From a4749501f4bf4a48aa79cd30d42abb0e45453430 Mon Sep 17 00:00:00 2001
From: Mars Ultor <krishna@ayyalasomayajula.net>
Date: Wed, 24 Dec 2025 20:00:30 -0600
Subject: [PATCH] UB fixed in parse_chat

---
 src/processing_utils.cpp | 30 +++++++++++++++++++++---------
 1 file changed, 21 insertions(+), 9 deletions(-)

diff --git a/src/processing_utils.cpp b/src/processing_utils.cpp
index fd791d7..3022efb 100644
--- a/src/processing_utils.cpp
+++ b/src/processing_utils.cpp
@@ -128,22 +128,34 @@ time_t extract_timestamp_utc(std::string_view filename,
     return Route::NONE;
   }
 
-  std::pair<std::string, std::string>
-parse_chat(std::string_view line) {
-    // [CHAT] <Steve> hello world
+std::pair<std::string, std::string>
+parse_chat(std::string_view line)
+{
+    constexpr std::string_view tag = "[CHAT]";
 
-    auto start = line.find("[CHAT] <");
-    start += 8;
+    auto tag_pos = line.find(tag);
+    if (tag_pos == std::string_view::npos)
+        return {"", ""};
 
-    auto end = line.find('>', start);
-    auto msg = end + 2;
+    std::size_t pos = tag_pos + tag.size();
+    if (pos >= line.size())
+        return {"", ""};
 
+    // Skip whitespace after [CHAT]
+    while (pos < line.size() && line[pos] == ' ')
+        ++pos;
+
+    if (pos >= line.size())
+        return {"", ""};
+
+    // Entire remainder is "message"
     return {
-        std::string(line.substr(start, end - start)),
-        std::string(line.substr(msg))
+        "",  // sender unknown / system / to be inferred elsewhere
+        std::string(line.substr(pos))
     };
 }
 
+
   void process_to_eof(std::string path, std::shared_ptr<clickhouse::Client>client, std::string target_user_id){
     spdlog::info("Processing file {} ", path);
     gzFile gz_file_log = gzopen(path.c_str(), "rb");