index-build

Author: monneyboi

poliloom/README.md

@@ -74,7 +74,7 @@ API documentation available at http://localhost:8000/docs
 | `poliloom import-politicians` | Import politicians linking to existing entities       |
 | `poliloom enrich-wikipedia`   | Extract politician data from Wikipedia using AI       |
 | `poliloom garbage-collect`    | Remove entities deleted from Wikidata                 |
-| `poliloom index-build`        | Build Meilisearch index (use `--rebuild` for first run) |
+| `poliloom index-build`        | Build Meilisearch index (`--rebuild` to start fresh)    |
 | `poliloom index-delete`       | Delete the Meilisearch index                          |
 | `poliloom index-stats`        | Show index status and task progress                   |
 

poliloom/poliloom/cli.py

@@ -978,15 +978,15 @@ def index_delete(confirm):
 @click.option(
     "--rebuild",
     is_flag=True,
-    help="Delete and recreate index before indexing (required for first run)",
+    help="Delete and recreate index before indexing",
 )
 def index_build(batch_size, rebuild):
     """Build Meilisearch index from database.
 
     Indexes all documents from Location, Country, Language, and Politician tables.
     Uses upsert semantics - unchanged documents won't be re-embedded.
 
-    Use --rebuild to delete and recreate the index (required for first run).
+    Use --rebuild to delete and recreate the index from scratch.
     """
     from poliloom.search import INDEX_NAME, SearchDocument, SearchService
 
@@ -999,6 +999,8 @@ def index_build(batch_size, rebuild):
         click.echo(f"   Recreated index '{INDEX_NAME}'")
     else:
         click.echo("⏳ Building Meilisearch index...")
+        if search_service.ensure_index():
+            click.echo(f"   Created index '{INDEX_NAME}'")
 
     # Reindex all documents from each model type
     models = _get_search_indexed_models()

poliloom/poliloom/search.py

@@ -103,6 +103,22 @@ def delete_index(self) -> None:
                 raise
             logger.debug(f"Index '{INDEX_NAME}' does not exist, nothing to delete")
 
+    def ensure_index(self) -> bool:
+        """Create the index if it doesn't exist.
+
+        Returns:
+            True if index was created, False if it already existed.
+        """
+        try:
+            self.client.get_index(INDEX_NAME)
+            logger.debug(f"Index '{INDEX_NAME}' already exists")
+            return False
+        except meilisearch.errors.MeilisearchApiError as e:
+            if "index_not_found" not in str(e):
+                raise
+            self.create_index()
+            return True
+
     def index_documents(self, documents: list[SearchDocument]) -> Optional[int]:
         """Index documents to Meilisearch.