/// <summary>
/// Initializes a new instance of the <see cref="RegexCapture"/> class.
/// </summary>
/// <param name="buffer">The buffer.</param>
public RegexCapture(RegexBuffer buffer)
{
if (buffer == null)
{
throw new ArgumentNullException(nameof(buffer), "RegexBuffer is null");
}
this.startLocation = buffer.Offset;
buffer.MoveNext();
// we're not in a series of normal characters, so clear
buffer.Clearinseries();
// if the first character of the capture is a '?',
// we need to decode what comes after it.
if (buffer.Current == '?')
{
bool decoded = this.CheckNamed(buffer);
if (!decoded)
{
decoded = this.CheckBalancedGroup(buffer);
}
if (!decoded)
{
decoded = this.CheckNonCapturing(buffer);
}
if (!decoded)
{
decoded = this.CheckOptions(buffer);
}
if (!decoded)
{
decoded = this.CheckLookahead(buffer);
}
if (!decoded)
{
decoded = this.CheckNonBacktracking(buffer);
}
if (!decoded)
{
this.CheckConditional(buffer);
}
}
else
{
// plain old capture...
if (!this.HandlePlainOldCapture(buffer))
{
throw new Exception($"Unrecognized capture: {buffer.String}");
}
}
buffer.AddLookup(this, this.startLocation, buffer.Offset - 1);
}
// eat the whole comment until the end of line...
private static void EatComment(RegexBuffer buffer)
{
while (buffer.Current != '\r')
{
buffer.MoveNext();
}
}
/// <summary>
/// Initializes a new instance of the <see cref="RegexCharClass"/> class.
/// </summary>
/// <param name="buffer">The buffer.</param>
public RegexCharClass(RegexBuffer buffer)
{
if (buffer == null)
{
throw new ArgumentNullException(nameof(buffer), "RegexBuffer is null");
}
int startLoc = buffer.Offset;
buffer.MoveNext();
Regex regex = new Regex(@"(?<Negated>\^?)(?<Class>.+?)\]");
Match match = regex.Match(buffer.String);
if (match.Success)
{
this.description = string.Format(match.Groups["Negated"].ToString() == "^" ? "Any character not in \"{0}\"" : "Any character in \"{0}\"", match.Groups["Class"]);
buffer.Offset += match.Groups[0].Length;
}
else
{
this.description = "missing ']' in character class";
}
buffer.AddLookup(this, startLoc, buffer.Offset - 1);
}
/// <summary>
/// Initializes a new instance of the <see cref="RegexCharClass"/> class.
/// </summary>
/// <param name="buffer">The buffer.</param>
public RegexCharClass(RegexBuffer buffer)
{
if (buffer == null)
{
throw new ArgumentNullException("buffer", "RegexBuffer is null");
}
int startLoc = buffer.Offset;
buffer.MoveNext();
Regex regex = new Regex(@"(?<Negated>\^?)(?<Class>.+?)\]");
Match match = regex.Match(buffer.String);
if (match.Success)
{
this.description = string.Format(match.Groups["Negated"].ToString() == "^" ? "Any character not in \"{0}\"" : "Any character in \"{0}\"", match.Groups["Class"]);
buffer.Offset += match.Groups[0].Length;
}
else
{
this.description = "missing ']' in character class";
}
buffer.AddLookup(this, startLoc, buffer.Offset - 1);
}
/// <summary>
/// Initializes a new instance of the <see cref="RegexCapture"/> class.
/// </summary>
/// <param name="buffer">The buffer.</param>
public RegexCapture(RegexBuffer buffer)
{
if (buffer == null)
{
throw new ArgumentNullException("buffer", "RegexBuffer is null");
}
this.startLocation = buffer.Offset;
buffer.MoveNext();
// we're not in a series of normal characters, so clear
buffer.Clearinseries();
// if the first character of the capture is a '?',
// we need to decode what comes after it.
if (buffer.Current == '?')
{
bool decoded = this.CheckNamed(buffer);
if (!decoded)
{
decoded = this.CheckBalancedGroup(buffer);
}
if (!decoded)
{
decoded = this.CheckNonCapturing(buffer);
}
if (!decoded)
{
decoded = this.CheckOptions(buffer);
}
if (!decoded)
{
decoded = this.CheckLookahead(buffer);
}
if (!decoded)
{
decoded = this.CheckNonBacktracking(buffer);
}
if (!decoded)
{
this.CheckConditional(buffer);
}
}
else
{
// plain old capture...
if (!this.HandlePlainOldCapture(buffer))
{
throw new Exception(string.Format("Unrecognized capture: {0}", buffer.String));
}
}
buffer.AddLookup(this, this.startLocation, buffer.Offset - 1);
}
/// <summary>
/// Initializes a new instance of the <see cref="RegexQuantifier"/> class.
/// </summary>
/// <param name="buffer">The buffer.</param>
public RegexQuantifier(RegexBuffer buffer)
{
if (buffer == null)
{
throw new ArgumentNullException("buffer", "RegexBuffer is null");
}
int startLoc = buffer.Offset;
buffer.MoveNext();
// look for "n}", "n,}", or "n,m}"
Regex regex = new Regex(@"(?<n>\d+)(?<Comma>,?)(?<m>\d*)\}");
Match match = regex.Match(buffer.String);
if (match.Success)
{
if (match.Groups["m"].Length != 0)
{
this.description = string.Format("At least {0}, but not more than {1} times", match.Groups["n"], match.Groups["m"]);
}
else if (match.Groups["Comma"].Length != 0)
{
this.description = string.Format("At least {0} times", match.Groups["n"]);
}
else
{
this.description = string.Format("Exactly {0} times", match.Groups["n"]);
}
buffer.Offset += match.Groups[0].Length;
if (!buffer.AtEnd && buffer.Current == '?')
{
this.description += " (non-greedy)";
buffer.MoveNext();
}
}
else
{
this.description = "missing '}' in quantifier";
}
buffer.AddLookup(this, startLoc, buffer.Offset - 1);
}
/// <summary>
/// Decodes the escape.
/// </summary>
/// <param name="buffer">The buffer.</param>
private void DecodeEscape(RegexBuffer buffer)
{
buffer.MoveNext();
this.character = (string)Escaped[buffer.Current];
if (this.character == null)
{
bool decoded = this.CheckBackReference(buffer);
if (!decoded)
{
// TODO: Handle other items below:
switch (buffer.Current)
{
case 'u':
buffer.MoveNext();
string unicode = buffer.String.Substring(0, 4);
this.character = "Unicode " + unicode;
buffer.Offset += 4;
break;
case ' ':
this.character = "' ' (space)";
this.special = false;
buffer.MoveNext();
break;
case 'c':
buffer.MoveNext();
this.character = "CTRL-" + buffer.Current;
buffer.MoveNext();
break;
case 'x':
buffer.MoveNext();
string number = buffer.String.Substring(0, 2);
this.character = "Hex " + number;
buffer.Offset += 2;
break;
default:
this.character = new string(buffer.Current, 1);
this.special = false;
buffer.MoveNext();
break;
}
}
}
else
{
this.special = true;
buffer.MoveNext();
}
}
/// <summary>
/// Initializes a new instance of the <see cref="RegexQuantifier"/> class.
/// </summary>
/// <param name="buffer">The buffer.</param>
public RegexQuantifier(RegexBuffer buffer)
{
if (buffer == null)
{
throw new ArgumentNullException("buffer", "RegexBuffer is null");
}
int startLoc = buffer.Offset;
buffer.MoveNext();
// look for "n}", "n,}", or "n,m}"
Regex regex = new Regex(@"(?<n>\d+)(?<Comma>,?)(?<m>\d*)\}");
Match match = regex.Match(buffer.String);
if (match.Success)
{
if (match.Groups["m"].Length != 0)
{
this.description = string.Format("At least {0}, but not more than {1} times", match.Groups["n"], match.Groups["m"]);
}
else if (match.Groups["Comma"].Length != 0)
{
this.description = string.Format("At least {0} times", match.Groups["n"]);
}
else
{
this.description = string.Format("Exactly {0} times", match.Groups["n"]);
}
buffer.Offset += match.Groups[0].Length;
if (!buffer.AtEnd && buffer.Current == '?')
{
this.description += " (non-greedy)";
buffer.MoveNext();
}
}
else
{
this.description = "missing '}' in quantifier";
}
buffer.AddLookup(this, startLoc, buffer.Offset - 1);
}
/// <summary>
/// Parses the specified buffer.
/// </summary>
/// <param name="buffer">The buffer.</param>
private void Parse(RegexBuffer buffer)
{
while (!buffer.AtEnd)
{
// if this regex ignores whitespace, we need to ignore these
if (buffer.IgnorePatternWhitespace && ((buffer.Current == ' ') || (buffer.Current == '\r') || (buffer.Current == '\n') || (buffer.Current == '\t')))
{
buffer.MoveNext();
}
else
{
switch (buffer.Current)
{
case '(':
this.items.Add(new RegexCapture(buffer));
break;
case ')':
// end of closure; just return.
return;
case '[':
this.items.Add(new RegexCharClass(buffer));
break;
case '{':
this.items.Add(new RegexQuantifier(buffer));
break;
case '|':
this.items.Add(new RegexAlternate(buffer));
break;
case '\\':
this.items.Add(new RegexCharacter(buffer));
break;
case '#':
if (buffer.IgnorePatternWhitespace)
{
EatComment(buffer);
}
else
{
this.items.Add(new RegexCharacter(buffer));
}
break;
default:
this.items.Add(new RegexCharacter(buffer));
break;
}
}
}
}
/// <summary>
/// Initializes a new instance of the <see cref="RegexCharacter"/> class.
/// </summary>
/// <param name="buffer">The buffer.</param>
public RegexCharacter(RegexBuffer buffer)
{
if (buffer == null)
{
throw new ArgumentNullException("buffer", "RegexBuffer is null");
}
int startLoc = buffer.Offset;
bool quantifier = false;
switch (buffer.Current)
{
case '.':
this.character = ". (any character)";
buffer.MoveNext();
this.special = true;
break;
case '+':
this.character = "+ (one or more times)";
buffer.MoveNext();
this.special = true;
quantifier = true;
break;
case '*':
this.character = "* (zero or more times)";
buffer.MoveNext();
this.special = true;
quantifier = true;
break;
case '?':
this.character = "? (zero or one time)";
buffer.MoveNext();
this.special = true;
quantifier = true;
break;
case '^':
this.character = "^ (anchor to start of string)";
buffer.MoveNext();
break;
case '$':
this.character = "$ (anchor to end of string)";
buffer.MoveNext();
break;
case ' ':
this.character = "' ' (space)";
buffer.MoveNext();
break;
case '\\':
this.DecodeEscape(buffer);
break;
default:
this.character = buffer.Current.ToString(CultureInfo.CurrentCulture);
buffer.MoveNext();
this.special = false;
break;
}
if (quantifier)
{
if (!buffer.AtEnd && buffer.Current == '?')
{
this.character += " (non-greedy)";
buffer.MoveNext();
}
}
buffer.AddLookup(this, startLoc, buffer.Offset - 1, this.character.Length == 1);
}
/// <summary>
/// Initializes a new instance of the <see cref="RegexCharacter"/> class.
/// </summary>
/// <param name="buffer">The buffer.</param>
public RegexCharacter(RegexBuffer buffer)
{
if (buffer == null)
{
throw new ArgumentNullException("buffer", "RegexBuffer is null");
}
int startLoc = buffer.Offset;
bool quantifier = false;
switch (buffer.Current)
{
case '.':
this.character = ". (any character)";
buffer.MoveNext();
this.special = true;
break;
case '+':
this.character = "+ (one or more times)";
buffer.MoveNext();
this.special = true;
quantifier = true;
break;
case '*':
this.character = "* (zero or more times)";
buffer.MoveNext();
this.special = true;
quantifier = true;
break;
case '?':
this.character = "? (zero or one time)";
buffer.MoveNext();
this.special = true;
quantifier = true;
break;
case '^':
this.character = "^ (anchor to start of string)";
buffer.MoveNext();
break;
case '$':
this.character = "$ (anchor to end of string)";
buffer.MoveNext();
break;
case ' ':
this.character = "' ' (space)";
buffer.MoveNext();
break;
case '\\':
this.DecodeEscape(buffer);
break;
default:
this.character = buffer.Current.ToString(CultureInfo.CurrentCulture);
buffer.MoveNext();
this.special = false;
break;
}
if (quantifier)
{
if (!buffer.AtEnd && buffer.Current == '?')
{
this.character += " (non-greedy)";
buffer.MoveNext();
}
}
buffer.AddLookup(this, startLoc, buffer.Offset - 1, this.character.Length == 1);
}
/// <summary>
/// Decodes the escape.
/// </summary>
/// <param name="buffer">The buffer.</param>
private void DecodeEscape(RegexBuffer buffer)
{
buffer.MoveNext();
this.character = (string)Escaped[buffer.Current];
if (this.character == null)
{
bool decoded = this.CheckBackReference(buffer);
if (!decoded)
{
// TODO: Handle other items below:
switch (buffer.Current)
{
case 'u':
buffer.MoveNext();
string unicode = buffer.String.Substring(0, 4);
this.character = "Unicode " + unicode;
buffer.Offset += 4;
break;
case ' ':
this.character = "' ' (space)";
this.special = false;
buffer.MoveNext();
break;
case 'c':
buffer.MoveNext();
this.character = "CTRL-" + buffer.Current;
buffer.MoveNext();
break;
case 'x':
buffer.MoveNext();
string number = buffer.String.Substring(0, 2);
this.character = "Hex " + number;
buffer.Offset += 2;
break;
default:
this.character = new string(buffer.Current, 1);
this.special = false;
buffer.MoveNext();
break;
}
}
}
else
{
this.special = true;
buffer.MoveNext();
}
}
