Created
May 27, 2023 13:05
-
-
Save Sid72020123/952fa5494b1492f1b5797810ee73a2b9 to your computer and use it in GitHub Desktop.
The Python program that fetched the activity of a Scratch user
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from requests import get | |
from bs4 import BeautifulSoup | |
from pprint import pprint | |
Scratch_URL = "https://scratch.mit.edu" | |
def get_activity_type(raw_activity_type): | |
activities = { | |
"added": "studio-add", | |
"became a curator of": "studio-curator", | |
"loved": "project-love", | |
"favorited": "project-favorite", | |
"is now following": "user-follow", | |
"is now following the studio": "studio-follow", | |
"shared the project": "project-share", | |
"was promoted to manager of": "studio-manager", | |
"remixed": "project-remix", | |
"joined Scratch": "scratch-join", | |
} | |
return activities[raw_activity_type] | |
def get_activity(username): | |
response = get( | |
f"https://scratch.mit.edu/messages/ajax/user-activity/?user={username}&max=1000000", | |
headers={"User-Agent": "Python Requests"}, | |
).content | |
soup = BeautifulSoup(response, "html.parser") | |
activities = soup.find_all("li") | |
result = [] | |
for activity in activities: | |
div_contents = activity.find("div").contents | |
a = {} | |
activity_type = get_activity_type(div_contents[2].strip()) | |
activity_time = str(activity.find("span", "time").contents[0]).replace( | |
"\xa0", " " | |
) | |
a = {"Type": activity_type, "Time": activity_time, "Action": {}} | |
a_act = a["Action"] | |
if activity_type == "studio-add": | |
a_act["ProjectURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["ProjectName"] = f"{str(div_contents[3].contents[0])}" | |
a_act["StudioURL"] = f"{Scratch_URL}{div_contents[5]['href']}" | |
a_act["StudioName"] = f"{str(div_contents[5].contents[0])}" | |
elif activity_type == "studio-curator": | |
a_act["StudioURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["StudioName"] = f"{str(div_contents[3].contents[0])}" | |
elif activity_type == "project-love": | |
a_act["ProjectURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["ProjectName"] = f"{str(div_contents[3].contents[0])}" | |
elif activity_type == "project-favorite": | |
a_act["ProjectURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["ProjectName"] = f"{str(div_contents[3].contents[0])}" | |
elif activity_type == "user-follow": | |
a_act["UsernameURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["Username"] = f"{str(div_contents[3].contents[0])}" | |
elif activity_type == "studio-follow": | |
a_act["StudioURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["StudioName"] = f"{str(div_contents[3].contents[0])}" | |
elif activity_type == "project-share": | |
a_act["ProjectURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["ProjectName"] = f"{str(div_contents[3].contents[0])}" | |
elif activity_type == "studio-manager": | |
a_act["StudioURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["StudioName"] = f"{str(div_contents[3].contents[0])}" | |
elif activity_type == "project-remix": | |
a_act["ParentProjectURL"] = f"{Scratch_URL}{div_contents[3]['href']}" | |
a_act["ParentProjectName"] = f"{str(div_contents[3].contents[0])}" | |
a_act["NewProjectURL"] = f"{Scratch_URL}{div_contents[5]['href']}" | |
a_act["NewProjectName"] = f"{str(div_contents[5].contents[0])}" | |
result.append(a) | |
return result | |
activity = get_activity("programORdie") | |
pprint(activity) # Just pretty print the result... |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment