Spaces:
Running
Running
Update Dockerfile
Browse files- Dockerfile +10 -43
Dockerfile
CHANGED
@@ -4,61 +4,28 @@ FROM python:3.10-slim
|
|
4 |
# Set the working directory in the container
|
5 |
WORKDIR /app
|
6 |
|
7 |
-
# *** CHANGED HOME DIRECTORY ***
|
8 |
-
# Set the HOME environment variable to /tmp which should be writable.
|
9 |
-
# This should influence where crawl4ai attempts to create its .crawl4ai directory.
|
10 |
-
ENV HOME=/tmp
|
11 |
-
|
12 |
-
# Install system dependencies required by Playwright/crawl4ai
|
13 |
-
# Based on Playwright's recommendations for Debian/Ubuntu
|
14 |
-
RUN apt-get update && apt-get install -y --no-install-recommends \
|
15 |
-
# Base dependencies
|
16 |
-
wget \
|
17 |
-
gnupg \
|
18 |
-
# Playwright browser dependencies (Chromium)
|
19 |
-
libnss3 \
|
20 |
-
libatk1.0-0 \
|
21 |
-
libatk-bridge2.0-0 \
|
22 |
-
libcups2 \
|
23 |
-
libdrm2 \
|
24 |
-
libdbus-1-3 \
|
25 |
-
libxkbcommon0 \
|
26 |
-
libatspi2.0-0 \
|
27 |
-
libx11-6 \
|
28 |
-
libxcomposite1 \
|
29 |
-
libxdamage1 \
|
30 |
-
libxext6 \
|
31 |
-
libxfixes3 \
|
32 |
-
libxrandr2 \
|
33 |
-
libgbm1 \
|
34 |
-
libpango-1.0-0 \
|
35 |
-
libcairo2 \
|
36 |
-
libasound2 \
|
37 |
-
# Clean up
|
38 |
-
&& rm -rf /var/lib/apt/lists/*
|
39 |
-
|
40 |
# Copy the requirements file into the container at /app
|
41 |
COPY requirements.txt .
|
42 |
|
|
|
|
|
|
|
|
|
43 |
# Install Python dependencies
|
44 |
# Using --no-cache-dir reduces image size slightly
|
45 |
RUN pip install --no-cache-dir -r requirements.txt
|
46 |
|
47 |
-
# Run crawl4ai post-installation setup (installs browser binaries)
|
48 |
-
# This should now download browsers relative to HOME=/tmp or its internal logic.
|
49 |
-
RUN crawl4ai-setup
|
50 |
-
|
51 |
# Copy the rest of the application code into the container at /app
|
52 |
COPY . .
|
53 |
|
54 |
-
# Make port
|
55 |
-
#
|
56 |
-
|
57 |
-
EXPOSE ${PORT}
|
58 |
|
59 |
# Define environment variable to ensure Python output is sent straight to logs
|
60 |
ENV PYTHONUNBUFFERED=1
|
61 |
|
62 |
# Command to run the application using Gunicorn
|
63 |
-
#
|
64 |
-
|
|
|
|
4 |
# Set the working directory in the container
|
5 |
WORKDIR /app
|
6 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
# Copy the requirements file into the container at /app
|
8 |
COPY requirements.txt .
|
9 |
|
10 |
+
# Install any needed system dependencies (if any - bs4, requests usually don't need much)
|
11 |
+
# RUN apt-get update && apt-get install -y --no-install-recommends some-package && rm -rf /var/lib/apt/lists/*
|
12 |
+
# For this bot, we likely don't need extra apt packages currently.
|
13 |
+
|
14 |
# Install Python dependencies
|
15 |
# Using --no-cache-dir reduces image size slightly
|
16 |
RUN pip install --no-cache-dir -r requirements.txt
|
17 |
|
|
|
|
|
|
|
|
|
18 |
# Copy the rest of the application code into the container at /app
|
19 |
COPY . .
|
20 |
|
21 |
+
# Make port 7860 available to the world outside this container
|
22 |
+
# Hugging Face Spaces typically expect apps to run on port 7860
|
23 |
+
EXPOSE 7860
|
|
|
24 |
|
25 |
# Define environment variable to ensure Python output is sent straight to logs
|
26 |
ENV PYTHONUNBUFFERED=1
|
27 |
|
28 |
# Command to run the application using Gunicorn
|
29 |
+
# It will run the Flask 'app' object found in the 'main' module (main.py)
|
30 |
+
# Listen on all interfaces (0.0.0.0) on the port specified by HF (usually 7860)
|
31 |
+
CMD ["gunicorn", "--bind", "0.0.0.0:7860", "main:app"]
|