From 8f7e5f639b6d6457213bc2507ec7668982b65689 Mon Sep 17 00:00:00 2001
From: mohiit1502 <mohiit1502@gmail.com>
Date: Sat, 3 Jan 2026 14:21:04 +0530
Subject: [PATCH] feat: Add base image to avoid 489MB TensorFlow downloads

- Add Dockerfile.base with pre-installed TensorFlow/FAISS dependencies
- Update Dockerfile.api and Dockerfile.streamlit to use base image
- Base image only needs rebuild when requirements.txt changes
---
 Dockerfile.api       | 32 ++------------------------------
 Dockerfile.base      | 26 ++++++++++++++++++++++++++
 Dockerfile.streamlit | 32 ++------------------------------
 Jenkinsfile          |  4 ++++
 4 files changed, 34 insertions(+), 60 deletions(-)
 create mode 100644 Dockerfile.base

diff --git a/Dockerfile.api b/Dockerfile.api
index b18fbd1..8fa3de9 100644
--- a/Dockerfile.api
+++ b/Dockerfile.api
@@ -1,36 +1,10 @@
 # Reverse Image Search API
-# Multi-stage build for optimized image size
+# Uses pre-built base image with TensorFlow to avoid repeated downloads
 
-FROM python:3.11-slim as builder
+FROM harbor.armco.dev/library/visualsearch-base:latest
 
 WORKDIR /app
 
-# Install build dependencies
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    build-essential \
-    && rm -rf /var/lib/apt/lists/*
-
-# Copy requirements and install dependencies
-COPY requirements.txt .
-RUN pip install --no-cache-dir --user -r requirements.txt
-
-# Production stage
-FROM python:3.11-slim
-
-WORKDIR /app
-
-# Install runtime dependencies (libgomp for OpenMP parallelism)
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    libgomp1 \
-    && rm -rf /var/lib/apt/lists/*
-
-# FAISS uses OpenMP for parallel search - set threads for Xeon
-ENV OMP_NUM_THREADS=8
-
-# Copy installed packages from builder
-COPY --from=builder /root/.local /root/.local
-ENV PATH=/root/.local/bin:$PATH
-
 # Copy application code
 COPY api/ ./api/
 COPY config.py .
@@ -45,8 +19,6 @@ COPY filenames.pkl .
 RUN mkdir -p uploads logs
 
 # Environment variables
-ENV PYTHONUNBUFFERED=1
-ENV PYTHONDONTWRITEBYTECODE=1
 ENV API_PORT=8081
 ENV DEBUG=false
 
diff --git a/Dockerfile.base b/Dockerfile.base
new file mode 100644
index 0000000..6f49c8d
--- /dev/null
+++ b/Dockerfile.base
@@ -0,0 +1,26 @@
+# Base image with TensorFlow and heavy dependencies pre-installed
+# Build once, reuse for API and Streamlit builds
+# Tag: harbor.armco.dev/library/visualsearch-base:latest
+
+FROM python:3.11-slim
+
+WORKDIR /app
+
+# Install build dependencies
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    build-essential \
+    libgomp1 \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy and install requirements
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# FAISS uses OpenMP for parallel search
+ENV OMP_NUM_THREADS=8
+ENV PATH=/root/.local/bin:$PATH
+ENV PYTHONUNBUFFERED=1
+ENV PYTHONDONTWRITEBYTECODE=1
+
+# Clean up build dependencies to reduce image size
+RUN apt-get purge -y build-essential && apt-get autoremove -y
diff --git a/Dockerfile.streamlit b/Dockerfile.streamlit
index f335955..b2466ec 100644
--- a/Dockerfile.streamlit
+++ b/Dockerfile.streamlit
@@ -1,36 +1,10 @@
 # Reverse Image Search Streamlit UI
-# Multi-stage build for optimized image size
+# Uses pre-built base image with TensorFlow to avoid repeated downloads
 
-FROM python:3.11-slim as builder
+FROM harbor.armco.dev/library/visualsearch-base:latest
 
 WORKDIR /app
 
-# Install build dependencies
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    build-essential \
-    && rm -rf /var/lib/apt/lists/*
-
-# Copy requirements and install dependencies
-COPY requirements.txt .
-RUN pip install --no-cache-dir --user -r requirements.txt
-
-# Production stage
-FROM python:3.11-slim
-
-WORKDIR /app
-
-# Install runtime dependencies (libgomp for OpenMP parallelism)
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    libgomp1 \
-    && rm -rf /var/lib/apt/lists/*
-
-# FAISS uses OpenMP for parallel search - set threads for Xeon
-ENV OMP_NUM_THREADS=8
-
-# Copy installed packages from builder
-COPY --from=builder /root/.local /root/.local
-ENV PATH=/root/.local/bin:$PATH
-
 # Copy application code
 COPY config.py .
 COPY reverse_icon_search.py .
@@ -44,8 +18,6 @@ COPY filenames.pkl .
 RUN mkdir -p uploads logs
 
 # Environment variables
-ENV PYTHONUNBUFFERED=1
-ENV PYTHONDONTWRITEBYTECODE=1
 ENV STREAMLIT_SERVER_PORT=3000
 ENV STREAMLIT_SERVER_ADDRESS=0.0.0.0
 
diff --git a/Jenkinsfile b/Jenkinsfile
index c9d773d..c180d3e 100644
--- a/Jenkinsfile
+++ b/Jenkinsfile
@@ -1,5 +1,9 @@
 @Library('jenkins-shared@main') _
 
+// NOTE: Build base image first when requirements.txt changes:
+//   [imageName: 'visualsearch-base', dockerfile: 'Dockerfile.base']
+// Then rebuild API and Streamlit images
+
 kanikoPipeline(
   repoName: 'visual-search-engine',
   branch: env.BRANCH_NAME ?: 'main',