a QName to specify Data Type");
}
}
public static char HexUnescape(String value, ref int index)
{
if (index + 2 >= value.Length)
{
throw new RdfParseException("Malformed Percent Encoded Escape");
}
if (value[index] != '%')
{
throw new RdfParseException("Malformed Percent Encoded Escape");
}
index = index + 3;
return(UnicodeSpecsHelper.ConvertToChar(value.Substring(index + 1, 2)));
}
private IToken TryGetVariable()
{
// Consume first Character which must be a ?/$
ConsumeCharacter();
// Consume other valid Characters
char next = Peek();
while (Char.IsLetterOrDigit(next) || UnicodeSpecsHelper.IsLetterOrDigit(next) || next == '-' || next == '_' || next == '\\')
{
if (next == '\\')
{
// Check its a valid Escape
HandleEscapes(TokeniserEscapeMode.QName);
}
else
{
ConsumeCharacter();
}
next = Peek();
}
// Validate
String value = Value;
if (value.EndsWith("."))
{
Backtrack();
value = value.Substring(0, value.Length - 1);
}
if (SparqlSpecsHelper.IsValidVarName(value))
{
return(new VariableToken(value, CurrentLine, StartPosition, EndPosition));
}
else
{
throw Error("The value '" + value + "' is not valid a Variable Name");
}
}
private bool IsValidVarName(String value)
{
if (_isValidVarName.IsMatch(value))
{
return(true);
}
else
{
// Have to validate Character by Character
char[] cs = value.ToCharArray();
char first = cs[0];
// First character must be an underscore or letter
if (first == '_' || Char.IsLetter(first) || UnicodeSpecsHelper.IsLetter(first))
{
// Remaining Characters must be underscores, letters, numbers or hyphens
for (int i = 1; i < cs.Length; i++)
{
char c = cs[i];
if (c == '_' || c == '-' || Char.IsLetterOrDigit(c) || UnicodeSpecsHelper.IsLetterOrDigit(c))
{
// OK
}
else
{
// Invalid Character
return(false);
}
}
// If we get here it's all fine
return(true);
}
else
{
// Invalid Start Character
return(false);
}
}
}
/// <summary>
/// Handles the special SPARQL escapes that can occur in a local name
/// </summary>
/// <remarks>
/// Unlike <see cref="BaseTokeniser.HandleEscapes()">HandleEscapes()</see> this only unescapes unicode escapes, other escapes are simply validated and passed through for later unescaping
/// </remarks>
protected void HandleSparqlLocalNameEscapes()
{
//Grab the first character which must be a \ or %
char next = this.SkipCharacter();
//Stuff for Unicode/Hex escapes
StringBuilder localOutput;
if (next == '\\')
{
//Backslash based escape
next = this.Peek();
switch (next)
{
case '_':
case '-':
case '.':
case '|':
case '$':
case '&':
case '\'':
case '(':
case ')':
case '*':
case '+':
case ',':
case ';':
case '=':
case ':':
case '/':
case '?':
case '#':
case '@':
case '%':
//Escapable Characters
this._output.Append('\\');
this.ConsumeCharacter();
return;
case 'u':
//Need to consume the u first
localOutput = new StringBuilder();
this.SkipCharacter();
next = this.Peek();
//Try to get Four Hex Digits
while (localOutput.Length < 4 && this.IsHexDigit(next))
{
localOutput.Append(next);
this.SkipCharacter();
next = this.Peek();
}
//Did we get four Hex Digits
if (localOutput.Length != 4)
{
throw Error("Unexpected Character (Code " + (int)next + "): " + next + " encountered while trying to parse Unicode Escape from Content:\n" + this._output.ToString() + "\nThe \\u Escape must be followed by four Hex Digits");
}
else if (localOutput.ToString().Equals("0000"))
{
//Ignore the null escape
}
else
{
this._output.Append(UnicodeSpecsHelper.ConvertToChar(localOutput.ToString()));
}
return;
case 'U':
//Need to consume the U first
localOutput = new StringBuilder();
this.SkipCharacter();
next = this.Peek();
//Try to get Eight Hex Digits
while (localOutput.Length < 8 && this.IsHexDigit(next))
{
localOutput.Append(next);
this.SkipCharacter();
next = this.Peek();
}
//Did we get eight Hex Digits
if (localOutput.Length != 8)
{
throw Error("Unexpected Character (Code " + (int)next + "): " + next + " encountered while trying to parse Unicode Escape from Content:\n" + this._output.ToString() + "\nThe \\U Escape must be followed by eight Hex Digits");
}
else if (localOutput.ToString().Equals("00000000"))
{
//Ignore the null escape
}
else
{
this._output.Append(UnicodeSpecsHelper.ConvertToChar(localOutput.ToString()));
}
return;
default:
throw Error("Unexpected Backslash Character encountered in a Local Name, the Backslash Character can only be used for Unicode escapes (\\u and \\U) and a limited set of special characters (_-.|&'()*+,;=/?#@%) in Local Names");
}
}
else if (next == '%')
{
localOutput = new StringBuilder();
localOutput.Append(next);
next = this.Peek();
while (localOutput.Length < 3 && this.IsHexDigit(next))
{
localOutput.Append(next);
this.SkipCharacter();
next = this.Peek();
}
//Did we get % followed by two hex digits
if (localOutput.Length != 3)
{
throw Error("Encountered a % character in a Local Name but the required two hex digits were not present after it, please use \\% if you wish to represent the percent character");
}
#if !SILVERLIGHT
else if (!Uri.IsHexEncoding(localOutput.ToString(), 0))
#else
else if (SilverlightExtensions.IsHexEncoding(localOutput.ToString(), 0))
#endif
{
throw Error("Invalid % encoded character encountered");
}
else
{
this._output.Append(localOutput.ToString());
}
}
else
{
throw Error("HandleSparqlLocalNameEscapes() was called but the next character is not a % or \\ as expected");
}
}
/// <summary>
/// Handles the standard escapes supported in all the UTF-8 based RDF serializations
/// </summary>
protected void HandleEscapes(TokeniserEscapeMode mode)
{
//Grab the first character which must be a \
char next = this.SkipCharacter();
if (next != '\\')
{
throw Error("HandleEscapes() was called but the first character was not a \\ as expected");
}
//Stuff for Unicode escapes
StringBuilder localOutput;
next = this.Peek();
switch (next)
{
case '\\':
//Backslash escape
if (mode != TokeniserEscapeMode.QName)
{
//Consume this one Backslash
this.ConsumeCharacter();
//If this was a backslash escape i.e. \\
//Then need to check whether the subsequent character could be confused with a valid escape
//in the tokenised output and if so insert another backslash into the output
next = this.Peek();
switch (next)
{
case 't':
case 'n':
case 'r':
case 'u':
case 'U':
this._output.Append("\\");
break;
}
return;
}
else
{
goto default;
}
case '"':
//Quote escape (only valid in Quoted Literals)
if (mode == TokeniserEscapeMode.QuotedLiterals || mode == TokeniserEscapeMode.QuotedLiteralsAlternate)
{
//Consume and return
this.ConsumeCharacter();
return;
}
else
{
goto default;
}
case '\'':
//Single Quote Escape (only valid in Alternate Quoted Literals)
if (mode == TokeniserEscapeMode.QuotedLiteralsAlternate)
{
//Consume and return
this.ConsumeCharacter();
return;
}
else
{
goto default;
}
case '>':
//End Uri Escape (only valid in URIs)
if (mode == TokeniserEscapeMode.Uri)
{
//Consume and return
this.ConsumeCharacter();
return;
}
else
{
goto default;
}
#region White Space Escapes
case 'n':
//New Line Escape
if (mode != TokeniserEscapeMode.QName)
{
//Discard and append a real New Line to the output
this.SkipCharacter();
this._output.Append("\n");
return;
}
else
{
goto default;
}
case 'r':
//New Line Escape
if (mode != TokeniserEscapeMode.QName)
{
//Discard and append a real New Line to the output
this.SkipCharacter();
this._output.Append("\r");
return;
}
else
{
goto default;
}
case 't':
//Tab Escape
if (mode != TokeniserEscapeMode.QName)
{
//Discard and append a real Tab to the output
this.SkipCharacter();
this._output.Append("\t");
return;
}
else
{
goto default;
}
#endregion
#region Unicode Escapes
case 'u':
//Need to consume the u first
localOutput = new StringBuilder();
this.SkipCharacter();
next = this.Peek();
//Try to get Four Hex Digits
while (localOutput.Length < 4 && this.IsHexDigit(next))
{
localOutput.Append(next);
this.SkipCharacter();
next = this.Peek();
}
//Did we get four Hex Digits
if (localOutput.Length != 4)
{
throw Error("Unexpected Character (Code " + (int)next + "): " + next + " encountered while trying to parse Unicode Escape from Content:\n" + this._output.ToString() + "\nThe \\u Escape must be followed by four Hex Digits");
}
else if (localOutput.ToString().Equals("0000"))
{
//Ignore the null escape
}
else
{
this._output.Append(UnicodeSpecsHelper.ConvertToChar(localOutput.ToString()));
}
return;
case 'U':
//Need to consume the U first
localOutput = new StringBuilder();
this.SkipCharacter();
next = this.Peek();
//Try to get Eight Hex Digits
while (localOutput.Length < 8 && this.IsHexDigit(next))
{
localOutput.Append(next);
this.SkipCharacter();
next = this.Peek();
}
//Did we get eight Hex Digits
if (localOutput.Length != 8)
{
throw Error("Unexpected Character (Code " + (int)next + "): " + next + " encountered while trying to parse Unicode Escape from Content:\n" + this._output.ToString() + "\nThe \\U Escape must be followed by eight Hex Digits");
}
else if (localOutput.ToString().Equals("00000000"))
{
//Ignore the null escape
}
else
{
this._output.Append(UnicodeSpecsHelper.ConvertToChar(localOutput.ToString()));
}
return;
#endregion
default:
//Not an escape character
if (mode != TokeniserEscapeMode.QName)
{
//Append the \ and then return
//Processing continues normally in the caller function
this._output.Append("\\");
return;
}
else
{
throw Error("Unexpected Backslash Character encountered in a QName, the Backslash Character can only be used for Unicode escapes (\\u and \\U) in QNames");
}
}
}
/// <summary>
/// Handles the standard escapes supported in all the UTF-8 based RDF serializations
/// </summary>
protected void HandleEscapes(TokeniserEscapeMode mode)
{
// Grab the first character which must be a \
char next = SkipCharacter();
if (next != '\\')
{
throw Error("HandleEscapes() was called but the first character was not a \\ as expected");
}
// Stuff for Unicode escapes
StringBuilder localOutput;
bool isLiteral = (mode == TokeniserEscapeMode.QuotedLiterals || mode == TokeniserEscapeMode.QuotedLiteralsAlternate || mode == TokeniserEscapeMode.QuotedLiteralsBoth);
next = Peek();
switch (next)
{
case '\\':
// Backslash escape
if (isLiteral || mode == TokeniserEscapeMode.PermissiveUri)
{
// Consume this one Backslash
ConsumeCharacter();
return;
}
goto default;
case '"':
// Quote escape (only valid in Quoted Literals)
if (mode == TokeniserEscapeMode.QuotedLiterals || mode == TokeniserEscapeMode.QuotedLiteralsBoth)
{
// Consume and return
ConsumeCharacter();
return;
}
goto default;
case '\'':
// Single Quote Escape (only valid in Alternate Quoted Literals)
if (mode == TokeniserEscapeMode.QuotedLiteralsAlternate || mode == TokeniserEscapeMode.QuotedLiteralsBoth)
{
// Consume and return
ConsumeCharacter();
return;
}
goto default;
case '>':
// End Uri Escape (only valid in URIs)
if (mode == TokeniserEscapeMode.Uri)
{
// Consume and return
ConsumeCharacter();
return;
}
goto default;
case 'n':
// New Line Escape
if (isLiteral || mode == TokeniserEscapeMode.PermissiveUri)
{
// Discard and append a real New Line to the output
SkipCharacter();
_output.Append('\n');
return;
}
goto default;
case 'r':
// New Line Escape
if (isLiteral || mode == TokeniserEscapeMode.PermissiveUri)
{
// Discard and append a real New Line to the output
SkipCharacter();
_output.Append('\r');
return;
}
goto default;
case 't':
// Tab Escape
if (isLiteral || mode == TokeniserEscapeMode.PermissiveUri)
{
// Discard and append a real Tab to the output
SkipCharacter();
_output.Append('\t');
return;
}
goto default;
case 'b':
// Backspace Escape
if (isLiteral)
{
// Discard and append a real backspace to the output
SkipCharacter();
_output.Append('\b');
return;
}
goto default;
case 'f':
// Form Feed Escape
if (isLiteral)
{
// Discard and append a real form feed to the output
SkipCharacter();
_output.Append('\f');
return;
}
goto default;
case 'u':
// Need to consume the u first
localOutput = new StringBuilder();
SkipCharacter();
next = Peek();
// Try to get Four Hex Digits
while (localOutput.Length < 4 && IsHexDigit(next))
{
localOutput.Append(next);
SkipCharacter();
next = Peek();
}
// Did we get four Hex Digits
if (localOutput.Length != 4)
{
throw Error("Unexpected Character (Code " + (int)next + "): " + next + " encountered while trying to parse Unicode Escape from Content:\n" + _output.ToString() + "\nThe \\u Escape must be followed by four Hex Digits");
}
_output.Append(UnicodeSpecsHelper.ConvertToChar(localOutput.ToString()));
return;
case 'U':
// Need to consume the U first
localOutput = new StringBuilder();
SkipCharacter();
next = Peek();
// Try to get Eight Hex Digits
while (localOutput.Length < 8 && IsHexDigit(next))
{
localOutput.Append(next);
SkipCharacter();
next = Peek();
}
// Did we get eight Hex Digits
if (localOutput.Length != 8)
{
throw Error("Unexpected Character (Code " + (int)next + "): " + next + " encountered while trying to parse Unicode Escape from Content:\n" + _output.ToString() + "\nThe \\U Escape must be followed by eight Hex Digits");
}
_output.Append(UnicodeSpecsHelper.ConvertToChars(localOutput.ToString()));
return;
default:
// Not an escape character
throw Error("Invalid escape sequence encountered, \\" + next + " is not a valid escape sequence in the current token");
}
}
