Continued:

[fba.git] / fetch_instances.py

diff --git a/fetch_instances.py b/fetch_instances.py
old mode 100644 (file)
new mode 100755 (executable)
index 346ece0..63a8fa7
--- a/fetch_instances.py
+++ b/fetch_instances.py
@@ -1,43 +1,89 @@
+#!/usr/bin/python3
+# -*- coding: utf-8 -*-
+
+# Fedi API Block - An aggregator for fetching blocking data from fediverse nodes
+# Copyright (C) 2023 Free Software Foundation
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published
+# by the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program.  If not, see <https://www.gnu.org/licenses/>.
+
 import sqlite3
 import sys
 import json
+import time
+import validators
 import fba
 
-domain = sys.argv[1]
+def fetch_instances(domain: str, origin: str, software: str, path: str = None):
+    # NOISY-DEBUG: print("DEBUG: domain,origin,software,path:", domain, origin, software, path)
+    if not fba.is_instance_registered(domain):
+        # NOISY-DEBUG: print("DEBUG: Adding new domain:", domain, origin)
+        fba.add_instance(domain, origin, sys.argv[0], path)
 
-peerlist = fba.get_peers(domain)
+    # NOISY-DEBUG: print("DEBUG: Fetching instances for domain:", domain, software)
+    peerlist = fba.get_peers(domain, software)
 
-if (peerlist is None):
-    print("FATAL: CANNOT FETCH PEERS:", domain)
-    sys.exit(255)
+    if (peerlist is None):
+        print("ERROR: Cannot fetch peers:", domain)
+        return
+    elif fba.has_pending_nodeinfos(domain):
+        # NOISY-DEBUG: print(f"DEBUG: domain='{domain}' has pending nodeinfo data, flushing ...")
+        fba.update_nodeinfos(domain)
 
-fba.c.execute(
-    "SELECT domain FROM instances WHERE 1"
-)
+    print(f"INFO: Checking {len(peerlist)} instances from {domain} ...")
+    for instance in peerlist:
+        # NOISY-DEBUG: print("DEBUG: BEFORE instance:", instance)
+        instance = fba.tidyup(instance)
+        # NOISY-DEBUG: print("DEBUG: AFTER instance:", instance)
 
-for instance in peerlist:
-    instance = instance.lower()
+        if instance == "":
+            print("WARNING: Empty instance after tidyup(), domain:", domain)
+            continue
+        elif not validators.domain(instance.split("/")[0]):
+            print(f"WARNING: Bad instance='{instance}' from domain='{domain}',origin='{origin}',software='{software}'")
+            continue
+        elif fba.is_blacklisted(instance):
+            # NOISY-DEBUG: print("DEBUG: instance is blacklisted:", instance)
+            continue
 
-    blacklisted = False
-    for domain in fba.blacklist:
-        if domain in instance:
-            blacklisted = True
+        # NOISY-DEBUG: print("DEBUG: Handling instance:", instance)
+        try:
+            if not fba.is_instance_registered(instance):
+                # NOISY-DEBUG: print("DEBUG: Adding new instance:", instance, domain)
+                fba.add_instance(instance, domain, sys.argv[0])
+        except BaseException as e:
+            print(f"ERROR: instance='{instance}',exception:'{str(e)}'")
+            continue
 
-    if blacklisted:
-        print("domain is blacklisted:", domain)
-        continue
+instance = sys.argv[1]
 
-    print("Handling instance:", instance)
-    try:
-        fba.c.execute(
-            "SELECT domain FROM instances WHERE domain = ?", (instance,)
-        )
+# Initial fetch
+fetch_instances(instance, None, None)
 
-        if fba.c.fetchone() == None:
-            fba.add_instance(instance)
+# Loop through some instances
+fba.cursor.execute(
+    "SELECT domain, origin, software, nodeinfo_url FROM instances WHERE software IN ('pleroma', 'mastodon', 'friendica', 'misskey', 'gotosocial', 'bookwyrm', 'takahe', 'lemmy') AND (last_instance_fetch IS NULL OR last_instance_fetch < ?) ORDER BY rowid DESC", [time.time() - fba.config["recheck_instance"]]
+)
+
+rows = fba.cursor.fetchall()
+print(f"INFO: Checking {len(rows)} entries ...")
+for row in rows:
+    # NOISY-DEBUG: print("DEBUG: domain:", row[0])
+    if fba.is_blacklisted(row[0]):
+        print("WARNING: domain is blacklisted:", row[0])
+        continue
 
-        fba.conn.commit()
-    except Exception as e:
-        print("error:", e, instance)
+    print(f"INFO: Fetching instances for instance '{row[0]}'('{row[2]}') of origin '{row[1]}',nodeinfo_url='{row[3]}'")
+    fetch_instances(row[0], row[1], row[2], row[3])
 
-fba.conn.close()
+fba.connection.close()