public IStructureMatchResultSet Match(TextSnapshotOffset snapshotOffset, IStructureMatchOptions options)
{
var snapshot = snapshotOffset.GetDocumentView();
if (snapshot == null)
{
return(null);
}
SyntaxTree syntaxTree;
if (!snapshot.Document.TryGetSyntaxTree(out syntaxTree))
{
return(null);
}
var position = snapshot.Position;
var result = syntaxTree.MatchBraces(position, Matchers);
if (!result.IsValid)
{
return(null);
}
var leftRange = snapshot.ToSnapshotRange(result.Left);
var rightRange = snapshot.ToSnapshotRange(result.Right);
var results = new StructureMatchResultCollection
{
new StructureMatchResult(leftRange),
new StructureMatchResult(rightRange)
};
return(new StructureMatchResultSet(results));
}
/////////////////////////////////////////////////////////////////////////////////////////////////////
// PUBLIC PROCEDURES
/////////////////////////////////////////////////////////////////////////////////////////////////////
/// <summary>
/// Attempts to locate structural text, such as bracket pairs, at the specified offset.
/// </summary>
/// <param name="snapshotOffset">The <see cref="TextSnapshotOffset"/> to examine.</param>
/// <param name="options">The <see cref="IStructureMatchOptions"/> to use.</param>
/// <returns>An <see cref="IStructureMatchResultSet"/> that contains the results that were found, if any.</returns>
/// <remarks>
/// When the result set is empty, no structural delimiter was next to the specified offset.
/// When the result set contains a single result, a structural delimiter such as a bracket was found next to the specified offset,
/// but no matching structural delimiter was found.
/// The result set may contain multiple entries in cases such as when a language wishes to flag <c>#if...#else...#endif</c> blocks as a structure.
/// </remarks>
public override IStructureMatchResultSet Match(TextSnapshotOffset snapshotOffset, IStructureMatchOptions options)
{
if (snapshotOffset.IsDeleted)
{
throw new ArgumentNullException("snapshotOffset");
}
// Get a snapshot reader and configure it for quick initial lookup
ITextSnapshotReader reader = snapshotOffset.Snapshot.GetReader(snapshotOffset.Offset);
reader.Options.DefaultTokenLoadBufferLength = 250;
reader.Options.InitialTokenLoadBufferLength = 4;
IToken token = reader.Token;
if (token != null)
{
// If the token is not a multi-line comment but is at the start of a token, check the previous token
if ((token.Id != SimpleTokenId.MultiLineCommentText) && (reader.IsAtTokenStart))
{
token = reader.ReadTokenReverse();
}
// If the token is a multi-line comment...
if (token.Id == SimpleTokenId.MultiLineCommentText)
{
// The Simple language programmatic lexer variant only has a single token for the entire comment so
// ensure the target offset is at a delimiter (and not within the body of the comment)...
// For most other languages, you'd want to scan tokens to find a matching delimiter token instead
bool isAtStart = (snapshotOffset.Offset <= token.StartOffset + 2);
bool isAtEnd = (snapshotOffset.Offset >= token.EndOffset - 2);
if (isAtStart || isAtEnd)
{
// Get the token's text and ensure it ends with a proper delimiter
string tokenText = reader.TokenText;
if ((token.Length >= 4) && (tokenText.EndsWith("*/", StringComparison.Ordinal)))
{
// Found a valid match
StructureMatchResultCollection results = new StructureMatchResultCollection();
results.Add(new StructureMatchResult(new TextSnapshotRange(reader.Snapshot, token.StartOffset, token.StartOffset + 2))
{
IsSource = isAtStart,
NavigationSnapshotOffset = new TextSnapshotOffset(reader.Snapshot, token.StartOffset)
});
results.Add(new StructureMatchResult(new TextSnapshotRange(reader.Snapshot, token.EndOffset - 2, token.EndOffset))
{
IsSource = !isAtStart,
NavigationSnapshotOffset = new TextSnapshotOffset(reader.Snapshot, token.EndOffset)
});
return(new StructureMatchResultSet(results));
}
}
}
}
// Call the base method
return(base.Match(snapshotOffset, options));
}
