From 90b8bb6193d36fddfcad2208135bf21f696ec276 Mon Sep 17 00:00:00 2001
From: wuhongsheng <664116298@qq.com>
Date: Fri, 8 Nov 2024 13:58:12 +0800
Subject: [PATCH 1/2] fix:JSON serialization problem

---
 nano_graphrag/_op.py                   |  4 ++--
 nano_graphrag/_storage/gdb_networkx.py |  6 +++---
 nano_graphrag/_utils.py                | 10 +++++-----
 3 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/nano_graphrag/_op.py b/nano_graphrag/_op.py
index 4ee9ed5..9714a63 100644
--- a/nano_graphrag/_op.py
+++ b/nano_graphrag/_op.py
@@ -1,5 +1,5 @@
 import re
-import json
+import json5
 import asyncio
 import tiktoken
 from typing import Union
@@ -680,7 +680,7 @@ async def _find_most_related_community_from_entities(
     for node_d in node_datas:
         if "clusters" not in node_d:
             continue
-        related_communities.extend(json.loads(node_d["clusters"]))
+        related_communities.extend(json5.loads(node_d["clusters"]))
     related_community_dup_keys = [
         str(dp["cluster"])
         for dp in related_communities
diff --git a/nano_graphrag/_storage/gdb_networkx.py b/nano_graphrag/_storage/gdb_networkx.py
index e29bf3e..38351d3 100644
--- a/nano_graphrag/_storage/gdb_networkx.py
+++ b/nano_graphrag/_storage/gdb_networkx.py
@@ -1,5 +1,5 @@
 import html
-import json
+import json5
 import os
 from collections import defaultdict
 from dataclasses import dataclass
@@ -154,7 +154,7 @@ async def community_schema(self) -> dict[str, SingleCommunitySchema]:
         for node_id, node_data in self._graph.nodes(data=True):
             if "clusters" not in node_data:
                 continue
-            clusters = json.loads(node_data["clusters"])
+            clusters = json5.loads(node_data["clusters"])
             this_node_edges = self._graph.edges(node_id)
 
             for cluster in clusters:
@@ -195,7 +195,7 @@ async def community_schema(self) -> dict[str, SingleCommunitySchema]:
 
     def _cluster_data_to_subgraphs(self, cluster_data: dict[str, list[dict[str, str]]]):
         for node_id, clusters in cluster_data.items():
-            self._graph.nodes[node_id]["clusters"] = json.dumps(clusters)
+            self._graph.nodes[node_id]["clusters"] = json5.dumps(clusters)
 
     async def _leiden_clustering(self):
         from graspologic.partition import hierarchical_leiden
diff --git a/nano_graphrag/_utils.py b/nano_graphrag/_utils.py
index ae772eb..17111ab 100644
--- a/nano_graphrag/_utils.py
+++ b/nano_graphrag/_utils.py
@@ -1,6 +1,6 @@
 import asyncio
 import html
-import json
+import json5
 import logging
 import os
 import re
@@ -45,8 +45,8 @@ def extract_first_complete_json(s: str):
                     first_json_str = s[first_json_start:i+1]
                     try:
                         # Attempt to parse the JSON string
-                        return json.loads(first_json_str.replace("\n", ""))
-                    except json.JSONDecodeError as e:
+                        return json5.loads(first_json_str.replace("\n", ""))
+                    except json5.JSONDecodeError as e:
                         logger.error(f"JSON decoding failed: {e}. Attempted string: {first_json_str[:50]}...")
                         return None
                     finally:
@@ -151,14 +151,14 @@ def compute_mdhash_id(content, prefix: str = ""):
 
 def write_json(json_obj, file_name):
     with open(file_name, "w", encoding="utf-8") as f:
-        json.dump(json_obj, f, indent=2, ensure_ascii=False)
+        json5.dump(json_obj, f, indent=2, ensure_ascii=False)
 
 
 def load_json(file_name):
     if not os.path.exists(file_name):
         return None
     with open(file_name, encoding="utf-8") as f:
-        return json.load(f)
+        return json5.load(f)
 
 
 # it's dirty to type, so it's a good way to have fun

From 92bc4475f69a169c2a5bb18d566584b9d0d225f6 Mon Sep 17 00:00:00 2001
From: wuhongsheng <664116298@qq.com>
Date: Mon, 11 Nov 2024 16:55:50 +0800
Subject: [PATCH 2/2] fix:JSON serialization problem

---
 nano_graphrag/_utils.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/nano_graphrag/_utils.py b/nano_graphrag/_utils.py
index 17111ab..e44222a 100644
--- a/nano_graphrag/_utils.py
+++ b/nano_graphrag/_utils.py
@@ -1,6 +1,7 @@
 import asyncio
 import html
 import json5
+import json
 import logging
 import os
 import re
@@ -45,8 +46,8 @@ def extract_first_complete_json(s: str):
                     first_json_str = s[first_json_start:i+1]
                     try:
                         # Attempt to parse the JSON string
-                        return json5.loads(first_json_str.replace("\n", ""))
-                    except json5.JSONDecodeError as e:
+                        return json.loads(first_json_str.replace("\n", ""))
+                    except json.JSONDecodeError as e:
                         logger.error(f"JSON decoding failed: {e}. Attempted string: {first_json_str[:50]}...")
                         return None
                     finally: