/** Returns true, if the given character should be considered on extraction. */
public static boolean considerPdfCharacter(PdfCharacter character) {
// Don't consider the character, if it is a diacritic.
if (character.isDiacritic()) {
return false;
}
PdfColor color = character.getColor();
// Don't consider the character, if its color is white.
if (color != null && color.isWhite(0.05f)) {
return false;
}
// Don't consider the character "|" because it is used as PARA_ADDENDUM in
// tex-paragraph-parser.
if ("|".equals(character.getUnicode().trim()) && character.getFontsize() < 6.5f) { // TODO
return false;
}
// Don't consider the character, if its orientation isn't 0.
if (character.getOrientation() != 0) {
return false;
}
// Don't consider the character, if its a proper rectangle.
if (character.getRectangle().getWidth() <= 0 || character.getRectangle().getHeight() <= 0) {
return false;
}
// If the character has a valid encoding, we have to consider it.
if (character.hasEncoding()) {
return true;
}
// Don't consider the character, if its doesn't hold any content.
if (character.getUnicode().trim().isEmpty()) {
return false;
}
return true;
}
/** Returns true, if the given shape should be considered on extraction. */
public static boolean considerPdfShape(PdfShape shape, Rectangle clippingPath) {
if (shape == null) {
return false;
}
PdfColor color = shape.getColor();
if (color == null) {
return false;
}
if (color.isWhite()) {
return false;
}
// If the clipping path is given, it must contain the shape.
if (clippingPath != null) {
return clippingPath.overlaps(shape.getRectangle());
}
return true;
}