From b4c5d50a77f22c1d273b86ce6d3f77ef30906ded Mon Sep 17 00:00:00 2001
From: rra <rscmbbng@riseup.net>
Date: Tue, 28 Apr 2020 14:30:18 +0200
Subject: [PATCH] updated with a way to get around gab.best enumeration and
 better error logging

---
 fedicrawler.py | 55 ++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 40 insertions(+), 15 deletions(-)

diff --git a/fedicrawler.py b/fedicrawler.py
index 1838d36..710cbc8 100644
--- a/fedicrawler.py
+++ b/fedicrawler.py
@@ -1,5 +1,5 @@
 #!/bin/env python3
-# fediscraper v2
+# fediscraper v3
 
 import json, requests
 from multiprocessing.dummy import Pool as ThreadPool 
@@ -8,7 +8,19 @@ start_url = 'https://post.lurk.org'
 activity = ''
 peers_info ='/api/v1/instance/peers'
 
-pool = ThreadPool(256)
+pool = ThreadPool(512)
+
+def not_gab(instance):
+    #gab does some weird stuff wrt enumerating subdomains
+    #example: epa1pu1qcxxyzcxher0u.gab.best
+    if instance:
+        if 'gab.best'in instance:
+            return False
+        else:
+            return True
+    else:
+        return False
+    #TODO filter ngrok
 
 
 def get_peers(instance):
@@ -18,7 +30,6 @@ def get_peers(instance):
             peers = r.json()
             print(instance, 'peers with', len(peers))
             return peers
-        
         else:
             print('fail: got {} on {}'.format(r.status_code, instance))
             return
@@ -29,16 +40,29 @@ def get_peers(instance):
 
 def get_instance_info(instance):
     instance_info = '/api/v1/instance'
-    print('getting info for', instance)
-    try:
-        ii = requests.get('https://'+instance+instance_info, timeout=10)
-        info = ii.json()
-        print('info request for {} succeeded'.format(instance))
-    except Exception as e:
-        print('failed to query instance info')
-        # print(e)
-        info = {'error': 'error'}
-    return info
+    if not_gab(instance):
+        print('getting info for', instance)
+        try:
+            r = requests.get('https://'+instance+instance_info, timeout=10)
+            if r.status_code == 200:
+                info = r.json()
+                print('info request for {} succeeded'.format(instance))
+            elif r.status_code == 400:
+                #try to see if its peertube, probably should use something better
+                pt = requests.get('https://'+instance+'/api/v1/config')
+                if r.status_code == 200:
+                    print('info request for {} succeeded, peertube'.format(instance))
+                    info = r.json()
+                else:
+                    info = {'error': r.status_code}
+            else:
+                info = {'error': r.status_code}
+
+        except Exception as e:
+            print('failed to query instance info')
+            # print(e)
+            info = {'error': str(e)}
+        return info
 
 
 instances = set([])
@@ -55,7 +79,8 @@ if r. status_code == 200:
     for peer_list in network:
         if peer_list:
             for instance in peer_list:
-                instances.add(instance)
+                if not_gab(instance):
+                    instances.add(instance)
 
 
 instance_info = pool.map(get_instance_info, instances)
@@ -76,4 +101,4 @@ pool.join()
 
 with open('instance_scrape.json','w') as f:
     f.write(json.dumps(scrape,indent=4))
-    #f.write('\n'.join(text)
\ No newline at end of file
+    #f.write('\n'.join(text)