Fix text unit incremental ID updates (#1734)

natoverse · web-flow · commit faa05b691ffe · 2025-02-24T14:58:00.000-08:00
* Increment text_unit ids during incremental

* Semver
diff --git a/.semversioner/next-release/patch-20250224201440103182.json b/.semversioner/next-release/patch-20250224201440103182.json
@@ -0,0 +1,4 @@
+{
+  "type": "patch",
+  "description": "Properly increment text unit IDs during updates."
+}
diff --git a/graphrag/index/update/incremental_index.py b/graphrag/index/update/incremental_index.py
@@ -301,6 +301,10 @@ def _update_and_merge_text_units(
             lambda x: [entity_id_mapping.get(i, i) for i in x] if x is not None else x
         )
 
+    initial_id = old_text_units["human_readable_id"].max() + 1
+    delta_text_units["human_readable_id"] = np.arange(
+        initial_id, initial_id + len(delta_text_units)
+    )
     # Merge the final text units
     return pd.concat([old_text_units, delta_text_units], ignore_index=True, copy=False)
 

-Original file line number
+Diff line change
@@ @@ -0,0 +1,4 @@ @@
 +{
 +  "type": "patch",
 +  "description": "Properly increment text unit IDs during updates."
 +}
Original file line number	Diff line number	Diff line change
`@@ -301,6 +301,10 @@ def _update_and_merge_text_units(`
`301`	`301`	`lambda x: [entity_id_mapping.get(i, i) for i in x] if x is not None else x`
`302`	`302`	`)`
`303`	`303`
	`304`	`+ initial_id = old_text_units["human_readable_id"].max() + 1`
	`305`	`+ delta_text_units["human_readable_id"] = np.arange(`
	`306`	`+ initial_id, initial_id + len(delta_text_units)`
	`307`	`+ )`
`304`	`308`	`# Merge the final text units`
`305`	`309`	`return pd.concat([old_text_units, delta_text_units], ignore_index=True, copy=False)`
`306`	`310`