fix(code-graph): v2 pipeline hardening — error propagation, stack guards, tracing (1ca434a8) · Commits · GitLab.org / orbit / GitLab Knowledge Graph

config/default.yaml

+1 −0

Original line number	Diff line number	Diff line
		@@ -68,6 +68,7 @@ engine:
		respect_gitignore: true
		worker_threads: 0
		max_concurrent_languages: 0
		per_file_timeout_ms: 5000
		namespace-deletion:
		concurrency_group: code
		max_attempts: 1 # re-dispatched daily, no need to retry

config/schemas/config.schema.json

+11 −0

Original line number	Diff line number	Diff line
		@@ -56,6 +56,7 @@
		"max_concurrent_languages": 0,
		"max_file_size_bytes": 5000000,
		"max_files": 1000000,
		"per_file_timeout_ms": 5000,
		"respect_gitignore": true,
		"worker_threads": 0
		},
		@@ -337,6 +338,13 @@
		"default": 1000000,
		"minimum": 0
		},
		"per_file_timeout_ms": {
		"description": "Global per-file resolution timeout in milliseconds.\nApplied to all languages unless the language's own DSL rules\nspecify a different value. 0 = no global timeout.",
		"type": "integer",
		"format": "uint64",
		"default": 5000,
		"minimum": 0
		},
		"respect_gitignore": {
		"type": "boolean",
		"default": true
		@@ -378,6 +386,7 @@
		"max_concurrent_languages": 0,
		"max_file_size_bytes": 5000000,
		"max_files": 1000000,
		"per_file_timeout_ms": 5000,
		"respect_gitignore": true,
		"worker_threads": 0
		}
		@@ -449,6 +458,7 @@
		"max_concurrent_languages": 0,
		"max_file_size_bytes": 5000000,
		"max_files": 1000000,
		"per_file_timeout_ms": 5000,
		"respect_gitignore": true,
		"worker_threads": 0
		},
		@@ -637,6 +647,7 @@
		"max_concurrent_languages": 0,
		"max_file_size_bytes": 5000000,
		"max_files": 1000000,
		"per_file_timeout_ms": 5000,
		"respect_gitignore": true,
		"worker_threads": 0
		},

crates/code-graph/src/v2/dsl/engine.rs

+1 −0

Original line number	Diff line number	Diff line
		@@ -681,6 +681,7 @@ impl LanguageSpec {
		if stacker::remaining_stack().unwrap_or(usize::MAX)
		< crate::legacy::parser::MINIMUM_STACK_REMAINING
		{
		tracing::warn!("stack limit reached during AST walk, subtree truncated");
		return;
		}

crates/code-graph/src/v2/error.rs

0 → 100644

+96 −0

Original line number	Diff line number	Diff line
		use std::fmt;

		/// Errors from the v2 code-graph pipeline.
		///
		/// Covers infrastructure failures (thread pools, I/O), per-file
		/// processing errors, and graph invariant violations.
		#[derive(Debug)]
		pub enum CodeGraphError {
		/// A file could not be read from disk.
		FileRead {
		path: String,
		source: std::io::Error,
		},

		/// A file could not be parsed (tree-sitter / OXC / Prism failure).
		ParseFailed { path: String, message: String },

		/// A rayon thread pool could not be created (OS resource limits).
		ThreadPoolCreation {
		language: String,
		source: rayon::ThreadPoolBuildError,
		},

		/// The sentinel watchdog thread could not be spawned.
		SentinelSpawn { source: std::io::Error },

		/// A graph node was expected to be a Definition but wasn't.
		UnexpectedNodeType { expected: &'static str, got: String },

		/// Arrow conversion failed for a graph.
		ArrowConversion { message: String },

		/// A sink write failed.
		SinkWrite { table: String, message: String },

		/// Generic internal error with context.
		Internal { context: String, message: String },
		}

		impl fmt::Display for CodeGraphError {
		fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
		match self {
		Self::FileRead { path, source } => {
		write!(f, "failed to read {path}: {source}")
		}
		Self::ParseFailed { path, message } => {
		write!(f, "failed to parse {path}: {message}")
		}
		Self::ThreadPoolCreation { language, source } => {
		write!(f, "failed to create thread pool for {language}: {source}")
		}
		Self::SentinelSpawn { source } => {
		write!(f, "failed to spawn sentinel thread: {source}")
		}
		Self::UnexpectedNodeType { expected, got } => {
		write!(f, "expected {expected} node, got {got}")
		}
		Self::ArrowConversion { message } => {
		write!(f, "arrow conversion failed: {message}")
		}
		Self::SinkWrite { table, message } => {
		write!(f, "sink write to {table} failed: {message}")
		}
		Self::Internal { context, message } => {
		write!(f, "{context}: {message}")
		}
		}
		}
		}

		impl std::error::Error for CodeGraphError {
		fn source(&self) -> Option<&(dyn std::error::Error + 'static)> {
		match self {
		Self::FileRead { source, .. } => Some(source),
		Self::SentinelSpawn { source, .. } => Some(source),
		Self::ThreadPoolCreation { source, .. } => Some(source),
		_ => None,
		}
		}
		}

		impl CodeGraphError {
		/// Returns a stable stage label for metrics recording.
		pub fn stage(&self) -> &'static str {
		match self {
		Self::FileRead { .. } => "file_read",
		Self::ParseFailed { .. } => "parse",
		Self::ThreadPoolCreation { .. } => "thread_pool",
		Self::SentinelSpawn { .. } => "sentinel",
		Self::UnexpectedNodeType { .. } => "graph_node",
		Self::ArrowConversion { .. } => "arrow_conversion",
		Self::SinkWrite { .. } => "sink_write",
		Self::Internal { .. } => "internal",
		}
		}
		}

crates/code-graph/src/v2/langs/custom/js/pipeline.rs

+1 −1

Original line number	Diff line number	Diff line
		@@ -60,7 +60,7 @@ impl LanguagePipeline for JsPipeline {

		if !errors.is_empty() {
		for error in &errors {
		log::warn!("[v2-js] skipped {}: {}", error.file_path, error.error);
		tracing::warn!(path = %error.file_path, error = %error.error, "js: skipped file");
		}
		}
		Ok(())