public static Paragraph AttachNewParagraph()
{
Vector3 pos;
if (_instance.paragraphs.Count > 0)
{
TextChar lastChar = _instance.paragraphs [_instance.paragraphs.Count - 1].IndexToTextChar(_instance.paragraphs [_instance.paragraphs.Count - 1].Count - 1);
pos = lastChar.transform.position;
pos.x += lastChar.spacing;
}
else
{
pos = Vector3.zero;
}
GameObject newParagraphObject = Instantiate(_instance.paragraphPrefab.gameObject, pos, Quaternion.identity) as GameObject;
Paragraph p = newParagraphObject.GetComponent <Paragraph> ();
_instance.paragraphs.Add(p);
return(p);
}
public void AddChar(TextCharDef def)
{
GameObject newCharObject = Instantiate(letterPrefab.gameObject) as GameObject;
if (textChars.Count == 0)
{
newCharObject.transform.position = transform.position;
}
else
{
newCharObject.transform.position = textChars[textChars.Count - 1].transform.position + new Vector3(def.spacing, 0, 0);
}
newCharObject.transform.SetParent(transform);
TextChar newChar = newCharObject.GetComponent <TextChar> ();
newChar.theChar = def.theChar;
newChar.textSize = def.size;
newChar.textColor = def.color;
newChar.fontSize = def.fontSize;
newChar.font = def.font;
newChar.spacing = def.spacing;
textChars.Add(newChar);
}
/**
* <summary>Sorts the extracted text strings.</summary>
* <remarks>Sorting implies text position ordering, integration and aggregation.</remarks>
* <param name="rawTextStrings">Source (lower-level) text strings.</param>
* <param name="textStrings">Target (higher-level) text strings.</param>
*/
private void Sort(
List <ContentScanner.TextStringWrapper> rawTextStrings,
List <ITextString> textStrings
)
{
// Sorting the source text strings...
{
TextStringPositionComparer <ContentScanner.TextStringWrapper> positionComparator = new TextStringPositionComparer <ContentScanner.TextStringWrapper>();
rawTextStrings.Sort(positionComparator);
}
// Aggregating and integrating the source text strings into the target ones...
TextString textString = null;
TextStyle textStyle = null;
TextChar previousTextChar = null;
bool dehyphenating = false;
foreach (ContentScanner.TextStringWrapper rawTextString in rawTextStrings)
{
/*
* NOTE: Contents on the same line are grouped together within the same text string.
*/
// Add a new text string in case of new line!
if (textString != null &&
textString.TextChars.Count > 0 &&
!TextStringPositionComparer <ITextString> .IsOnTheSameLine(textString.Box.Value, rawTextString.Box.Value))
{
if (dehyphenated &&
previousTextChar.Value == '-') // Hyphened word.
{
textString.TextChars.Remove(previousTextChar);
dehyphenating = true;
}
else // Full word.
{
// Add synthesized space character!
textString.TextChars.Add(
new TextChar(
' ',
new RectangleF(
previousTextChar.Box.Right,
previousTextChar.Box.Top,
0,
previousTextChar.Box.Height
),
textStyle,
true
)
);
textString = null;
dehyphenating = false;
}
previousTextChar = null;
}
if (textString == null)
{
textStrings.Add(textString = new TextString());
}
textStyle = rawTextString.Style;
double spaceWidth = textStyle.GetWidth(' ') * .5;
foreach (TextChar textChar in rawTextString.TextChars)
{
if (previousTextChar != null)
{
/*
* NOTE: PDF files may have text contents omitting space characters,
* so they must be inferred and synthesized, marking them as virtual
* in order to allow the user to distinguish between original contents
* and augmented ones.
*/
if (!textChar.Contains(' ') &&
!previousTextChar.Contains(' '))
{
float charSpace = textChar.Box.X - previousTextChar.Box.Right;
if (charSpace > spaceWidth)
{
// Add synthesized space character!
textString.TextChars.Add(
previousTextChar = new TextChar(
' ',
new RectangleF(
previousTextChar.Box.Right,
textChar.Box.Y,
charSpace,
textChar.Box.Height
),
textStyle,
true
)
);
}
}
else if (dehyphenating &&
previousTextChar.Contains(' '))
{
textStrings.Add(textString = new TextString());
dehyphenating = false;
}
}
textString.TextChars.Add(previousTextChar = textChar);
}
}
}
/**
* <summary>Sorts the extracted text strings.</summary>
* <remarks>Sorting implies text position ordering, integration and aggregation.</remarks>
* <param name="rawTextStrings">Source (lower-level) text strings.</param>
* <param name="textStrings">Target (higher-level) text strings.</param>
*/
private void Sort(
List <ContentScanner.TextStringWrapper> rawTextStrings,
List <ITextString> textStrings
)
{
// Sorting the source text strings...
{
TextStringPositionComparer <ContentScanner.TextStringWrapper> positionComparator = new TextStringPositionComparer <ContentScanner.TextStringWrapper>();
rawTextStrings.Sort(positionComparator);
}
// Aggregating and integrating the source text strings into the target ones...
TextString textString = null;
TextChar previousTextChar = null;
foreach (ContentScanner.TextStringWrapper rawTextString in rawTextStrings)
{
/*
* NOTE: Contents on the same line are grouped together within the same text string.
*/
// Add a new text string in case of new line!
if (textString == null ||
(textString.TextChars.Count > 0 &&
!TextStringPositionComparer <ITextString> .IsOnTheSameLine(
textString.Box.Value,
rawTextString.Box.Value)))
{
textStrings.Add(textString = new TextString());
previousTextChar = null;
}
TextStyle textStyle = rawTextString.Style;
float spaceWidth = 0;
try
{ spaceWidth = textStyle.Font.GetWidth(' ', textStyle.FontSize); }
catch
{ /* NOOP. */ }
if (spaceWidth == 0)
{
spaceWidth = textStyle.FontSize * .25f;
} // NOTE: as a rule of thumb, space width is estimated according to the font size.
foreach (TextChar textChar in rawTextString.TextChars)
{
if (previousTextChar != null)
{
/*
* NOTE: PDF files may have text contents omitting space characters,
* so they must be inferred and synthesized, marking them as virtual
* in order to allow the user to distinguish between original contents
* and augmented ones.
*/
float characterSpace = textChar.Box.X - previousTextChar.Box.Right;
if (characterSpace >= spaceWidth)
{
// Add synthesized space character!
textString.TextChars.Add(
new TextChar(
' ',
new RectangleF(
previousTextChar.Box.Right,
textChar.Box.Y,
characterSpace,
textChar.Box.Height
),
textStyle,
true
)
);
}
}
textString.TextChars.Add(previousTextChar = textChar);
}
}
}
