Fix non-latin languages processing when generating PDF files

stellarnode · stellarnode · commit 932f94eeb677 · 2025-07-20T16:41:12.000+03:00
diff --git a/youtubevideotranscriptbot/Dockerfile_template b/youtubevideotranscriptbot/Dockerfile_template
@@ -1,7 +1,32 @@
-# This is Dockerfile
 # Use an official Python runtime as a parent image
 FROM python:3.10-slim
 
+# Install system dependencies for WeasyPrint and fonts
+RUN apt-get update && apt-get install -y \
+    # WeasyPrint dependencies
+    libpango-1.0-0 \
+    libharfbuzz0b \
+    libpangoft2-1.0-0 \
+    libfontconfig1 \
+    libcairo2 \
+    libgdk-pixbuf2.0-0 \
+    libgtk-3-0 \
+    # Font packages for emoji support
+    fonts-dejavu-core \
+    fonts-dejavu-extra \
+    fonts-noto-core \
+    fonts-noto-cjk \
+    fonts-noto-color-emoji \
+    fonts-noto-extra \
+    fonts-liberation \
+    fontconfig \
+    # Clean up
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+# Update font cache
+RUN fc-cache -fv
+
 # Set the working directory in the container
 WORKDIR /app
 
@@ -28,4 +53,6 @@ RUN python3 -c "import config; import os; " \
     "print('Environment variables set from config.py')"
 
 # Run the bot
-CMD ["python", "telegram_bot.py"]
+CMD ["python", "telegram_bot.py"]
+
+
diff --git a/youtubevideotranscriptbot/utils.py b/youtubevideotranscriptbot/utils.py
@@ -117,14 +117,14 @@ def create_emoji_friendly_pdf_with_weasyprint(html_content, output_filename=None
     }
     
     body {
-        font-family: "DejaVu Sans", sans-serif;
+        font-family: "Noto Sans","DejaVu Sans", sans-serif;
         font-size: 12pt;
         line-height: 1.5;
         color: #333;
     }
     
     h1, h2, h3 {
-        font-family: "DejaVu Sans", sans-serif;
+        font-family: "Noto Sans","DejaVu Sans", sans-serif;
         color: #222;
     }
     

Original file line number	Diff line number	Diff line change
`@@ -117,14 +117,14 @@ def create_emoji_friendly_pdf_with_weasyprint(html_content, output_filename=None`
`117`	`117`	`}`
`118`	`118`
`119`	`119`	`body {`
`120`		`- font-family: "DejaVu Sans", sans-serif;`
	`120`	`+ font-family: "Noto Sans","DejaVu Sans", sans-serif;`
`121`	`121`	`font-size: 12pt;`
`122`	`122`	`line-height: 1.5;`
`123`	`123`	`color: #333;`
`124`	`124`	`}`
`125`	`125`
`126`	`126`	`h1, h2, h3 {`
`127`		`- font-family: "DejaVu Sans", sans-serif;`
	`127`	`+ font-family: "Noto Sans","DejaVu Sans", sans-serif;`
`128`	`128`	`color: #222;`
`129`	`129`	`}`
`130`	`130`