A Vertex Life in Nebula Graph Storaged¶
约 230 个字 73 行代码 1 张图片 预计阅读时间 2 分钟
在 Nebula Graph 中，一个 Vertex 的生命周期从创建到删除，涉及到多个组件和流程。本文将详细介绍一个 Vertex 的生命周期，由此来借鉴实现 VECTOR 类型属性的存储和处理流程。
Vertex Creation¶
在 Nebula Graph 中，Vertex 的创建通常通过 AddVerticesProcessor executor 来完成。它会将 Vertex 要插入的数据已经元数据打包成 raft-wal log，提交到 Raft Part 中，最后在 Part::commitLogs()中将数据写入到 RocksDB。
Vertex Search¶
Vertex 的查询可以通过 Match，GO 以及 Fetch 进行查询，这里我们先讨论最简单的 Fetch 查询。
Fetch 查询¶
FETCH PROP ON test1 "t1" YIELD properties(vertex);语句执行计划如下：
C++
-----+-------------+--------------+----------------+------------------------------
| id | name | dependencies | profiling data | operator info |
-----+-------------+--------------+----------------+------------------------------
| 2 | Project | 1 | | outputVar: { |
| | | | | "colNames": [ |
| | | | | "properties(VERTEX)" |
| | | | | ], |
| | | | | "name": "__Project_2", |
| | | | | "type": "DATASET" |
| | | | | } |
| | | | | inputVar: __GetVertices_1 |
| | | | | columns: [ |
| | | | | "properties(VERTEX)" |
| | | | | ] |
-----+-------------+--------------+----------------+------------------------------
| 1 | GetVertices | 0 | | outputVar: { |
| | | | | "colNames": [], |
| | | | | "type": "DATASET", |
| | | | | "name": "__GetVertices_1" |
| | | | | } |
| | | | | inputVar: __VAR_0 |
| | | | | space: 2 |
| | | | | dedup: 0 |
| | | | | limit: -1 |
| | | | | filter: |
| | | | | orderBy: [] |
| | | | | src: COLUMN[0] |
| | | | | props: [ |
| | | | | { |
| | | | | "props": [ |
| | | | | "_tag", |
| | | | | "embedding", |
| | | | | "name" |
| | | | | ], |
| | | | | "tagId": 3 |
| | | | | } |
| | | | | ] |
| | | | | exprs: |
-----+-------------+--------------+----------------+------------------------------
| 0 | Start | | | outputVar: { |
| | | | | "colNames": [], |
| | | | | "type": "DATASET", |
| | | | | "name": "__Start_0" |
| | | | | } |
-----+-------------+--------------+----------------+------------------------------
Fetch 查询会最终调用GetVerticesProcessor executor。这个 executor 会生成 Node 的执行计划，执行 GetTagPropNode 的计划，我们需要在GetTagPropNode::doExecute方法中解析在TagNode::doExecute中从 kvstore 中获取的 value。
C++
List row;
for (auto* tagNode : tagNodes_) {
ret = tagNode->collectTagPropsIfValid(
[&row, tagNode, this](const std::vector<PropContext>* props) -> nebula::cpp2::ErrorCode {
for (const auto& prop : *props) {
if (prop.returned_) {
row.emplace_back(Value());
}
if (prop.filtered_ && expCtx_ != nullptr) {
expCtx_->setTagProp(tagNode->getTagName(), prop.name_, Value());
}
}
return nebula::cpp2::ErrorCode::SUCCEEDED;
},
[&row, vIdLen, isIntId, tagNode, this](
folly::StringPiece key,
RowReaderWrapper* reader,
RowReaderWrapper* vectorReader,
const std::vector<PropContext>* props) -> nebula::cpp2::ErrorCode {
// 向row中添加解析后vertex的属性对应的value
auto status = QueryUtils::collectVertexProps(
key, vIdLen, isIntId, reader, vecReader, props, row, expCtx_.get(), tagNode->getTagName());
if (!status.ok()) {
return nebula::cpp2::ErrorCode::E_TAG_PROP_NOT_FOUND;
}
return nebula::cpp2::ErrorCode::SUCCEEDED;
});
}