Update Dockerfile

Dockerfile

@@ -1,165 +1,47 @@
-# LibreTranslate Dockerfile with all 22 Indian Regional Languages
-FROM libretranslate/libretranslate:latest
+# Multi-stage build for optimal image size
+FROM python:3.11.11-slim-bullseye AS builder
 
-# Set environment variables for Indian languages
-ENV LT_HOST=0.0.0.0
-ENV LT_PORT=7860
-ENV LT_THREADS=6
-ENV LT_CHAR_LIMIT=20000
-ENV LT_LOAD_ONLY=en,hi,bn,te,mr,ta,ur,gu,kn,ml,or,pa,as,mai,mag,bh,sa,ne,ks,sd,kok,mni,sat,doi,bo
+WORKDIR /app
 
-# Create necessary directories
-USER root
-RUN mkdir -p /app/db /home/libretranslate/.local/share/argos-translate/packages
-RUN chown -R libretranslate:libretranslate /app/db /home/libretranslate/.local/share/argos-translate/packages
+ARG DEBIAN_FRONTEND=noninteractive
+RUN apt-get update -qq \
+  && apt-get -qqq install --no-install-recommends -y pkg-config gcc g++ \
+  && apt-get upgrade --assume-yes \
+  && apt-get clean \
+  && rm -rf /var/lib/apt
 
-# Switch back to libretranslate user
-USER libretranslate
-
-# Pre-install all 22 Indian regional languages plus English
-RUN python -c "
-import argostranslate.package
-import argostranslate.translate
-
-# Download and install language packages
-available_packages = argostranslate.package.get_available_packages()
-
-# All 22 official Indian languages (ISO 639-1/639-3 codes)
-indian_languages = [
-    'hi',  # Hindi
-    'bn',  # Bengali
-    'te',  # Telugu
-    'mr',  # Marathi
-    'ta',  # Tamil
-    'ur',  # Urdu
-    'gu',  # Gujarati
-    'kn',  # Kannada
-    'ml',  # Malayalam
-    'or',  # Odia
-    'pa',  # Punjabi
-    'as',  # Assamese
-    'mai', # Maithili
-    'mag', # Magahi
-    'bh',  # Bhojpuri
-    'sa',  # Sanskrit
-    'ne',  # Nepali
-    'ks',  # Kashmiri
-    'sd',  # Sindhi
-    'kok', # Konkani
-    'mni', # Manipuri
-    'sat', # Santali
-    'doi', # Dogri
-    'bo',  # Bodo
-]
-
-# Create comprehensive language pairs
-language_pairs = []
+RUN python -mvenv venv && ./venv/bin/pip install --no-cache-dir --upgrade pip
 
-# Primary focus: English to/from all Indian languages (PRIORITY PAIRS)
-print('Adding English <-> Indian language pairs...')
-for lang in indian_languages:
-    language_pairs.extend([
-        ('en', lang),  # English to Indian language - HIGH PRIORITY
-        (lang, 'en'),  # Indian language to English - HIGH PRIORITY
-    ])
+COPY . .
 
-# Add ALL possible inter-Indian language combinations
-print('Adding inter-Indian language pairs...')
-for i, lang1 in enumerate(indian_languages):
-    for j, lang2 in enumerate(indian_languages):
-        if i != j:  # Don't add same language pairs
-            language_pairs.append((lang1, lang2))
+# Install package from source code, compile translations
+RUN ./venv/bin/pip install Babel==2.12.1 && ./venv/bin/python scripts/compile_locales.py \
+  && ./venv/bin/pip install torch==2.2.0 --extra-index-url https://download.pytorch.org/whl/cpu \
+  && ./venv/bin/pip install "numpy<2" \
+  && ./venv/bin/pip install . \
+  && ./venv/bin/pip cache purge
 
-# Additional international language pairs with Indian languages
-international_languages = ['es', 'fr', 'de', 'it', 'pt', 'ru', 'zh', 'ja', 'ar', 'ko']
-print('Adding international language pairs with Indian languages...')
-for indian_lang in indian_languages:
-    for intl_lang in international_languages:
-        language_pairs.extend([
-            (indian_lang, intl_lang),  # Indian to International
-            (intl_lang, indian_lang),  # International to Indian
-        ])
+FROM python:3.11.11-slim-bullseye
 
-# Install available packages with priority system
-installed_count = 0
-failed_count = 0
-priority_installed = 0
+ARG with_models=false
+ARG models=""
 
-print('Starting installation of language models...')
-print(f'Total language pairs to attempt: {len(language_pairs)}')
-print('Priority: English <-> Indian languages first')
-
-# Separate English-Indian pairs for priority installation
-english_indian_pairs = [(f, t) for f, t in language_pairs if 'en' in (f, t)]
-other_pairs = [(f, t) for f, t in language_pairs if 'en' not in (f, t)]
-
-print(f'English-Indian priority pairs: {len(english_indian_pairs)}')
-print(f'Other language pairs: {len(other_pairs)}')
-
-# Install English-Indian pairs first (PRIORITY)
-print('\\n=== Installing Priority Pairs (English <-> Indian) ===')
-for from_code, to_code in english_indian_pairs:
-    package_to_install = next(
-        (pkg for pkg in available_packages 
-         if pkg.from_code == from_code and pkg.to_code == to_code), 
-        None
-    )
-    if package_to_install:
-        try:
-            argostranslate.package.install_from_path(package_to_install.download())
-            print(f'✓ PRIORITY: Successfully installed {from_code}->{to_code}')
-            installed_count += 1
-            priority_installed += 1
-        except Exception as e:
-            print(f'✗ PRIORITY: Failed to install {from_code}->{to_code}: {e}')
-            failed_count += 1
-    else:
-        print(f'⚠ PRIORITY: Package not available: {from_code}->{to_code}')
-        failed_count += 1
-
-# Install other pairs (inter-Indian and international)
-print('\\n=== Installing Additional Pairs (Inter-Indian & International) ===')
-for from_code, to_code in other_pairs:
-    package_to_install = next(
-        (pkg for pkg in available_packages 
-         if pkg.from_code == from_code and pkg.to_code == to_code), 
-        None
-    )
-    if package_to_install:
-        try:
-            argostranslate.package.install_from_path(package_to_install.download())
-            print(f'✓ Successfully installed {from_code}->{to_code}')
-            installed_count += 1
-        except Exception as e:
-            print(f'✗ Failed to install {from_code}->{to_code}: {e}')
-            failed_count += 1
-    else:
-        # Don't print warnings for unavailable inter-language pairs to reduce noise
-        failed_count += 1
+RUN addgroup --system --gid 1032 libretranslate && adduser --system --uid 1032 libretranslate && mkdir -p /home/libretranslate/.local && chown -R libretranslate:libretranslate /home/libretranslate/.local
+USER libretranslate
 
-print(f'\\n=== Final Installation Summary ===')
-print(f'Priority pairs (English <-> Indian): {priority_installed} installed')
-print(f'Total successfully installed: {installed_count} language pairs')
-print(f'Failed or unavailable: {failed_count} language pairs')
-print(f'\\n🎯 FOCUS: English to Indian Regional Languages Translation Server Ready!')
-print(f'📋 Supported: All available English <-> Indian language pairs')
-print(f'🌐 Bonus: Inter-Indian and international language pairs where available')
-print(f'\\n⚠️  Note: Some regional languages may not be available in Argos Translate yet.')
-print(f'✅ Available languages will be loaded automatically on server start.')
-print(f'🚀 Server optimized for English to Indian regional language translation!')
-"
+COPY --from=builder --chown=1032:1032 /app /app
+WORKDIR /app
 
-# Expose port
-EXPOSE 5000
+COPY --from=builder --chown=1032:1032 /app/venv/bin/ltmanage /usr/bin/
 
-# Health check optimized for Indian languages
-HEALTHCHECK --interval=30s --timeout=15s --start-period=60s --retries=3 \
-    CMD curl -f http://localhost:5000/health || exit 1
+RUN if [ "$with_models" = "true" ]; then  \
+  # initialize the language models
+  if [ ! -z "$models" ]; then \
+  ./venv/bin/python scripts/install_models.py --load_only_lang_codes "$models";   \
+  else \
+  ./venv/bin/python scripts/install_models.py;  \
+  fi \
+  fi
 
-# Default command with Indian language optimization
-CMD ["libretranslate", \
-     "--host", "0.0.0.0", \
-     "--port", "7860", \
-     "--threads", "6", \
-     "--char-limit", "20000", \
-     "--load-only", "en,hi,bn,te,mr,ta,ur,gu,kn,ml,or,pa,as,ne,sa"]
+EXPOSE 7860
+ENTRYPOINT [ "./venv/bin/libretranslate", "--host", "*" ]