From 2c563ad28048fc4672bb631f0605f5080761c892 Mon Sep 17 00:00:00 2001
From: Debanjum Singh Solanky <debanjum@gmail.com>
Date: Sat, 27 Apr 2024 01:55:08 +0530
Subject: [PATCH] Use hash of query in process lock id to standardize id format

- Using inferred_query directly was brittle (like previous job id)
- And process lock id had a limited size, so wouldn't work for larger
  inferred query strings
---
 src/khoj/routers/api_chat.py | 6 ++++--
 src/khoj/routers/email.py    | 7 +++++--
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/src/khoj/routers/api_chat.py b/src/khoj/routers/api_chat.py
index 58cc72bc..4e2f38ab 100644
--- a/src/khoj/routers/api_chat.py
+++ b/src/khoj/routers/api_chat.py
@@ -407,6 +407,7 @@ async def websocket_endpoint(
                 continue
             # Generate the job id from the hash of inferred_query and crontime
             job_id = hashlib.md5(f"{inferred_query}_{crontime}".encode("utf-8")).hexdigest()
+            query_id = hashlib.md5(f"{inferred_query}".encode("utf-8")).hexdigest()
             partial_scheduled_chat = functools.partial(
                 scheduled_chat, inferred_query, q, websocket.user.object, websocket.url
             )
@@ -416,7 +417,7 @@ async def websocket_endpoint(
                     trigger=trigger,
                     args=(
                         partial_scheduled_chat,
-                        f"{ProcessLock.Operation.SCHEDULED_JOB}_{user.uuid}_{inferred_query}",
+                        f"{ProcessLock.Operation.SCHEDULED_JOB}_{user.uuid}_{query_id}",
                     ),
                     id=f"job_{user.uuid}_{job_id}",
                     name=f"{inferred_query}",
@@ -682,12 +683,13 @@ async def chat(
 
         # Generate the job id from the hash of inferred_query and crontime
         job_id = hashlib.md5(f"{inferred_query}_{crontime}".encode("utf-8")).hexdigest()
+        query_id = hashlib.md5(f"{inferred_query}".encode("utf-8")).hexdigest()
         partial_scheduled_chat = functools.partial(scheduled_chat, inferred_query, q, request.user.object, request.url)
         try:
             job = state.scheduler.add_job(
                 run_with_process_lock,
                 trigger=trigger,
-                args=(partial_scheduled_chat, f"{ProcessLock.Operation.SCHEDULED_JOB}_{user.uuid}_{inferred_query}"),
+                args=(partial_scheduled_chat, f"{ProcessLock.Operation.SCHEDULED_JOB}_{user.uuid}_{query_id}"),
                 id=f"job_{user.uuid}_{job_id}",
                 name=f"{inferred_query}",
                 max_instances=2,  # Allow second instance to kill any previous instance with stale lock
diff --git a/src/khoj/routers/email.py b/src/khoj/routers/email.py
index 96eb5d85..8e6464d3 100644
--- a/src/khoj/routers/email.py
+++ b/src/khoj/routers/email.py
@@ -60,11 +60,14 @@ def send_task_email(name, email, query, result):
     html_result = markdown_it.MarkdownIt().render(result)
     html_content = template.render(name=name, query=query, result=html_result)
 
-    resend.Emails.send(
+    query_for_subject_line = query.replace("\n", " ").replace('"', "").replace("'", "")
+
+    r = resend.Emails.send(
         {
             "from": "Khoj <khoj@khoj.dev>",
             "to": email,
-            "subject": f'✨ Your Task Results for "{query}"',
+            "subject": f'✨ Your Task Results for "{query_for_subject_line}"',
             "html": html_content,
         }
     )
+    return r