From 0284d15d4286d07ef06cf0fd1bdacd204f3f68a8 Mon Sep 17 00:00:00 2001
From: cl117 <cl117@rice.edu>
Date: Tue, 30 Jun 2026 16:18:50 -0600
Subject: [PATCH] skip clustering

---
 flask/cluster.py  | 11 +++++++++++
 flask/config.json | 14 +-------------
 flask/explorer.py |  3 ++-
 3 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/flask/cluster.py b/flask/cluster.py
index 724e852..440d901 100644
--- a/flask/cluster.py
+++ b/flask/cluster.py
@@ -115,6 +115,17 @@ def uclust2clusters():
 
 def update_clusters():
     logger_.log('------------ Updating clusters ------------', True)
+
+    # Re-read config (force a fresh load, not the value cached at import) so a
+    # flag set at runtime via /config is honored. When skip_clustering is set,
+    # skip the (very slow) uclust run and keep the clusters from the previous
+    # run -- update_index will leave the existing clusters_dump untouched.
+    config_manager._config = None
+    if config_manager.load_config().get('skip_clustering'):
+        logger_.log('******** skip_clustering=true: skipping uclust, reusing previous clusters ********', True)
+        logger_.log('------------ Clustering skipped (previous clusters kept) ------------\n', True)
+        return None
+
     logger_.log('******** Query for sequences ********', True)
     sequences_response = query.query_sparql(sequence_query)
     logger_.log('******** Query for sequences complete ********', True)
diff --git a/flask/config.json b/flask/config.json
index a188e64..486899c 100644
--- a/flask/config.json
+++ b/flask/config.json
@@ -1,13 +1 @@
-{
-   "uclust_identity": "0.8",
-   "elasticsearch_index_name": "part",
-   "pagerank_tolerance": "0.0001",
-   "elasticsearch_endpoint": "http://localhost:9200/",
-   "sparql_endpoint": "http://localhost:8890/sparql?",
-   "last_update_start": "none",
-   "last_update_end": "none",
-   "distributed_search": false,
-   "which_search": "vsearch",
-   "autoUpdateIndex": false,
-   "updateTimeInDays": "1"
-}
+{"uclust_identity": "0.8", "elasticsearch_index_name": "part", "pagerank_tolerance": "0.0001", "elasticsearch_endpoint": "http://localhost:9200/", "sparql_endpoint": "http://localhost:8890/sparql?", "last_update_start": "2026-06-30T15:42:11.762203", "last_update_end": "none", "distributed_search": false, "which_search": "vsearch", "autoUpdateIndex": false, "updateTimeInDays": "1", "skip_clustering": true}
\ No newline at end of file
diff --git a/flask/explorer.py b/flask/explorer.py
index 56b0377..2166f9c 100644
--- a/flask/explorer.py
+++ b/flask/explorer.py
@@ -76,7 +76,8 @@ def update_index():
     config_manager.save_update_start_time()
 
     clusters = cluster.update_clusters()
-    data_manager.save_clusters(clusters)
+    if clusters is not None:  # None => skip_clustering, keep existing clusters_dump
+        data_manager.save_clusters(clusters)
     
     uri2rank = pagerank.update_pagerank()
     data_manager.save_uri2rank(uri2rank)