Slavic characters are replaced by ■ #792

Netvia · 2025-01-16T11:17:36Z

Describe the Bug

Slavic characters are replaced by ■. Standard Arial font knows this letters, where should be an error?

pdf is created with unreadable chars. I also tried updated this html (with file "arial.ttf" in the same directory), but there is the same ouptut:

<!DOCTYPE html>
<html>
    <head>
        <meta charset="UTF-8">
            <title>My HTML Document</title>
            <style>
                @font-face {
                    font-family: Arial;
                    src: url('/arial.ttf');
                }
                *{
                    font-family: Arial;
                    color: red;
                    }
            </style
        </head>
    <body>
        <h1>+ěščřžýáíéí</h1>
    </body>
</html>

Minimal Example to Reproduce

file input.html :

<!DOCTYPE html>
<html>
    <head>
        <meta charset="UTF-8">
            <title>My HTML Document</title>
            <style>
                *{
                    font-family: Arial;
                    color: red;
                    }
            </style
        </head>
    <body>
        <h1>+ěščřžýáíéí</h1>
    </body>
</html>

python code (utf-8 enconding too):

  OUT = "output.pdf"
  IN = "input.html"

  from xhtml2pdf import pisa
  with open(IN, 'r', encoding='utf-8') as html_file:
      html_content = html_file.read()

  with open(OUT, 'wb') as pdf_file:
      pisa_status = pisa.CreatePDF(html_content, dest=pdf_file, encoding='utf-8')

  if pisa_status.err:
      return False

Expected Behavior

+ěščřžýáíéí

Actual Behavior

+■š■■žýáíéí

System Information

OS version: Windows 11 24H2
Python version: 3.13.0, also on 3.10.7
XHTML2PDF version: 0.2.16

The text was updated successfully, but these errors were encountered:

PiotrCzapla · 2025-03-20T10:32:56Z

it seems the font is missing this character set, claude-code figure out a solution on its :), here is a working example:

import urllib.request
from io import BytesIO

from xhtml2pdf import pisa

font_url = "https://github.com/notofonts/noto-fonts/raw/main/hinted/ttf/NotoSans/NotoSans-Regular.ttf"
font_path = "./NotoSans-Regular.ttf"
# it is used by link_callback to find the font file
urllib.request.urlretrieve(font_url, font_path)

html = """
<!DOCTYPE html>
<html>
<head>
    <meta charset="UTF-8">
    <style>
        @font-face {
            font-family: 'NotoSans-Regular';
            src: url('{{font_url}}');
        }

        body {
            font-family: NotoSans-Regular, Arial, sans-serif;
            font-size: 10pt;
            line-height: 1.5;
            color: #333;
        }
        
        
        h1 {
            font-size: 18px;
            font-weight: bold;
        }
        
        .test-section {
            margin: 20px 0;
            padding: 10px;
            border: 1px solid #ccc;
        }
    </style>
</head>
<body>

    <h1>UTF-8 and Font Test</h1>
    
    <div class="test-section">
        <h2>Polish Characters:</h2>
        <p>ąćęłńóśźż ĄĆĘŁŃÓŚŹŻ</p>
    </div>
    
    <div class="test-section">
        <h2>Czech Characters:</h2>
        <p>ěščřžýáíéů ĚŠČŘŽÝÁÍÉŮ</p>
    </div>
    
    <div class="test-section">
        <h2>Russian Characters:</h2>
        <p>абвгдеёжзийклмнопрстуфхцчшщъыьэюя</p>
        <p>АБВГДЕЁЖЗИЙКЛМНОПРСТУФХЦЧШЩЪЫЬЭЮЯ</p>
    </div>
    
    <div class="test-section">
        <h2>Additional Test:</h2>
        <p>zażółć gęślą jaźń</p>
    </div>
</body>
</html>
""".replace(
    "{{font_url}}", font_url
)


# this bit is crucial to make the font work
def link_callback(uri, rel):
    if uri.startswith("file:///"):
        path = uri[7:]
        return path
    if uri == font_url:
        return "NotoSans-Regular.ttf"
    return uri


# Convert HTML to PDF
pdf_data = BytesIO()
conversion_result = pisa.CreatePDF(
    html, dest=pdf_data, encoding="UTF-8", link_callback=link_callback
)

# Check for errors
if conversion_result.err:
    print(f"Error during PDF creation: {conversion_result.err}")
else:
    # Write PDF to file
    with open("test_noto_font.pdf", "wb") as f:
        f.write(pdf_data.getvalue())
    print("PDF created: test_noto_font.pdf")

I've simplified the Claude-code solution to the relevant bits.

Netvia added the bug Something isn't working label Jan 16, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Uh oh!

Slavic characters are replaced by ■ #792

Slavic characters are replaced by ■ #792

Netvia commented Jan 16, 2025 •

edited

Loading

PiotrCzapla commented Mar 20, 2025 •

edited

Loading

Uh oh!

Uh oh!

Slavic characters are replaced by ■ #792

Slavic characters are replaced by ■ #792

Comments

Netvia commented Jan 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Describe the Bug

Minimal Example to Reproduce

Expected Behavior

Actual Behavior

System Information

PiotrCzapla commented Mar 20, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Netvia commented Jan 16, 2025 •

edited

Loading

PiotrCzapla commented Mar 20, 2025 •

edited

Loading