# Preventing Text with Special Characters from Being Cut Off when Converting HTML to PDF

## Environment

| Version    | Product            | Author                                                                          |
| ---------- | ------------------ | ------------------------------------------------------------------------------- |
| 2024.3.806 | RadWordsProcessing | [Desislava Yordanova](https://www.telerik.com/blogs/author/desislava-yordanova) |

## Description

When you convert HTML documents to PDF format by using [RadWordsProcessing](https://www.telerik.com/document-processing-libraries/documentation/libraries/radwordsprocessing/overview) and its [PdfFormatProvider](https://www.telerik.com/document-processing-libraries/documentation/libraries/radwordsprocessing/formats-and-conversion/pdf/pdfformatprovider), text that contains special characters such as **å, ä, or ö** gets cut off. This issue occurs because the library requires explicit access to [font](https://www.telerik.com/document-processing-libraries/documentation/libraries/radpdfprocessing/concepts/fonts) data, which is **not** automatically available in the .NET Standard version of Telerik Document Processing.

This KB article also answers the following questions:

* How can I include special characters in PDFs using RadWordsProcessing?
* What steps are needed to support non-standard fonts in PDF conversion?
* How do I ensure all text displays correctly when converting HTML to PDF?

| Before                                                                                                                                                                    | After                                                                                                                                                                |
| ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| ![HTML to Pdf with Cut Off Text](https://www.telerik.com/document-processing-libraries/documentation/assets/e98ed3978af90d453605c5142117a073/html-to-pdf-cutoff-text.png) | ![HTML to Pdf with Full Text](https://www.telerik.com/document-processing-libraries/documentation/assets/ca70a9a9b4e0623b6f5870833a04e6fc/html-to-pdf-full-text.png) |

## Solution

To resolve the text cut-off issue and ensure all characters display correctly, implement a custom `FontsProvider`. This provider supplies the necessary font data to the PdfProcessing library and enables it to render all characters correctly.

1. Implement a custom [FontsProvider](https://www.telerik.com/document-processing-libraries/documentation/knowledge-base/pdfprocessing-implement-fontsprovider) by extending `FontsProviderBase` and override the `GetFontData` method. This method returns the font data for the required fonts, including those with special characters.

The following example shows how to handle *Tahoma, Arial, and Segoe UI* fonts. When you use other fonts, modify and extend the custom implementation with the respective fonts.

```csharp
    internal class FontsProvider : Telerik.Windows.Documents.Extensibility.FontsProviderBase
    {
        private readonly string fontFolder = Environment.GetFolderPath(Environment.SpecialFolder.Fonts);

        public override byte[] GetFontData(Telerik.Windows.Documents.Core.Fonts.FontProperties fontProperties)
        {
            string fontFamilyName = fontProperties.FontFamilyName;
            bool isBold = fontProperties.FontWeight == Telerik.Documents.Core.Fonts.FontWeights.Bold;
            if (fontFamilyName == "Tahoma" && isBold)
            {
                return GetFontDataFromFontFolder("tahomabd.ttf");
            }
            else if (fontFamilyName == "Tahoma")
            {
                return GetFontDataFromFontFolder("tahoma.ttf");
            }
            else if (fontFamilyName == "Arial" && isBold)
            {
                return GetFontDataFromFontFolder("arialbd.ttf");
            }
            else if (fontFamilyName == "Arial")
            {
                return GetFontDataFromFontFolder("arial.ttf");
            }
            else if (fontFamilyName == "Segoe UI")
            {
                return GetFontDataFromFontFolder("segoeui.ttf");
            }

            return null;
        }

        private byte[] GetFontDataFromFontFolder(string fontFileName)
        {
            using (FileStream fileStream = File.OpenRead(this.fontFolder + "\\" + fontFileName))
            {
                using (MemoryStream memoryStream = new MemoryStream())
                {
                    fileStream.CopyTo(memoryStream);
                    return memoryStream.ToArray();
                }
            }
        }
    }
```

2. Before you convert the HTML document to PDF, set the custom `FontsProvider` to the `FontsProvider` property of the `FixedExtensibilityManager`.

```csharp
    Telerik.Windows.Documents.Extensibility.FontsProviderBase fontsProvider = new FontsProvider();
    Telerik.Windows.Documents.Extensibility.FixedExtensibilityManager.FontsProvider = fontsProvider;
```

These steps provide access to the necessary font data and ensure all characters, including special ones, render correctly in the PDF document.

## See Also

* [PDF Format Provider](https://www.telerik.com/document-processing-libraries/documentation/libraries/radwordsprocessing/formats-and-conversion/pdf/pdfformatprovider)
* [Fonts in RadPdfProcessing](https://www.telerik.com/document-processing-libraries/documentation/libraries/radpdfprocessing/concepts/fonts)
* [How to Implement a FontsProvider](https://www.telerik.com/document-processing-libraries/documentation/knowledge-base/pdfprocessing-implement-fontsprovider)
* [RadWordsProcessing Documentation](https://www.telerik.com/document-processing-libraries/documentation/libraries/radwordsprocessing/overview)
* [Preserving the Font in PDF Export from Excel](https://www.telerik.com/document-processing-libraries/documentation/knowledge-base/preserve-font-boldness-pdf-export-radspreadprocessing)