textWidth or font problem when exporting to PDF

Core Library Questions

ali_mi

textWidth or font problem when exporting to PDF

in Core Library Questions • 1 month ago

The following code writes two lines with the same font and settings, but the first one using the existing text() method, while the other one uses a function called drawChars() which writes a given string character by character. Each line of text has a line underneath (I draw it as a way of measuring the length of the strings) that is supposed to have the same length as the text string, however one of the strings is displayed differently, looks shorter, like more squeezed as in the following image:

When I switch from size(800, 400, PDF, "test.pdf") at the beginning to simply size(800, 400) and also comment out the exit() command in order to display the text on the screen instead of writing it to a PDF file, everything looks as expected: Both lines of strings have the same length. So there must be something wrong with the way the font is drawn on the PDF file and Processing not being able to give the right information about it when textWidth() is called. I had the same results both with and without textMode(SHAPE). Here is the code with PDF export that displays the above results, with Processing 2.0.1 on OSX 10.7.5:

import processing.pdf.*;

String msg = "Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Nam cursus. Morbi ut mi. Nullam enim leo, egestas id, condimentum at, laoreet mattis, massa.";

PFont f;

void setup() {

size(800, 400, PDF, "test.pdf");

//size(800, 400);

textMode(SHAPE);

f = createFont("HelveticaNeue", 11, true);

smooth();

noLoop();

}

void draw() {

background(255);

noFill();

int xLine,yLine;

textFont(f);

yLine=50;

xLine=10;

//drawChars(msg, 10, yLine);

fill(0);

text(msg, xLine, yLine);

noFill();

strokeWeight(1);

line(xLine, yLine, xLine+textWidth(msg), yLine);

yLine=100;

drawChars(msg, 10, yLine);

noFill();

strokeWeight(1);

line(xLine, yLine, xLine+textWidth(msg), yLine);

println("Finished.");

exit();

}

void drawChars(String s, int px, int py) {

println(textWidth(s));

fill(0);

float dx=0;

char currentChar;

for (int i = 0; i < s.length(); i++)

{

currentChar = s.charAt(i);

text(currentChar, px+dx, py);

dx += textWidth(currentChar);

}

}

I need this drawChars() function to draw strings character by character, while I also need to know the exact pixel length of the complete string being drawn. Right now there is an inconsistency between them. Any ideas about how to solve or work around this?

Replies(1)

dimkir

Re: textWidth or font problem when exporting to PDF

1 month ago

my best guess is that the reason for this difference in length is because:

when you output the whole string then Kerning is applied to your string, so the length of string WITH Kerning applied is LESS than sum of the individual characters.

http://en.wikipedia.org/wiki/Kerning

when you draw char by char, then NO KERNING is applied, and rendered string is just sum of the lengths of individual characters.

The simpliest thing for you to do, is to make your own function called textNoKerning(String, float, float) which will just print character by character, and thus your strings will be compliant.

Other solution is based on determening length's of words with textWidth() and avoid operating on the level of individual characters.

Top Reply