From efc4de0bf0e6b5197eb2f1e008d4b7d521ab2e29 Mon Sep 17 00:00:00 2001
From: Cameron Pfiffer <cameron@pfiffer.org>
Date: Thu, 12 Mar 2026 01:53:44 -0700
Subject: [PATCH] fix(cli): parse facets in ingest-thread to prevent URL
 truncation

---
 src/void_tools/cli.py | 58 +++++++++++++++++++++++++++++++++++--------
 1 file changed, 48 insertions(+), 10 deletions(-)

diff --git a/src/void_tools/cli.py b/src/void_tools/cli.py
index a4c9d40..b67d6bf 100644
--- a/src/void_tools/cli.py
+++ b/src/void_tools/cli.py
@@ -449,9 +449,7 @@ def cmd_dispatch(args):
             if not text:
                 print("Error: annotation requires 'text' field. Skipping.")
                 continue
-                
-            target_uri = mapping[target_id]
-            
+
             # Convert AT-URI to bsky.app HTTP URL for Margin compatibility
             if target_uri.startswith("at://"):
                 parts = target_uri.replace("at://", "").split("/")
@@ -732,13 +730,44 @@ def cmd_ingest_thread(args):
         timestamp = datetime.datetime.now(datetime.UTC).strftime('%Y%m%d_%H%M%S')
         filename = os.path.join(threads_dir, f"thread_{post_id}_{timestamp}.md")
         
+        def replace_facets_with_markdown(text, facets):
+            if not facets:
+                return text
+            try:
+                text_bytes = text.encode('utf-8')
+                links = []
+                for facet in facets:
+                    for feature in getattr(facet, 'features', []):
+                        if hasattr(feature, 'uri'):
+                            links.append({
+                                'start': facet.index.byte_start,
+                                'end': facet.index.byte_end,
+                                'uri': feature.uri
+                            })
+                links.sort(key=lambda x: x['start'], reverse=True)
+                for link in links:
+                    start, end, uri = link['start'], link['end'], link['uri']
+                    link_text = text_bytes[start:end].decode('utf-8')
+                    markdown_link = f"[{link_text}]({uri})".encode('utf-8')
+                    text_bytes = text_bytes[:start] + markdown_link + text_bytes[end:]
+                return text_bytes.decode('utf-8')
+            except Exception as e:
+                # Fallback to plain text if parsing fails
+                return text
+
         def format_thread_node(node, level=0):
             content = ""
             indent = "  " * level
             if hasattr(node, 'post'):
                 post = node.post
                 author = post.author.handle
-                text = post.record.text.replace('\n', '\n' + indent + '> ')
+                
+                # Apply facets to get full URLs
+                raw_text = getattr(post.record, 'text', '')
+                facets = getattr(post.record, 'facets', [])
+                text = replace_facets_with_markdown(raw_text, facets)
+                
+                text = text.replace('\n', '\n' + indent + '> ')
                 content += f"{indent}- **@{author}**: {text}\n"
                 
                 if hasattr(node, 'replies') and node.replies:
@@ -779,11 +808,8 @@ def cmd_fetch_history(args):
         cursor = None
         
         import datetime
-        from dateutil import parser as date_parser
-        
-        target_date = date_parser.isoparse(args.before)
-        if target_date.tzinfo is None:
-            target_date = target_date.replace(tzinfo=datetime.UTC)
+        # Simple ISO parser fallback since dateutil might not be installed in the venv
+        target_date = datetime.datetime.fromisoformat(args.before.replace('Z', '+00:00'))
 
         print(f"Fetching posts for {args.handle} before {target_date}...")
         
@@ -800,7 +826,7 @@ def cmd_fetch_history(args):
             for item in response.feed:
                 # We only want posts by the author, not reposts
                 if hasattr(item.post.record, 'text') and item.post.author.did == did:
-                    post_time = date_parser.isoparse(item.post.indexed_at)
+                    post_time = datetime.datetime.fromisoformat(item.post.indexed_at.replace('Z', '+00:00'))
                     if post_time < target_date:
                         posts.append({
                             'uri': item.post.uri,
@@ -825,6 +851,13 @@ def cmd_fetch_history(args):
     except Exception as e:
         print(f"Error fetching history: {e}", file=sys.stderr)
 
+def cmd_monitor(args):
+    """Execute the Horizon Radar monitoring cycle."""
+    from void_tools.monitor import run_monitor_cycle
+    config_path = os.path.join(os.getcwd(), 'monitoring.yaml')
+    output_path = os.path.join(os.getcwd(), 'horizon.yaml')
+    run_monitor_cycle(config_path, output_path)
+
 def main():
     parser = argparse.ArgumentParser(description="Void ATProto CLI Tools")
     subparsers = parser.add_subparsers(dest="command", required=True)
@@ -880,6 +913,9 @@ def main():
     annotate_parser.add_argument("--motivation", default="commenting", help="W3C motivation (default: commenting)")
     annotate_parser.add_argument("--quote", help="Exact text to anchor the annotation to")
 
+    # Monitor
+    monitor_parser = subparsers.add_parser("monitor", help="Run the Horizon Radar to discover relevant network activity")
+
     args = parser.parse_args()
     
     if args.command == "notifs":
@@ -900,6 +936,8 @@ def main():
         cmd_profile(args)
     elif args.command == "annotate":
         cmd_annotate(args)
+    elif args.command == "monitor":
+        cmd_monitor(args)
 
 if __name__ == "__main__":
     main()
-- 
2.51.2