Verified Commit 14a3a63e authored by Michael Usachenko's avatar Michael Usachenko Committed by GitLab
Browse files

refactor(code-graph): reorganize into parent crate with sub-crates

parent 4ea5ca11

AGENTS.md

+4 −3
Original line number Diff line number Diff line
@@ -81,15 +81,16 @@ Single binary: `gkg-server` (4 modes: Webserver, Indexer, DispatchIndexing, Heal 
| `query-engine/formatters` | Result formatters (graph, raw row, goon) |

| `indexer` | NATS consumer, SDLC + code + namespace deletion handler modules, worker pools, scheduler, `testkit/` |

| `ontology` | Loads/validates YAML ontology, query validation helpers |

| `code-parser` | Multi-language parser (7 langs), tree-sitter + swc, extracts definitions/imports/references |

| `code-graph` | Builds in-memory property graphs from parsed code |

| `code-graph` | Parent crate for code parsing and graph construction; re-exports `treesitter-visit`, `parser-core`, `code-graph-linker` |

| `code-graph/treesitter-visit` | Tree-sitter language bindings wrapper |

| `code-graph/parser` | Multi-language parser (7 langs), tree-sitter + swc, extracts definitions/imports/references |

| `code-graph/linker` | Builds in-memory property graphs from parsed code |

| `utils` | Shared ClickHouse parameter types (`ChScalar`, `ChType`) and Arrow extraction utilities |

| `clickhouse-client` | Async ClickHouse client, Arrow-IPC streaming, `QueryProfiler` for per-query stats |

| `query-engine/profiler` | Standalone CLI for profiling GKG queries directly against ClickHouse |

| `siphon-proto` | Protobuf types for CDC replication events |

| `labkit-rs` | Logging, correlation IDs, OpenTelemetry metrics |

| `health-check` | K8s readiness/liveness probes |

| `treesitter-visit` | Tree-sitter language bindings wrapper |

| `cli` | Local `orbit index` and `orbit query` commands |

| `gitlab-client` | GitLab REST/JWT client for Rails API calls |

| `integration-testkit` | Shared ClickHouse testcontainer helpers, `MockRedactionService`, and `ResponseView` assertion framework for integration tests |

CLAUDE.md

+4 −3
Original line number Diff line number Diff line
@@ -81,15 +81,16 @@ Single binary: `gkg-server` (4 modes: Webserver, Indexer, DispatchIndexing, Heal 
| `query-engine/formatters` | Result formatters (graph, raw row, goon) |

| `indexer` | NATS consumer, SDLC + code + namespace deletion handler modules, worker pools, scheduler, `testkit/` |

| `ontology` | Loads/validates YAML ontology, query validation helpers |

| `code-parser` | Multi-language parser (7 langs), tree-sitter + swc, extracts definitions/imports/references |

| `code-graph` | Builds in-memory property graphs from parsed code |

| `code-graph` | Parent crate for code parsing and graph construction; re-exports `treesitter-visit`, `parser-core`, `code-graph-linker` |

| `code-graph/treesitter-visit` | Tree-sitter language bindings wrapper |

| `code-graph/parser` | Multi-language parser (7 langs), tree-sitter + swc, extracts definitions/imports/references |

| `code-graph/linker` | Builds in-memory property graphs from parsed code |

| `utils` | Shared ClickHouse parameter types (`ChScalar`, `ChType`) and Arrow extraction utilities |

| `clickhouse-client` | Async ClickHouse client, Arrow-IPC streaming, `QueryProfiler` for per-query stats |

| `query-engine/profiler` | Standalone CLI for profiling GKG queries directly against ClickHouse |

| `siphon-proto` | Protobuf types for CDC replication events |

| `labkit-rs` | Logging, correlation IDs, OpenTelemetry metrics |

| `health-check` | K8s readiness/liveness probes |

| `treesitter-visit` | Tree-sitter language bindings wrapper |

| `cli` | Local `orbit index` and `orbit query` commands |

| `gitlab-client` | GitLab REST/JWT client for Rails API calls |

| `integration-testkit` | Shared ClickHouse testcontainer helpers, `MockRedactionService`, and `ResponseView` assertion framework for integration tests |

Cargo.lock

+9 −0
Original line number Diff line number Diff line
@@ -1209,6 +1209,15 @@ dependencies = [ 
[[package]]

name = "code-graph"

version = "0.1.0"

dependencies = [

 "code-graph-linker",

 "parser-core",

 "treesitter-visit",

]



[[package]]

name = "code-graph-linker"

version = "0.1.0"

dependencies = [

 "anyhow",

 "chrono",

Cargo.toml

+4 −3
Original line number Diff line number Diff line
@@ -4,9 +4,10 @@ version = "0.1.0" 
[workspace]

resolver = "2"

members = [

  "crates/treesitter-visit",

  "crates/code-parser",

  "crates/code-graph",

  "crates/code-graph/treesitter-visit",

  "crates/code-graph/parser",

  "crates/code-graph/linker",

  "crates/clickhouse-client",

  "crates/indexer",

  "crates/query-engine",
@@ -84,7 +85,7 @@ criterion = { version = "0.8.1", features = ["html_reports"] } 
tracing-test = "0.2.5"



# Parser-core

parser-core = { path = "crates/code-parser" }

parser-core = { path = "crates/code-graph/parser" }

clickhouse-client = { path = "crates/clickhouse-client" }

duckdb = { version = "1.10501.0", features = ["vtab-arrow", "appender-arrow"] }

gitalisk-core = { git = "https://gitlab.com/gitlab-org/rust/gitalisk.git", tag = "v0.7.0" }

crates/cli/src/main.rs

+3 −3
Original line number Diff line number Diff line
@@ -2,8 +2,8 @@ mod workspace; 


use anyhow::{Context, Result};

use clap::{Parser, Subcommand};

use code_graph::indexer::{IndexingConfig, RepositoryIndexer};

use code_graph::loading::DirectoryFileSource;

use code_graph::linker::indexer::{IndexingConfig, RepositoryIndexer, RepositoryIndexingResult};

use code_graph::linker::loading::DirectoryFileSource;

use ontology::Ontology;

use query_engine::compiler::SecurityContext;

use serde::Serialize;
@@ -220,7 +220,7 @@ async fn run_index(path: PathBuf, threads: usize, show_stats: bool) -> Result<() 
fn build_index_output(

    repo_name: &str,

    path: &str,

    result: &code_graph::indexer::RepositoryIndexingResult,

    result: &RepositoryIndexingResult,

    show_stats: bool,

) -> IndexOutput {

    let (graph, rel_counts, def_counts) = match result.graph_data {