Upload 4 files
Browse files- Dockerfile +20 -0
- main.py +35 -0
- requirements.txt +6 -0
- temp.py +21 -0
Dockerfile
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
FROM python:3.10
|
2 |
+
|
3 |
+
RUN useradd -m -u 1000 user
|
4 |
+
USER user
|
5 |
+
ENV PATH="/home/user/.local/bin:${PATH}"
|
6 |
+
WORKDIR /app
|
7 |
+
COPY --chown=user . /app
|
8 |
+
USER root
|
9 |
+
|
10 |
+
RUN apt-get update && apt-get install -y wget unzip && \
|
11 |
+
wget https://dl.google.com/linux/direct/google-chrome-stable_current_amd64.deb && \
|
12 |
+
apt install -y ./google-chrome-stable_current_amd64.deb && \
|
13 |
+
rm google-chrome-stable_current_amd64.deb && \
|
14 |
+
apt-get clean
|
15 |
+
|
16 |
+
USER user
|
17 |
+
RUN pip install --trusted-host pypi.python.org -r requirements.txt
|
18 |
+
COPY --chown=user . .
|
19 |
+
|
20 |
+
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]
|
main.py
ADDED
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
from fastapi import FastAPI
|
3 |
+
import time
|
4 |
+
from bs4 import BeautifulSoup
|
5 |
+
from selenium import webdriver
|
6 |
+
from selenium.webdriver.chrome.service import Service
|
7 |
+
from selenium.webdriver.chrome.options import Options
|
8 |
+
from webdriver_manager.chrome import ChromeDriverManager
|
9 |
+
chrome_options = Options()
|
10 |
+
chrome_options.add_argument('--no-sandbox')
|
11 |
+
chrome_options.add_argument('--headless')
|
12 |
+
chrome_options.add_argument('--disable-gpu')
|
13 |
+
chrome_options.add_argument('--disable-dev-shm-usage')
|
14 |
+
|
15 |
+
app = FastAPI()
|
16 |
+
favicon_path='https://www.iconarchive.com/download/i49313/martin-berube/sport/Soccer.ico'
|
17 |
+
|
18 |
+
@app.get("/")
|
19 |
+
def read_root():
|
20 |
+
return {"Hello": "World"}
|
21 |
+
@app.get("/selenium/")
|
22 |
+
async def read_item(reqUrl,tag):
|
23 |
+
driver = webdriver.Chrome(service=Service(ChromeDriverManager().install()),options=chrome_options)
|
24 |
+
url= reqUrl
|
25 |
+
driver.get(url)
|
26 |
+
time.sleep(3)
|
27 |
+
soup = BeautifulSoup(driver.page_source,'lxml')
|
28 |
+
headings = soup.find_all( name= tag)
|
29 |
+
a=[]
|
30 |
+
for heading in headings:
|
31 |
+
a.append(heading.getText())
|
32 |
+
driver.quit()
|
33 |
+
return {"Hello": a}
|
34 |
+
|
35 |
+
|
requirements.txt
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
selenium
|
2 |
+
beautifulsoup4
|
3 |
+
webdriver_manager
|
4 |
+
lxml
|
5 |
+
fastapi
|
6 |
+
uvicorn
|
temp.py
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import time
|
2 |
+
from bs4 import BeautifulSoup
|
3 |
+
from selenium import webdriver
|
4 |
+
from selenium.webdriver.chrome.service import Service
|
5 |
+
from selenium.webdriver.chrome.options import Options
|
6 |
+
from webdriver_manager.chrome import ChromeDriverManager
|
7 |
+
chrome_options = Options()
|
8 |
+
chrome_options.add_argument('--no-sandbox')
|
9 |
+
chrome_options.add_argument('--headless')
|
10 |
+
chrome_options.add_argument('--disable-gpu')
|
11 |
+
chrome_options.add_argument('--disable-dev-shm-usage')
|
12 |
+
driver = webdriver.Chrome(service=Service(ChromeDriverManager().install()),options=chrome_options)
|
13 |
+
url= 'https://navpan2-dockerbot.hf.space/'
|
14 |
+
driver.get(url)
|
15 |
+
time.sleep(5)
|
16 |
+
soup = BeautifulSoup(driver.page_source,'lxml')
|
17 |
+
headings = soup.find_all( name= 'span')
|
18 |
+
for heading in headings:
|
19 |
+
print(heading.getText())
|
20 |
+
time.sleep(2)
|
21 |
+
driver.quit()
|