public List <string> GetColumnAsStrings(string column, bool fuzzy = true)
{
PdfTextElementColumn columnData = GetColumn(column, fuzzy);
// Emit result
var result = new List <string>();
foreach (PdfTextElement elem in columnData.Elements)
{
result.Add(elem.VisibleText);
}
return(result);
}
public PdfTextElementColumn GetColumn(string column, bool fuzzy = true)
{
PdfTextElement columnHead = FindElementByText(column, fuzzy);
if (columnHead == null)
{
return(PdfTextElementColumn.Empty);
}
double headY = columnHead.GetY() - columnHead.VisibleHeight;
double headX1 = columnHead.GetX();
double headX2 = headX1 + columnHead.VisibleWidth;
// Determine horizontal extent
double extentX1 = double.MinValue;
double extentX2 = double.MaxValue;
foreach (PdfTextElement elem in _textElements)
{
if (elem == columnHead)
{
continue;
}
if (TextElementHorizontalIntersection(columnHead, elem) == false)
{
continue;
}
double elemX1 = elem.GetX();
double elemX2 = elemX1 + elem.VisibleWidth;
if (elemX2 < headX1)
{
if (elemX2 > extentX1)
{
extentX1 = elemX2;
}
}
if (elemX1 > headX2)
{
if (elemX1 < extentX2)
{
extentX2 = elemX1;
}
}
}
PdfTextElementColumn columnData = GetColumn(columnHead, headY, headX1, headX2, extentX1, extentX2);
return(columnData);
}
public Bitmap RenderColumn(PdfTextElementColumn columnData, Bitmap bmp = null)
{
Graphics gc;
if (bmp == null)
{
bmp = new Bitmap(_pageWidth * _scale, _pageHeight * _scale, PixelFormat.Format32bppArgb);
gc = Graphics.FromImage(bmp);
gc.Clear(Color.White);
}
else
{
gc = Graphics.FromImage(bmp);
}
// Draw text elements of the column header
using (Pen penTextElem = new Pen(Color.Green))
using (Pen penCharElem = new Pen(Color.DarkGreen))
{
DrawTextElement(columnData.HeadTextElement, gc, penTextElem, penCharElem, _scale, _pageHeight, _pageRect.XMin, _pageRect.YMin, Brushes.Olive);
}
// Draw text elements of the column
using (Pen penTextElem = new Pen(Color.Red))
using (Pen penCharElem = new Pen(Color.DarkRed))
{
foreach (PdfTextElement textElement in columnData.Elements)
{
DrawTextElement(textElement, gc, penTextElem, penCharElem, _scale, _pageHeight, _pageRect.XMin, _pageRect.YMin, Brushes.OrangeRed);
}
}
// Draw column extents
using (Pen penColumn = new Pen(Color.Red))
{
float y = (float)(_pageRect.YMax - columnData.Y);
float x1 = (float)(columnData.X1 - _pageRect.XMin);
float x2 = (float)(columnData.X2 - _pageRect.XMin);
gc.DrawLine(penColumn, x1 * _scale, y * _scale, x2 * _scale, y * _scale);
gc.DrawLine(penColumn, x1 * _scale, y * _scale, x1 * _scale, _pageHeight * _scale);
gc.DrawLine(penColumn, x2 * _scale, y * _scale, x2 * _scale, _pageHeight * _scale);
}
gc.Dispose();
return(bmp);
}
public PdfTextElementColumn GetColumn(PdfTextElement columnHead, double headY, double headX1, double headX2, double extentX1, double extentX2)
{
// Get all the elements that intersects vertically, are down and sort results
var columnDataRaw = new List <PdfTextElement>();
foreach (PdfTextElement elem in _textElements)
{
if (TextElementVerticalIntersection(elem, headX1, headX2) == false)
{
continue;
}
// Only intems down the column
double elemY = elem.GetY();
if (elemY >= headY)
{
continue;
}
columnDataRaw.Add(elem);
}
columnDataRaw = columnDataRaw.OrderByDescending(elem => elem.GetY()).ToList();
// Only items completelly inside extents, try spliting big elements and break on big elements that can't be splitted
var columnElements = new List <PdfTextElement>();
foreach (PdfTextElement elem in columnDataRaw)
{
double elemX1 = elem.GetX();
double elemX2 = elemX1 + elem.VisibleWidth;
// Add elements completely inside
if (elemX1 > extentX1 && elemX2 < extentX2)
{
columnElements.Add(elem);
continue;
}
// Try to split elements intersecting extents of the column
double maxSpacing = elem.Characters.Average(c => c.Width) / 10;
int indexStart = 0;
int indexEnd = elem.Characters.Count - 1;
bool indexStartValid = true;
bool indexEndValid = true;
if (elemX1 < extentX1)
{
// Search best start
int index = 0;
double characterPosition = elemX1 + elem.Characters[index].Displacement;
while (characterPosition < extentX1 && index < (elem.Characters.Count - 1))
{
index++;
characterPosition = elemX1 + elem.Characters[index].Displacement;
}
double spacing = elem.GetCharacterPreviousSpacing(index);
while (spacing < maxSpacing && index < (elem.Characters.Count - 1))
{
index++;
spacing = elem.GetCharacterPreviousSpacing(index);
}
if (spacing < maxSpacing)
{
indexStartValid = false;
}
indexStart = index;
}
if (elemX2 > extentX2)
{
// Search best end
int index = elem.Characters.Count - 1;
double characterPosition = elemX1 + elem.Characters[index].Displacement + elem.Characters[index].Width;
while (characterPosition > extentX2 && index > 0)
{
index--;
characterPosition = elemX1 + elem.Characters[index].Displacement + elem.Characters[index].Width;
}
double spacing = elem.GetCharacterPrecedingSpacing(index);
while (spacing < maxSpacing && index > 0)
{
index--;
spacing = elem.GetCharacterPrecedingSpacing(index);
}
if (spacing < maxSpacing)
{
indexEndValid = false;
}
indexEnd = index;
}
// Break when there is no good split, spaning all extent
if (indexStartValid == false && indexEndValid == false)
{
break;
}
// Continue when only one of the sides is invalid. (outside elements intersecting extents of the column)
if (indexStartValid == false || indexEndValid == false)
{
continue;
}
// Add splitted element
columnElements.Add(elem.SubPart(indexStart, indexEnd + 1));
}
var columnData = new PdfTextElementColumn(columnHead, columnElements, headY, extentX1, extentX2);
return(columnData);
}