gitlore/src/core/timeline_seed.rs

use std::collections::HashSet;

use rusqlite::Connection;
use tracing::debug;

use crate::core::error::Result;
use crate::core::timeline::{
    EntityRef, MatchedDiscussion, TimelineEvent, TimelineEventType, resolve_entity_by_iid,
    resolve_entity_ref, truncate_to_chars,
};
use crate::embedding::ollama::OllamaClient;
use crate::search::{FtsQueryMode, SearchFilters, SearchMode, search_hybrid, to_fts_query};

/// Result of the seed + hydrate phases.
pub struct SeedResult {
    pub seed_entities: Vec<EntityRef>,
    pub evidence_notes: Vec<TimelineEvent>,
    /// Discussions matched during seeding, to be collected as full threads.
    pub matched_discussions: Vec<MatchedDiscussion>,
    /// The search mode actually used (hybrid with fallback info).
    pub search_mode: String,
}

/// Run the SEED + HYDRATE phases of the timeline pipeline.
///
/// 1. SEED: Hybrid search (FTS + vector via RRF) over documents -> matched document IDs
/// 2. HYDRATE: Map document IDs -> source entities + top matched notes as evidence
///
/// When `client` is `None` or Ollama is unavailable, falls back to FTS-only search.
/// Discussion documents are resolved to their parent entity (issue or MR).
/// Entities are deduplicated. Evidence notes are capped at `max_evidence`.
pub async fn seed_timeline(
    conn: &Connection,
    client: Option<&OllamaClient>,
    query: &str,
    project_id: Option<i64>,
    since_ms: Option<i64>,
    max_seeds: usize,
    max_evidence: usize,
) -> Result<SeedResult> {
    let fts_query = to_fts_query(query, FtsQueryMode::Safe);
    if fts_query.is_empty() {
        return Ok(SeedResult {
            seed_entities: Vec::new(),
            evidence_notes: Vec::new(),
            matched_discussions: Vec::new(),
            search_mode: "lexical".to_owned(),
        });
    }

    // Use hybrid search for seed entity discovery (better recall than FTS alone).
    // search_hybrid gracefully falls back to FTS-only when Ollama is unavailable.
    let filters = SearchFilters {
        project_id,
        updated_since: since_ms,
        limit: max_seeds.saturating_mul(3),
        ..SearchFilters::default()
    };

    let (hybrid_results, warnings) = search_hybrid(
        conn,
        client,
        query,
        SearchMode::Hybrid,
        &filters,
        FtsQueryMode::Safe,
    )
    .await?;

    let search_mode = if warnings
        .iter()
        .any(|w| w.contains("falling back") || w.contains("FTS only"))
    {
        "lexical (hybrid fallback)".to_owned()
    } else if client.is_some() && !hybrid_results.is_empty() {
        "hybrid".to_owned()
    } else {
        "lexical".to_owned()
    };

    for w in &warnings {
        debug!(warning = %w, "hybrid search warning during timeline seeding");
    }

    let (seed_entities, matched_discussions) = resolve_documents_to_entities(
        conn,
        &hybrid_results
            .iter()
            .map(|r| r.document_id)
            .collect::<Vec<_>>(),
        max_seeds,
    )?;

    // Evidence notes stay FTS-only (supplementary context, not worth a second embedding call)
    let evidence_notes = find_evidence_notes(conn, &fts_query, project_id, since_ms, max_evidence)?;

    Ok(SeedResult {
        seed_entities,
        evidence_notes,
        matched_discussions,
        search_mode,
    })
}

/// Seed the timeline directly from an entity IID, bypassing search entirely.
///
/// Used for `issue:42` / `mr:99` syntax. Resolves the entity, gathers ALL its
/// discussions, and returns a `SeedResult` compatible with the rest of the pipeline.
pub fn seed_timeline_direct(
    conn: &Connection,
    entity_type: &str,
    iid: i64,
    project_id: Option<i64>,
) -> Result<SeedResult> {
    let entity_ref = resolve_entity_by_iid(conn, entity_type, iid, project_id)?;

    // Gather all discussions for this entity (not search-matched, ALL of them)
    let entity_id_col = match entity_type {
        "issue" => "issue_id",
        "merge_request" => "merge_request_id",
        _ => {
            return Ok(SeedResult {
                seed_entities: vec![entity_ref],
                evidence_notes: Vec::new(),
                matched_discussions: Vec::new(),
                search_mode: "direct".to_owned(),
            });
        }
    };

    let sql = format!("SELECT id, project_id FROM discussions WHERE {entity_id_col} = ?1");
    let mut stmt = conn.prepare(&sql)?;
    let matched_discussions: Vec<MatchedDiscussion> = stmt
        .query_map(rusqlite::params![entity_ref.entity_id], |row| {
            Ok(MatchedDiscussion {
                discussion_id: row.get(0)?,
                entity_type: entity_type.to_owned(),
                entity_id: entity_ref.entity_id,
                project_id: row.get(1)?,
            })
        })?
        .collect::<std::result::Result<Vec<_>, _>>()?;

    Ok(SeedResult {
        seed_entities: vec![entity_ref],
        evidence_notes: Vec::new(),
        matched_discussions,
        search_mode: "direct".to_owned(),
    })
}

/// Resolve a list of document IDs to deduplicated entity refs and matched discussions.
/// Discussion and note documents are resolved to their parent entity (issue or MR).
/// Returns (entities, matched_discussions).
fn resolve_documents_to_entities(
    conn: &Connection,
    document_ids: &[i64],
    max_entities: usize,
) -> Result<(Vec<EntityRef>, Vec<MatchedDiscussion>)> {
    if document_ids.is_empty() {
        return Ok((Vec::new(), Vec::new()));
    }

    let placeholders: String = document_ids
        .iter()
        .map(|_| "?")
        .collect::<Vec<_>>()
        .join(",");
    let sql = format!(
        r"
        SELECT d.source_type, d.source_id, d.project_id,
               COALESCE(disc.issue_id, note_disc.issue_id) AS issue_id,
               COALESCE(disc.merge_request_id, note_disc.merge_request_id) AS mr_id,
               COALESCE(disc.id, note_disc.id) AS discussion_id
        FROM documents d
        LEFT JOIN discussions disc ON disc.id = d.source_id AND d.source_type = 'discussion'
        LEFT JOIN notes n ON n.id = d.source_id AND d.source_type = 'note'
        LEFT JOIN discussions note_disc ON note_disc.id = n.discussion_id AND d.source_type = 'note'
        WHERE d.id IN ({placeholders})
        ORDER BY CASE d.id {order_clause} END
        ",
        order_clause = document_ids
            .iter()
            .enumerate()
            .map(|(i, id)| format!("WHEN {id} THEN {i}"))
            .collect::<Vec<_>>()
            .join(" "),
    );

    let mut stmt = conn.prepare(&sql)?;
    let params: Vec<&dyn rusqlite::types::ToSql> = document_ids
        .iter()
        .map(|id| id as &dyn rusqlite::types::ToSql)
        .collect();
    let rows = stmt.query_map(params.as_slice(), |row| {
        Ok((
            row.get::<_, String>(0)?,      // source_type
            row.get::<_, i64>(1)?,         // source_id
            row.get::<_, i64>(2)?,         // project_id
            row.get::<_, Option<i64>>(3)?, // issue_id (coalesced)
            row.get::<_, Option<i64>>(4)?, // mr_id (coalesced)
            row.get::<_, Option<i64>>(5)?, // discussion_id (coalesced)
        ))
    })?;

    let mut seen_entities = HashSet::new();
    let mut seen_discussions = HashSet::new();
    let mut entities = Vec::new();
    let mut matched_discussions = Vec::new();

    for row_result in rows {
        let (source_type, source_id, proj_id, disc_issue_id, disc_mr_id, discussion_id) =
            row_result?;

        let (entity_type, entity_id) = match source_type.as_str() {
            "issue" => ("issue".to_owned(), source_id),
            "merge_request" => ("merge_request".to_owned(), source_id),
            "discussion" | "note" => {
                if let Some(issue_id) = disc_issue_id {
                    ("issue".to_owned(), issue_id)
                } else if let Some(mr_id) = disc_mr_id {
                    ("merge_request".to_owned(), mr_id)
                } else {
                    continue; // orphaned discussion/note
                }
            }
            _ => continue,
        };

        // Capture matched discussion (deduplicated)
        if let Some(disc_id) = discussion_id
            && (source_type == "discussion" || source_type == "note")
            && seen_discussions.insert(disc_id)
        {
            matched_discussions.push(MatchedDiscussion {
                discussion_id: disc_id,
                entity_type: entity_type.clone(),
                entity_id,
                project_id: proj_id,
            });
        }

        // Entity dedup
        let key = (entity_type.clone(), entity_id);
        if !seen_entities.insert(key) {
            continue;
        }

        if let Some(entity_ref) = resolve_entity_ref(conn, &entity_type, entity_id, Some(proj_id))?
        {
            entities.push(entity_ref);
        }

        if entities.len() >= max_entities {
            break;
        }
    }

    Ok((entities, matched_discussions))
}

/// Find evidence notes: FTS5-matched discussion notes that provide context.
fn find_evidence_notes(
    conn: &Connection,
    fts_query: &str,
    project_id: Option<i64>,
    since_ms: Option<i64>,
    max_evidence: usize,
) -> Result<Vec<TimelineEvent>> {
    let sql = r"
        SELECT n.id AS note_id, n.body, n.created_at, n.author_username,
               disc.id AS discussion_id,
               CASE WHEN disc.issue_id IS NOT NULL THEN 'issue' ELSE 'merge_request' END AS parent_type,
               COALESCE(disc.issue_id, disc.merge_request_id) AS parent_entity_id,
               d.project_id
        FROM documents_fts
        JOIN documents d ON d.id = documents_fts.rowid
        JOIN discussions disc ON disc.id = d.source_id AND d.source_type = 'discussion'
        JOIN notes n ON n.discussion_id = disc.id AND n.is_system = 0
        WHERE documents_fts MATCH ?1
          AND (?2 IS NULL OR d.project_id = ?2)
          AND (?3 IS NULL OR d.updated_at >= ?3)
        ORDER BY rank
        LIMIT ?4
    ";

    let mut stmt = conn.prepare(sql)?;
    let rows = stmt.query_map(
        rusqlite::params![fts_query, project_id, since_ms, max_evidence as i64],
        |row| {
            Ok((
                row.get::<_, i64>(0)?,            // note_id
                row.get::<_, Option<String>>(1)?, // body
                row.get::<_, i64>(2)?,            // created_at
                row.get::<_, Option<String>>(3)?, // author
                row.get::<_, i64>(4)?,            // discussion_id
                row.get::<_, String>(5)?,         // parent_type
                row.get::<_, i64>(6)?,            // parent_entity_id
                row.get::<_, i64>(7)?,            // project_id
            ))
        },
    )?;

    let mut events = Vec::new();

    for row_result in rows {
        let (
            note_id,
            body,
            created_at,
            author,
            discussion_id,
            parent_type,
            parent_entity_id,
            proj_id,
        ) = row_result?;

        let snippet = truncate_to_chars(body.as_deref().unwrap_or(""), 200);

        let entity_ref = resolve_entity_ref(conn, &parent_type, parent_entity_id, Some(proj_id))?;
        let (iid, project_path) = match entity_ref {
            Some(ref e) => (e.entity_iid, e.project_path.clone()),
            None => {
                debug!(
                    parent_type,
                    parent_entity_id,
                    proj_id,
                    "Skipping evidence note: parent entity not found (orphaned discussion)"
                );
                continue;
            }
        };

        events.push(TimelineEvent {
            timestamp: created_at,
            entity_type: parent_type,
            entity_id: parent_entity_id,
            entity_iid: iid,
            project_path,
            event_type: TimelineEventType::NoteEvidence {
                note_id,
                snippet,
                discussion_id: Some(discussion_id),
            },
            summary: format!("Note by {}", author.as_deref().unwrap_or("unknown")),
            actor: author,
            url: None,
            is_seed: true,
        });
    }

    Ok(events)
}

#[cfg(test)]
#[path = "timeline_seed_tests.rs"]
mod tests;