Missing characters in pdfs when you try to display text in Noto Fonts with different languages?
This is by design in noto fonts. Noto Tools already provides a way to merge fonts but this package is designed to streamline the process to generate the font files needed for pdfs without dropping characters.
In noto fonts latin characters are separated out from downloadable
.ttf files for non latin language sets. Specifically when using a tool like pdfmake we need to merge latin character sets with other languages to properly see latin (including english and number) characters.
Clone repository locally or install:
As npm module
npm install --save-dev nototools-pdf-fonts
- Then run
npm installwithin nototools-pdf-fonts module directory
- Then run
On Mac OS X, install dependencies with homebrew
brew install --with-cairo harfbuzz cairo pango pygtk imagemagick
Install python dependencies in directory,
pip install -r requirements.txt
Installing using 'editable' mode is recommended:
pip install -e .
Download fonts from noto fonts
Merge fonts using
merge_fonts.py - modify as needed(add downloaded font files to merge or skip this step to use already merged fonts provided in merged_fonts directory for the next step):
python3 dir_to_merge_fonts.py -d dir_to_notofonts_to_merge
.otf files from CJK (Chinese, Japanese and Korean) cannot be merged and instead should be built directly to base64 using the CJK font files found in the
merged_fonts directory which has been conveniently provided by Kaigen Fonts. Additional font weights can be supplied to this tool from there.
Convert merged fonts to base64 using this tool:
As a cloned repo
npx gulp buildFonts
As a npm module from root level of package.json
npx gulp --cwd 'directory_to_nototools-pdf-fonts_module' buildFonts
Find built files of base64 in
build directory of nototools-pdf-fonts module.