kuzu
kuzu copied to clipboard
kuzudb
Embeddable property graph database management system built for query speed and scalability. Implements Cypher.
Hey guys, please add a function that simplifies the construction of graphs from pure text. A possible solution would be to use the [LangChain GraphDocuments](https://api.python.langchain.com/en/latest/graphs/langchain_community.graphs.kuzu_graph.KuzuGraph.html#langchain_community.graphs.kuzu_graph.KuzuGraph) Object and provide a way...
Currently, `CommentOn` is a special operator on its own. Instead, it should be reworked as part of `Alter` operator under DDL.
Given an example: ``` MATCH (n0:Note {id: 11})-[r:Links]->(n:Note {id:12}) RETURN ID(n0), r, ID(n); ------------------------------------------------------------------------------- | n0._ID | r | n._ID | ------------------------------------------------------------------------------- | 1:435121 | (1:435121)-{_LABEL: Links, _ID: 2:841305}->(1:435122) |...
I have twitter related data in my database which includes `Tweets`, `Users`, `Hashtags` etc , and I have relations that include `Posts` (User Posting a tweet), `Mentions` (Tweet mentioning a...
Given the following queries: ``` MATCH (n0:Note {id: 11})-[r:Links]->(n:Note {id:12}) RETURN ID(n0), r, ID(n); ------------------------------------------------------------------------------- | n0._ID | r | n._ID | ------------------------------------------------------------------------------- | 1:435121 | (1:435121)-{_LABEL: Links, _ID: 2:841305}->(1:435122)...
We should support copy query statements. After we support it, we should use `std::unique_ptr` of statements rather than `std::shared_ptr`.
see #3062 TODO: - do not do much on rollback; for example, rollback hash index in CopyTable. - fail with inserting nodes after creating an empty table. Related to local...
We currently force the size of the selection vector to be exactly DEFAULT_VECTOR_CAPACITY, however the list data vector size can be greater than DEFAULT_VECTOR_CAPACITY. We should consider removing this restriction.
To support bypassing the WAL when writing new pages to disk arrays (an important performance optimization for #2938), it will be necessary to track the end of the file within...
Creating new nodes with a UNION type is not possible. ``` CREATE NODE TABLE tab(id SERIAL, un UNION(a INT32), PRIMARY KEY(id)); CREATE (t:tab {un: union_value(a := 1)}); ``` This gives...
