chore: rename get_id to id, add doc comments, pre-allocate format_history

common/src/storage/db.rs

@@ -143,7 +143,7 @@ impl SurrealDbClient {
         T: StoredObject + Send + Sync + 'static,
     {
         self.client
-            .create((T::table_name(), item.get_id()))
+            .create((T::table_name(), item.id()))
             .content(item)
             .await
     }
@@ -159,7 +159,7 @@ impl SurrealDbClient {
     where
         T: StoredObject + Send + Sync + 'static,
     {
-        let id = item.get_id().to_string();
+        let id = item.id().to_string();
         self.client
             .upsert((T::table_name(), id))
             .content(item)

common/src/storage/indexes.rs

@@ -581,14 +581,22 @@ async fn poll_index_build_status(
     Ok(())
 }
 
+/// Snapshot of an index build progress as reported by SurrealDB's `INFO FOR INDEX`.
 #[derive(Debug, PartialEq)]
 struct IndexBuildSnapshot {
+    /// Current build status string (e.g., `"indexing"`, `"ready"`, `"error"`).
     status: String,
+    /// Number of rows present when the build started.
     initial: u64,
+    /// Number of rows still pending processing.
     pending: u64,
+    /// Number of rows updated since the build started.
     updated: u64,
+    /// Total rows processed so far (`initial + updated`).
     processed: u64,
+    /// Total rows expected (from `SELECT count()` before the build), if available.
     total_rows: Option<u64>,
+    /// Progress as a percentage of `processed / total_rows`, if `total_rows` is known.
     progress_pct: Option<f64>,
 }
 

common/src/storage/types/analytics.rs

@@ -16,7 +16,7 @@ impl StoredObject for Analytics {
         "analytics"
     }
 
-    fn get_id(&self) -> &str {
+    fn id(&self) -> &str {
         &self.id
     }
 }

common/src/storage/types/file_info.rs

@@ -19,26 +19,34 @@ use crate::{
     stored_object,
 };
 
+/// Errors that can occur during file storage operations.
 #[derive(Error, Debug)]
 pub enum FileError {
+    /// No file record found with the given UUID.
     #[error("file not found for uuid: {0}")]
     FileNotFound(String),
 
+    /// Underlying I/O operation failed.
     #[error("io error occurred: {0}")]
     Io(#[from] std::io::Error),
 
+    /// A file with the same SHA-256 hash already exists.
     #[error("duplicate file detected with sha256: {0}")]
     DuplicateFile(String),
 
+    /// Database operation on the file record failed.
     #[error("surrealdb error: {0}")]
     SurrealError(#[from] surrealdb::Error),
 
+    /// Failed to persist the temporary file to its final location.
     #[error("failed to persist file: {0}")]
     PersistError(#[from] tempfile::PersistError),
 
+    /// Upload metadata did not include a file name.
     #[error("file name missing in metadata")]
     MissingFileName,
 
+    /// The underlying object store operation failed.
     #[error("object store error: {0}")]
     ObjectStore(#[from] ObjectStoreError),
 }

common/src/storage/types/ingestion_task.rs

@@ -68,12 +68,16 @@ impl TaskState {
     }
 }
 
+/// Information about an error that occurred during task processing.
 #[derive(Debug, Clone, serde::Serialize, serde::Deserialize, PartialEq, Eq, Default)]
 pub struct TaskErrorInfo {
+    /// Machine-readable error code (e.g., `"pipeline_error"`).
     pub code: Option<String>,
+    /// Human-readable error description.
     pub message: String,
 }
 
+/// Internal events that drive the task state machine transitions.
 #[derive(Debug, Clone, Copy)]
 enum TaskTransition {
     StartProcessing,

common/src/storage/types/message.rs

@@ -60,7 +60,11 @@ impl fmt::Display for Message {
 // helper function to format a vector of messages
 #[must_use]
 pub fn format_history(history: &[Message]) -> String {
-    let mut out = String::new();
+    let estimated: usize = history
+        .iter()
+        .map(|m| m.content.len() + 10)
+        .sum();
+    let mut out = String::with_capacity(estimated);
     for (i, msg) in history.iter().enumerate() {
         if i > 0 {
             out.push('\n');

common/src/storage/types/mod.rs

@@ -20,7 +20,7 @@ pub mod user;
 
 pub trait StoredObject: Serialize + for<'de> Deserialize<'de> {
     fn table_name() -> &'static str;
-    fn get_id(&self) -> &str;
+    fn id(&self) -> &str;
 }
 
 #[macro_export]
@@ -52,7 +52,7 @@ macro_rules! stored_object {
                 $table
             }
 
-            fn get_id(&self) -> &str {
+            fn id(&self) -> &str {
                 &self.id
             }
         }

common/src/storage/types/system_settings.rs

@@ -28,7 +28,7 @@ impl StoredObject for SystemSettings {
         "system_settings"
     }
 
-    fn get_id(&self) -> &str {
+    fn id(&self) -> &str {
         &self.id
     }
 }

common/src/utils/ingest_limits.rs

@@ -1,8 +1,11 @@
 use super::config::AppConfig;
 
+/// Errors raised when validating ingestion payloads against configured limits.
 #[derive(Debug, Clone, PartialEq, Eq)]
 pub enum IngestValidationError {
+    /// The payload exceeds a configured size limit (content, context, or category).
     PayloadTooLarge(String),
+    /// The request violates a non-size constraint (e.g., too many files).
     BadRequest(String),
 }
 

evaluations/src/corpus/orchestrator.rs

@@ -340,7 +340,7 @@ pub async fn ensure_corpus(
         corpus_questions.push(CorpusQuestion {
             question_id: case.question.id.clone(),
             paragraph_id: case.paragraph.id.clone(),
-            text_content_id: record.shard.text_content.get_id().to_string(),
+            text_content_id: record.shard.text_content.id().to_string(),
             question_text: case.question.question.clone(),
             answers: case.question.answers.clone(),
             is_impossible: case.question.is_impossible,

evaluations/src/corpus/store.rs

@@ -575,7 +575,7 @@ fn validate_answers(
             if chunk_text.contains(&needle)
                 || (!needle_norm.is_empty() && chunk_norm.contains(&needle_norm))
             {
-                matches.insert(chunk.chunk.get_id().to_string());
+                matches.insert(chunk.chunk.id().to_string());
                 found_any = true;
             }
         }

evaluations/src/namespace.rs

@@ -216,7 +216,7 @@ pub(crate) async fn ensure_eval_user(db: &SurrealDbClient) -> Result<User> {
         theme: Theme::System,
     };
 
-    if let Some(existing) = db.get_item::<User>(user.get_id()).await? {
+    if let Some(existing) = db.get_item::<User>(user.id()).await? {
         return Ok(existing);
     }
 

evaluations/src/pipeline/stages/run_queries.rs

@@ -247,7 +247,7 @@ pub(crate) async fn run_queries(
                             || candidate
                                 .chunks
                                 .iter()
-                                .any(|chunk| expected_chunk_ids_set.contains(&chunk.chunk.get_id()))
+                                .any(|chunk| expected_chunk_ids_set.contains(&chunk.chunk.id()))
                     } else {
                         true
                     };

evaluations/src/types.rs

@@ -194,7 +194,7 @@ impl EvaluationCandidate {
     fn from_entity(entity: RetrievedEntity) -> Self {
         let entity_category = Some(format!("{:?}", entity.entity.entity_type));
         Self {
-            entity_id: entity.entity.get_id().to_string(),
+            entity_id: entity.entity.id().to_string(),
             source_id: entity.entity.source_id.clone(),
             entity_name: entity.entity.name.clone(),
             entity_description: Some(entity.entity.description.clone()),
@@ -207,7 +207,7 @@ impl EvaluationCandidate {
     fn from_chunk(chunk: RetrievedChunk) -> Self {
         let snippet = chunk_snippet(&chunk.chunk.chunk);
         Self {
-            entity_id: chunk.chunk.get_id().to_string(),
+            entity_id: chunk.chunk.id().to_string(),
             source_id: chunk.chunk.source_id.clone(),
             entity_name: chunk.chunk.source_id.clone(),
             entity_description: Some(snippet),
@@ -423,11 +423,11 @@ pub fn build_case_diagnostics(
         let mut chunk_entries = Vec::new();
         for chunk in &candidate.chunks {
             let contains_answer = text_contains_answer(&chunk.chunk.chunk, answers_lower);
-            let expected_chunk = expected_set.contains(chunk.chunk.get_id());
-            seen_chunks.insert(chunk.chunk.get_id().to_string());
-            attached_chunk_ids.push(chunk.chunk.get_id().to_string());
+            let expected_chunk = expected_set.contains(chunk.chunk.id());
+            seen_chunks.insert(chunk.chunk.id().to_string());
+            attached_chunk_ids.push(chunk.chunk.id().to_string());
             chunk_entries.push(ChunkDiagnosticsEntry {
-                chunk_id: chunk.chunk.get_id().to_string(),
+                chunk_id: chunk.chunk.id().to_string(),
                 score: chunk.score,
                 contains_answer,
                 expected_chunk,

ingestion-pipeline/src/pipeline/services.rs

@@ -240,7 +240,7 @@ impl PipelineServices for DefaultPipelineServices {
     ) -> Result<(Vec<EmbeddedKnowledgeEntity>, Vec<KnowledgeRelationship>), AppError> {
         analysis
             .to_database_entities(
-                content.get_id(),
+                content.id(),
                 &content.user_id,
                 &self.openai_client,
                 &self.db,
@@ -271,7 +271,7 @@ impl PipelineServices for DefaultPipelineServices {
                 .await
                 .map_err(|e| AppError::InternalError(format!("FastEmbed embedding for chunk failed: {e}")))?;
             let chunk_struct = TextChunk::new(
-                content.get_id().to_string(),
+                content.id().to_string(),
                 chunk_text,
                 content.user_id.clone(),
             );

retrieval-pipeline/src/scoring.rs

@@ -188,7 +188,7 @@ pub fn merge_scored_by_id<T, S: std::hash::BuildHasher>(
     T: StoredObject + Clone,
 {
     for scored in incoming {
-        let id = scored.item.get_id().to_owned();
+        let id = scored.item.id().to_owned();
         target
             .entry(id)
             .and_modify(|existing| {
@@ -218,7 +218,7 @@ where
         b.fused
             .partial_cmp(&a.fused)
             .unwrap_or(Ordering::Equal)
-            .then_with(|| a.item.get_id().cmp(b.item.get_id()))
+            .then_with(|| a.item.id().cmp(b.item.id()))
     });
 }
 
@@ -250,11 +250,11 @@ where
             b_score
                 .partial_cmp(&a_score)
                 .unwrap_or(Ordering::Equal)
-                .then_with(|| a.item.get_id().cmp(b.item.get_id()))
+                .then_with(|| a.item.id().cmp(b.item.id()))
         });
 
         for (rank, candidate) in vector_ranked.into_iter().enumerate() {
-            let id = candidate.item.get_id().to_owned();
+            let id = candidate.item.id().to_owned();
             let entry = merged
                 .entry(id.clone())
                 .or_insert_with(|| Scored::new(candidate.item.clone()));
@@ -280,11 +280,11 @@ where
             b_score
                 .partial_cmp(&a_score)
                 .unwrap_or(Ordering::Equal)
-                .then_with(|| a.item.get_id().cmp(b.item.get_id()))
+                .then_with(|| a.item.id().cmp(b.item.id()))
         });
 
         for (rank, candidate) in fts_ranked.into_iter().enumerate() {
-            let id = candidate.item.get_id().to_owned();
+            let id = candidate.item.id().to_owned();
             let entry = merged
                 .entry(id.clone())
                 .or_insert_with(|| Scored::new(candidate.item.clone()));