Add experiment to see ye's follower's creation dates

Parler/__init__.py

@@ -87,7 +87,7 @@ def handle_response(self, response):
                 {
                     "status": response.status_code,
                     "error": response.reason,
-                    "message": self.Errors.NoAuth,
+                    "message": response.text,
                 }
             )
 

README.md

@@ -15,6 +15,7 @@ This **UNOFFICIAL** library designed to programatically fetch data from parler.c
 - [X] Implement unfinished authed functions
 - [ ] Rework experiments
 - [ ] Have a github action (several!) to archive posts, run analysises, etc..
+- [ ] Implement controls for header `x-ratelimit-limit` and `x-ratelimit-limit-remaining`
 - [X] Testing, testing
 - [X] Publish to pypi
 

experiments/08_ye.py

@@ -0,0 +1,143 @@
+# Script used to generate graphs and data for the article "Can Ye Save Parler?"
+
+from os import path
+import Parler
+from Parler import with_auth as authed
+import os
+from Parler import utils
+import time
+import datetime
+import json
+import itertools
+
+import matplotlib.pyplot as plt
+import matplotlib.dates as mdates
+import pandas as pd
+
+p = Parler.Parler(debug=True)
+au = authed.AuthSession(debug=True)
+assert os.getenv("PARLER_USERNAME") is not None
+assert os.getenv("PARLER_PASSWORD") is not None
+
+FLAG_WRITE_FILES = True
+
+au.login(
+    identifier=os.getenv("PARLER_USERNAME"), password=os.getenv("PARLER_PASSWORD")
+)
+assert au.is_logged_in
+
+# Hypothesis 1:
+# Ye"s followers" creation / joined dates are largely very recent, as in, many people created accounts on parler and followed him after the acquisition announcement.
+# Because Parler API does not provide an endpoint to get followers, we gather the comments from his post and check if the commenter follows Ye.
+# Unfortunately Ye"s entire list of ~37k followers are out of reach.
+# Results:
+
+
+def Hypothesis1():
+
+    comments = []
+
+    ye_followers_creation_dates = []
+
+    comments_limit = 20
+    # Get Ye's posts, which are a few, for now
+    r = au.created_items("ye", 20, 1)
+    assert utils.is_ok(r)
+    for post in r.get("data"):
+        cursor = 0
+        d = datetime.datetime.now()
+        timestamp = d.strftime("%Y-%m-%d %H:%M:%S")
+
+        comments_response = au.comments(
+            post.get("postuuid"), cursor, comments_limit, timestamp)
+        utils.is_ok(comments_response)
+        print(len(comments_response.get("data")))
+        comments.append(comments_response.get("data"))
+        while utils.is_ok(comments_response) and len(comments_response.get("data")) >= 1:
+            cursor += comments_limit
+            comments_response = au.comments(
+                post.get("postuuid"), cursor, comments_limit, timestamp)
+            comments.append(comments_response.get("data"))
+            time.sleep(0.5)
+
+    if FLAG_WRITE_FILES:
+        with open(path.join("output",  "comments.json"), "w") as f:
+            json.dump(comments, f, ensure_ascii=False, indent=4)
+
+    for comment in list(itertools.chain(*comments)):
+        cursor = 1
+
+        try:
+            r = au.following(comment.get("user").get("username"), 1, 20)
+        except:
+            continue
+        has_found_ye = False
+        is_end = False
+        while (not has_found_ye) and (not is_end):
+
+            try:
+                r = au.following(comment.get(
+                    "user").get("username"), cursor, 20)
+            except:
+                time.sleep(10)
+                r = au.following(comment.get(
+                    "user").get("username"), cursor, 20)
+            if not utils.is_ok(r):
+                is_end = True
+            if len(r.get("data")) == 0:
+                is_end = True
+            if utils.is_ok(r) and len(r.get("data")) >= 1:
+                following_users = r.get("data")
+                for x in following_users:
+                    if x.get("username") == "Ye":
+                        has_found_ye = True
+            cursor += 20
+        if has_found_ye:
+            try:
+                profile_response = p.profile(
+                    comment.get("user").get("username"))
+                if utils.is_ok(profile_response):
+                    ye_followers_creation_dates.append(
+                        profile_response.get("data").get("joined_date"))
+            except:
+                pass
+
+    if FLAG_WRITE_FILES:
+        with open(path.join("output",  "ye_followers_creation_dates.json"), "w") as f:
+            json.dump(ye_followers_creation_dates, f,
+                      ensure_ascii=False, indent=4)
+
+    # Make a chart
+
+    # aggregate by date
+    dates_agg = {}
+
+    for x in sorted(ye_followers_creation_dates):
+
+        parsed = x.split("T")[0]
+        if parsed not in dates_agg:
+            dates_agg[parsed] = 0
+        dates_agg[parsed] += 1
+
+    dataframe_dates = pd.to_datetime(list(dates_agg.keys()))
+    dataframe_amounts = list(dates_agg.values())
+
+    fig = plt.figure(figsize=(10, 7))
+    fig.tight_layout()
+    plt.xticks(rotation=90)
+    plt.plot(dataframe_dates, dataframe_amounts)
+    plt.title("Ye's Parler Account's Followers' creation dates")
+
+    plt.ylabel("Amount")
+    ax = plt.gca()
+    days_fmt = mdates.DateFormatter("%D")
+    ax.xaxis.set_major_locator(mdates.DayLocator(interval=2))
+    ax.xaxis.set_major_formatter(days_fmt)
+    ax.set_xlim([datetime.date(2022, 9, 1), datetime.date(2022, 10, 25)])
+    ax.axvline(datetime.date(2022, 10, 17), ls="-", color="red",
+               label="Parler Purchase Announcement (17/10)")
+    plt.gca().legend()
+    plt.savefig("chart.png")
+
+
+Hypothesis1()