added User.loves & User.loves_count

Author: FAReTek1

scratchattach/site/session.py

@@ -520,13 +520,14 @@ def connect_tw_cloud(self, project_id, *, purpose="", contact="", cloud_host="ws
         Returns:
             scratchattach.cloud.TwCloud: An object representing the TurboWarp cloud of a project.
         """
-        return cloud.TwCloud(project_id=project_id, purpose=purpose, contact=contact, cloud_host=cloud_host, _session=self)
+        return cloud.TwCloud(project_id=project_id, purpose=purpose, contact=contact, cloud_host=cloud_host) #, _session=self)
+                                                                                                             # PyCharm is giving a warning for this argument - unexpected argument
 
     # --- Connect classes inheriting from BaseSiteComponent ---
 
     def _make_linked_object(self, identificator_name, identificator, Class, NotFoundException):
         """
-        The Session class doesn't save the login in a ._session attribut, but IS the login ITSELF.
+        The Session class doesn't save the login in a ._session attribute, but IS the login ITSELF.
 
         Therefore the _make_linked_object method has to be adjusted
         to get it to work for in the Session class.
@@ -536,7 +537,7 @@ def _make_linked_object(self, identificator_name, identificator, Class, NotFound
         return commons._get_object(identificator_name, identificator, Class, NotFoundException, self)
 
 
-    def connect_user(self, username):
+    def connect_user(self, username) -> 'user.User':
         """
         Gets a user using this session, connects the session to the User object to allow authenticated actions
 
@@ -664,7 +665,7 @@ def connect_topic_list(self, category_id, *, page=1):
 
         Args:
             category_id (str): ID of the forum category
-        
+
         Keyword Arguments:
             page (str): Page of the category topics that should be returned
 
@@ -714,10 +715,10 @@ def connect_message_events(self, *, update_interval=2):
 
     def connect_filterbot(self, *, log_deletions=True):
         return filterbot.Filterbot(user.User(username=self.username, _session=self), log_deletions=log_deletions)
-        
+
 # ------ #
 
-def login_by_id(session_id, *, username=None, password=None, xtoken=None):
+def login_by_id(session_id, *, username=None, password=None, xtoken=None) -> Session:
     """
     Creates a session / log in to the Scratch website with the specified session id.
     Structured similarly to Session._connect_object method.
@@ -754,7 +755,7 @@ def login_by_id(session_id, *, username=None, password=None, xtoken=None):
             print(f"Warning: Logged in by id, but couldn't fetch session info. This won't affect any other features.")
     return _session
 
-def login(username, password, *, timeout=10):
+def login(username, password, *, timeout=10) -> Session:
     """
     Creates a session / log in to the Scratch website with the specified username and password.
 

scratchattach/site/user.py

@@ -3,6 +3,7 @@
 import json
 import random
 import string
+import math
 
 from ..eventhandlers import message_events
 from . import project
@@ -273,6 +274,97 @@ def projects(self, *, limit=40, offset=0):
             p["author"] = {"username":self.username}
         return commons.parse_object_list(_projects, project.Project, self._session)
 
+    def loves(self, *, limit=40, offset=0, get_full_project: bool=False) -> [project.Project]:
+        """
+        Returns:
+            list<projects.projects.Project>: The user's loved projects
+        """
+        # We need to use beautifulsoup webscraping so we cant use the api_iterative function
+        if offset < 0:
+            raise exceptions.BadRequest("offset parameter must be >= 0")
+        if limit < 0:
+            raise exceptions.BadRequest("limit parameter must be >= 0")
+
+        # There are 40 projects on display per page
+        # So number of pages to view the limit is ceil(limit / 40)
+
+        # The first page you need to view is 1 + offset // 40
+        # (You have to add one because the first page is idx 1 instead of 0)
+        pages = range(1 + offset // 40,
+                      1 + offset // 40 + math.ceil(limit / 40))
+        _projects = []
+
+        for page in pages:
+            # The index of the first project on page #n is just (n-1) * 40
+            first_idx = (page - 1) * 40
+
+            page_content = requests.get(f"https://scratch.mit.edu/projects/all/{self.username}/loves/"
+                             f"?page={page}", headers=self._headers).content
+            if b"Whoops! Our server is Scratch'ing its head" in page_content:
+                # We've either tried to access a non-existent account or
+                # the user hasn't loved enough projects (for the requested page to exist)
+                # So we can just break out of the loop early
+                break
+
+            soup = BeautifulSoup(
+                page_content,
+                "html.parser"
+            )
+
+            # Each project element is a list item with the class name 'project thumb item' so we can just use that
+            for i, project_element in enumerate(
+                    soup.find_all("li", {"class": "project thumb item"})):
+                # Remember we only want certain projects:
+                # The current project idx = first_idx + i
+                # We want to start at {offset} and end at {offset + limit}
+
+                # So the offset <= current project idx <= offset + limit
+                if offset <= first_idx + i <= offset + limit:
+                    # Each of these elements provides:
+                    # A project id
+                    # A thumbnail link (no need to webscrape this)
+                    # A title
+                    # An Author (called an owner for some reason)
+
+                    project_anchors = project_element.find_all("a")
+                    # Each list item has three <a> tags, the first two linking the project
+                    # 1st contains <img> tag
+                    # 2nd contains project title
+                    # 3rd links to the author & contains their username
+
+                    # This function is pretty handy!
+                    # I'll use it for an id from a string like: /projects/1070616180/
+                    project_id = commons.webscrape_count(project_anchors[0].attrs["href"],
+                                                         "/projects/", "/")
+                    title = project_anchors[1].contents[0]
+                    author = project_anchors[2].contents[0]
+
+                    # Instantiating a project with the properties that we know
+                    # This may cause issues (see below)
+                    _project = project.Project(id=project_id,
+                                               _session=self._session,
+                                               title=title,
+                                               author_name=author,
+                                               url=f"https://scratch.mit.edu/projects/{project_id}/")
+                    if get_full_project:
+                        # Put this under an if statement since making api requests for every single
+                        # project will cause the function to take a lot longer
+                        _project.update()
+
+                    _projects.append(
+                        _project
+                    )
+
+        return _projects
+
+    def loves_count(self):
+        text = requests.get(
+            f"https://scratch.mit.edu/projects/all/{self.username}/loves/",
+            headers=self._headers
+        ).text
+
+        return commons.webscrape_count(text, "&raquo;\n\n (", ")")
+
     def favorites(self, *, limit=40, offset=0):
         """
         Returns:
@@ -443,7 +535,7 @@ def reply_comment(self, content, *, parent_id, commentee_id=""):
 
         Warning:
             Only replies to top-level comments are shown on the Scratch website. Replies to replies are actually replies to the corresponding top-level comment in the API.
-            
+
             Therefore, parent_id should be the comment id of a top level comment.
 
         Args:
@@ -607,7 +699,7 @@ def comment_by_id(self, comment_id):
 
         Warning:
             For comments very far down on the user's profile, this method will take a while to find the comment. Very old comment are deleted from Scratch's database and may not appear.
-        
+
         Returns:
             scratchattach.comments.Comment: The request comment.
         """
@@ -675,11 +767,11 @@ def ocular_status(self):
             dict
         """
         return requests.get(f"https://my-ocular.jeffalo.net/api/user/{self.username}").json()
-    
+
     def verify_identity(self, *, verification_project_id=395330233):
         """
         Can be used in applications to verify a user's identity.
-        
+
         This function returns a Verifactor object. Attributs of this object:
         :.projecturl: The link to the project where the user has to go to verify
         :.project: The project where the user has to go to verify as scratchattach.Project object