Skip to content

chunking & polardb metadata bug fix #1485

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
CaralHsi merged 3 commits into from
Apr 17, 2026
Merged

chunking & polardb metadata bug fix #1485

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
037cc95
fix: polardb metadata bug fix
bittergreen Apr 13, 2026
9483b3c
fix: fix chunking bug in memreader
bittergreen Apr 13, 2026
0337860
fix: also fix get_node for nested metadata
bittergreen Apr 17, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 6 additions & 20 deletions src/memos/graph_dbs/polardb.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -899,13 +899,7 @@ def get_node(
logger.info(
f" polardb [get_node] get_node completed time in {elapsed_time:.2f}s"
)
return self._parse_node(
{
"id": id,
"memory": properties.get("memory", ""),
**properties,
}
)
return self._parse_node(properties)
return None

except Exception as e:
Expand Down Expand Up @@ -976,15 +970,7 @@ def get_nodes(self, ids: list[str], user_name: str, **kwargs) -> list[dict[str,
properties["embedding"] = embedding
except (json.JSONDecodeError, TypeError):
logger.warning(f"Failed to parse embedding for node {node_id}")
nodes.append(
self._parse_node(
{
"id": properties.get("id", node_id),
"memory": properties.get("memory", ""),
"metadata": properties,
}
)
)
nodes.append(self._parse_node(properties))
return nodes

@timed
Expand Down Expand Up @@ -1534,7 +1520,7 @@ def search_by_keywords_like(
user_name_conditions = self._build_user_name_and_kb_ids_conditions_sql(
user_name=user_name,
knowledgebase_ids=knowledgebase_ids,
default_user_name=self.config.user_name,
default_user_name=self._get_config_value("user_name"),
)

# Add OR condition if we have any user_name conditions
Expand Down Expand Up @@ -1633,7 +1619,7 @@ def search_by_keywords_tfidf(
user_name_conditions = self._build_user_name_and_kb_ids_conditions_sql(
user_name=user_name,
knowledgebase_ids=knowledgebase_ids,
default_user_name=self.config.user_name,
default_user_name=self._get_config_value("user_name"),
)

# Add OR condition if we have any user_name conditions
Expand Down Expand Up @@ -1751,7 +1737,7 @@ def search_by_fulltext(
user_name_conditions = self._build_user_name_and_kb_ids_conditions_sql(
user_name=user_name,
knowledgebase_ids=knowledgebase_ids,
default_user_name=self.config.user_name,
default_user_name=self._get_config_value("user_name"),
)

if user_name_conditions:
Expand Down Expand Up @@ -1873,7 +1859,7 @@ def search_by_embedding(
user_name_conditions = self._build_user_name_and_kb_ids_conditions_sql(
user_name=user_name,
knowledgebase_ids=knowledgebase_ids,
default_user_name=self.config.user_name,
default_user_name=self._get_config_value("user_name"),
)

if user_name_conditions:
Expand Down
5 changes: 3 additions & 2 deletions src/memos/mem_reader/multi_modal_struct.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,8 +102,9 @@ def _split_large_memory_item(
split_items = []

def _create_chunk_item(chunk):
# Chunk objects have a 'text' attribute
chunk_text = chunk.text
# Different chunkers are not fully consistent:
# some return Chunk-like objects with `.text`, while others return raw strings.
chunk_text = chunk.text if hasattr(chunk, "text") else chunk
if not chunk_text or not chunk_text.strip():
return None
# Create a new memory item for each chunk, preserving original metadata
Expand Down
Loading