Merge pull request #155 from SynBioDex/153-use-minhash-for-similar-and-hash-for-exact-sequence-clustering

set skip_cluster to true by default

Author: cl117

flask/cluster.py

@@ -117,11 +117,13 @@ 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.
+    # flag set at runtime via /config is honored. skip_clustering DEFAULTS TO
+    # TRUE: the (very slow) uclust run is skipped and the previous clusters are
+    # reused unless config explicitly sets skip_clustering to false. This lets
+    # deployments (e.g. Azure) that can't edit config.json avoid the multi-hour
+    # clustering by default; a frontend can later set it false to force a rerun.
     config_manager._config = None
-    if config_manager.load_config().get('skip_clustering'):
+    if config_manager.load_config().get('skip_clustering', True):
         logger_.log('******** skip_clustering=true: skipping uclust, reusing previous clusters ********', True)
         logger_.log('------------ Clustering skipped (previous clusters kept) ------------\n', True)
         return None

flask/config.json

@@ -1 +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}
+{"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": "2026-06-30T18:45:50.382470", "last_update_end": "2026-06-30T16:47:36.892366", "distributed_search": false, "which_search": "vsearch", "autoUpdateIndex": false, "updateTimeInDays": "1"}