public virtual string GetResultantText()
{
if (LocationTextExtractionStrategy2.DUMP_STATE)
{
this.DumpState();
}
IList <LocationTextExtractionStrategy2.TextChunk> list = new List <LocationTextExtractionStrategy2.TextChunk>(this.locationalResult);
this.SortWithMarks(list);
StringBuilder stringBuilder = new StringBuilder();
LocationTextExtractionStrategy2.TextChunk textChunk = null;
foreach (LocationTextExtractionStrategy2.TextChunk current in list)
{
if (textChunk == null)
{
stringBuilder.Append(current.text);
}
else
{
if (current.SameLine(textChunk))
{
if (this.IsChunkAtWordBoundary(current, textChunk) && !this.StartsWithSpace(current.text) && !this.EndsWithSpace(textChunk.text))
{
stringBuilder.Append(";;");
}
stringBuilder.Append(current.text);
}
else
{
stringBuilder.Append('\n');
stringBuilder.Append(current.text);
}
}
textChunk = current;
}
return(stringBuilder.ToString());
}
public virtual void EventOccurred(IEventData data, EventType type)
{
if (type.Equals(EventType.RENDER_TEXT))
{
TextRenderInfo textRenderInfo = (TextRenderInfo)data;
LineSegment lineSegment = textRenderInfo.GetBaseline();
if (textRenderInfo.GetRise() != 0f)
{
Matrix m = new Matrix(0f, -textRenderInfo.GetRise());
lineSegment = lineSegment.TransformBy(m);
}
if (this.useActualText)
{
CanvasTag canvasTag = (this.lastTextRenderInfo != null) ? this.FindLastTagWithActualText(this.lastTextRenderInfo.GetCanvasTagHierarchy()) : null;
if (canvasTag != null && canvasTag == this.FindLastTagWithActualText(textRenderInfo.GetCanvasTagHierarchy()))
{
LocationTextExtractionStrategy2.TextChunk textChunk = this.locationalResult[this.locationalResult.Count - 1];
Vector startPoint = new Vector(Math.Min(textChunk.GetLocation().GetStartLocation().Get(0), lineSegment.GetStartPoint().Get(0)), Math.Min(textChunk.GetLocation().GetStartLocation().Get(1), lineSegment.GetStartPoint().Get(1)), Math.Min(textChunk.GetLocation().GetStartLocation().Get(2), lineSegment.GetStartPoint().Get(2)));
Vector endPoint = new Vector(Math.Max(textChunk.GetLocation().GetEndLocation().Get(0), lineSegment.GetEndPoint().Get(0)), Math.Max(textChunk.GetLocation().GetEndLocation().Get(1), lineSegment.GetEndPoint().Get(1)), Math.Max(textChunk.GetLocation().GetEndLocation().Get(2), lineSegment.GetEndPoint().Get(2)));
LocationTextExtractionStrategy2.TextChunk value = new LocationTextExtractionStrategy2.TextChunk(textChunk.GetText(), this.tclStrat.CreateLocation(textRenderInfo, new LineSegment(startPoint, endPoint)));
this.locationalResult[this.locationalResult.Count - 1] = value;
}
else
{
string actualText = textRenderInfo.GetActualText();
LocationTextExtractionStrategy2.TextChunk item = new LocationTextExtractionStrategy2.TextChunk((actualText != null) ? actualText : textRenderInfo.GetText(), this.tclStrat.CreateLocation(textRenderInfo, lineSegment));
this.locationalResult.Add(item);
}
}
else
{
LocationTextExtractionStrategy2.TextChunk item2 = new LocationTextExtractionStrategy2.TextChunk(textRenderInfo.GetText(), this.tclStrat.CreateLocation(textRenderInfo, lineSegment));
this.locationalResult.Add(item2);
}
this.lastTextRenderInfo = textRenderInfo;
}
}
/// <summary>Determines if a space character should be inserted between a previous chunk and the current chunk.
/// </summary>
/// <remarks>
/// Determines if a space character should be inserted between a previous chunk and the current chunk.
/// This method is exposed as a callback so subclasses can fine time the algorithm for determining whether a space should be inserted or not.
/// By default, this method will insert a space if the there is a gap of more than half the font space character width between the end of the
/// previous chunk and the beginning of the current chunk. It will also indicate that a space is needed if the starting point of the new chunk
/// appears *before* the end of the previous chunk (i.e. overlapping text).
/// </remarks>
/// <param name="chunk">the new chunk being evaluated</param>
/// <param name="previousChunk">the chunk that appeared immediately before the current chunk</param>
/// <returns>true if the two chunks represent different words (i.e. should have a space between them). False otherwise.
/// </returns>
protected internal virtual bool IsChunkAtWordBoundary(LocationTextExtractionStrategy2.TextChunk chunk, LocationTextExtractionStrategy2.TextChunk previousChunk)
{
return(chunk.GetLocation().IsAtWordBoundary(previousChunk.GetLocation()));
}
