From 17e68080b799c70c9ad31846add78eaa744ca525 Mon Sep 17 00:00:00 2001
From: Daohan Chong <wildcat.name@gmail.com>
Date: Sun, 6 Mar 2022 18:14:30 +0800
Subject: [PATCH] Fix docker deployment (#166)

* Docker fix

* Update docker build script
---
 Dockerfile                              | 44 +++++++++++++++----------
 requirements.txt                        |  2 +-
 scylla/providers/proxy_list_provider.py | 14 +++++---
 3 files changed, 37 insertions(+), 23 deletions(-)

diff --git a/Dockerfile b/Dockerfile
index d7f5aa6..b79fcd8 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -1,36 +1,44 @@
 FROM node:lts-buster as node-build
+
 WORKDIR /root
+
 COPY package.json .
 RUN yarn install
 COPY . .
 RUN make assets-build
 
+FROM ubuntu:focal as python-build
 
-FROM python:3.9-slim as python-build
-RUN apt-get update && apt-get install -y g++ gcc libxslt-dev make libcurl4-openssl-dev build-essential
-RUN apt-get install -y libssl-dev
-WORKDIR /root
+ENV DEBIAN_FRONTEND=noninteractive
+ENV TZ=America/Los_Angeles
+
+RUN apt-get update && \
+    apt-get install -y python3 python3-distutils libpython3.8-dev curl g++ gcc libxslt-dev make libcurl4-openssl-dev build-essential libssl-dev && \
+    update-alternatives --install /usr/bin/python python /usr/bin/python3 1 && \
+    curl -sSL https://bootstrap.pypa.io/get-pip.py -o get-pip.py && \
+    python get-pip.py && \
+    rm get-pip.py && \
+    # Feature-parity with node.js base images.
+    apt-get install -y --no-install-recommends git openssh-client && \
+    # clean apt cache
+    rm -rf /var/lib/apt/lists/* && \
+    # Create the pwuser
+    adduser pwuser
+
+WORKDIR /app
 
 COPY --from=node-build /root/scylla/assets ./scylla/assets
 COPY requirements.txt .
-RUN pip install -r requirements.txt
-RUN python -m playwright install chromium
+RUN pip3 install -r requirements.txt
 COPY . .
-RUN python setup.py install
+RUN python3 setup.py install
 
-FROM python:3.9-slim as prod
-
-LABEL maintainer="WildCat <wildcat.name@gmail.com>"
-
-RUN apt-get update && apt-get install -y libxslt-dev libssl-dev libcurl4-openssl-dev
-
-COPY --from=python-build /usr/local/lib/python3.9/site-packages/ /usr/local/lib/python3.9/site-packages/
-COPY --from=python-build /root/.cache/ms-playwright /root/.cache/ms-playwright
-
-WORKDIR /var/www/scylla
+RUN mkdir -p /var/www/scylla
 VOLUME /var/www/scylla
 
+RUN python3 -m playwright install chromium --with-deps
+
 EXPOSE 8899
 EXPOSE 8081
 
-CMD python -m scylla
\ No newline at end of file
+CMD python3 -m scylla --db-path /var/www/scylla/scylla.db
diff --git a/requirements.txt b/requirements.txt
index a68c576..4d91139 100755
--- a/requirements.txt
+++ b/requirements.txt
@@ -6,5 +6,5 @@ sanic==21.6.2
 sanic-cors
 schedule==1.1.0
 six==1.16.0
-playwright==1.13.1
+playwright==1.19.1
 pyquery==1.4.3
diff --git a/scylla/providers/proxy_list_provider.py b/scylla/providers/proxy_list_provider.py
index c3b70ba..0f245db 100644
--- a/scylla/providers/proxy_list_provider.py
+++ b/scylla/providers/proxy_list_provider.py
@@ -1,5 +1,6 @@
 import base64
 import re
+from typing import List
 
 from pyquery import PyQuery
 
@@ -14,14 +15,19 @@ def __init__(self):
         super().__init__()
         self.w = Worker()
 
-    def parse(self, document: PyQuery) -> [ProxyIP]:
-        ip_list: [ProxyIP] = []
+    def parse(self, document: PyQuery) -> List[ProxyIP]:
+        ip_list: List[ProxyIP] = []
 
         if document is None:
             return []
 
         for ul in document.find('#proxy-table > div.table-wrap ul'):
-            js_code = ul.find('li.proxy script').text()
+            js_code_element = ul.find('li.proxy script')
+
+            if not js_code_element:
+                return []
+
+            js_code = js_code_element.text()
             matched = re.findall(r"Proxy\('(.+)'\)", js_code)
             if matched and len(matched) > 0:
                 encoded = matched[0]
@@ -32,7 +38,7 @@ def parse(self, document: PyQuery) -> [ProxyIP]:
 
         return ip_list
 
-    def urls(self) -> [str]:
+    def urls(self) -> List[str]:
         ret = []
         first_url = 'http://proxy-list.org/english/index.php?p=1'
         first_page = self.w.get_html(first_url, False)