Unicode multi_cell line wrap breaks after parenthesis

[davidnoyes]

I'm having an issue when using multi_cell with a unicode font, it seems like I get corrupted characters when a line break occurs and the line contains a parenthesis before the line break. This doesn't occur when I use pdf.write.

Python: 3.9.2
fpdf2: 2.3.1

I've been able to reproduce the issue with a sample project:

#!/usr/bin/env python
# -*- coding: utf-8 -*-

from fpdf import FPDF

pdf = FPDF()
pdf.add_page()

# Add a DejaVu Unicode font (uses UTF-8)
# Supports more than 200 languages. For a coverage status see:
# http://dejavu.svn.sourceforge.net/viewvc/dejavu/trunk/dejavu-fonts/langcover.txt
pdf.add_font('DejaVu', '', 'DejaVuSans.ttf', uni=True)
pdf.set_font('DejaVu', '', 14)

text = """

Line wrap : lorem () dolor sit amet, consectetur sit amet adipiscing elit et dolore
No wrap : ipsum () dolor sit amet, consectetur sit amet adipiscing
Parenthesis after line break : ipsum dolor sit amet, consectetur adipiscing et dolore elit () lorem

"""

pdf.set_font('helvetica', '', 18)

pdf.write(8, "using pdf.multi_cell with built in font")

pdf.ln(0)

pdf.multi_cell(w=0, h=8, txt=text, ln=1)

pdf.ln(8)

pdf.add_font('DejaVu', '', 'DejaVuSans.ttf', uni=True)
pdf.set_font('DejaVu', '', 14)

pdf.write(8, "using pdf.multi_cell with unicode font")

pdf.ln(0)

pdf.multi_cell(w=0, h=8, txt=text, ln=1)

pdf.ln(8)

pdf.write(10, "using pdf.write, line by line with unicode font")
for line in text.split('\n'):
    pdf.write(8, line)
    pdf.ln(8)

pdf.write(8, "using pdf.write for whole literal with unicode font")
pdf.write(8, text)

pdf.ln(8)

pdf.output("unicode.pdf")

The code outputs this pdf:
unicode.pdf

Any help you could provide would be greatly appreciated. Thanks.

[Lucas-C]

This is indeed a bug. It seems to happen only in justified mode (align='J') with unicode fonts.
There is some special "manual" word spacing in that case:
https://github.com/PyFPDF/fpdf2/blob/master/fpdf/fpdf.py#L1209

[Lucas-C]

This bug will be solved by #119

Thanks for reporting it @raspberrypython !