sqlrite-engine 0.10.0

//! On-disk persistence for a `Database`, using fixed-size paged files.
//!
//! The file is a sequence of 4 KiB pages. Page 0 holds the header
//! (magic, version, page count, schema-root pointer). Every other page carries
//! a small per-page header (type tag + next-page pointer + payload length)
//! followed by a payload of up to 4089 bytes.
//!
//! **Storage strategy (format version 2, Phase 3c.5).**
//!
//! - Each `Table`'s rows live as **cells** in a chain of `TableLeaf` pages.
//!   Cell layout and slot directory are in `cell.rs` / `table_page.rs`;
//!   cells that exceed the inline threshold spill into an overflow chain
//!   via `overflow.rs`.
//! - The schema catalog is itself a regular table named `sqlrite_master`,
//!   with one row per user table:
//!       `(name TEXT PRIMARY KEY, sql TEXT NOT NULL,
//!         rootpage INTEGER NOT NULL, last_rowid INTEGER NOT NULL)`
//!   This is the SQLite-style approach: the schema of `sqlrite_master`
//!   itself is hardcoded into the engine so the open path can bootstrap.
//! - Page 0's `schema_root_page` field points at the first leaf of
//!   `sqlrite_master`.
//!
//! **Format version.** Version 2 is not compatible with files produced by
//! earlier commits. Opening a v1 file returns a clean error — users on
//! old files have to regenerate them from CREATE/INSERT, as there's no
//! production data to migrate yet.

// Data-layer modules. Not every helper in these modules is used by save/open
// yet — some exist for tests, some for future maintenance operations.
// Module-level #[allow(dead_code)] keeps the build quiet without dotting
// the modules with per-item attributes.
#[allow(dead_code)]
pub mod allocator;
#[allow(dead_code)]
pub mod cell;
pub mod file;
#[allow(dead_code)]
pub mod freelist;
#[allow(dead_code)]
pub mod fts_cell;
pub mod header;
#[allow(dead_code)]
pub mod hnsw_cell;
#[allow(dead_code)]
pub mod index_cell;
#[allow(dead_code)]
pub mod interior_page;
pub mod overflow;
pub mod page;
pub mod pager;
#[allow(dead_code)]
pub mod table_page;
#[allow(dead_code)]
pub mod varint;
#[allow(dead_code)]
pub mod wal;

use std::collections::{BTreeMap, HashMap};
use std::path::Path;
use std::sync::{Arc, Mutex};

use crate::sql::dialect::SqlriteDialect;
use sqlparser::parser::Parser;

use crate::error::{Result, SQLRiteError};
use crate::sql::db::database::Database;
use crate::sql::db::secondary_index::{IndexOrigin, SecondaryIndex};
use crate::sql::db::table::{Column, DataType, Row, Table, Value};
use crate::sql::hnsw::DistanceMetric;
use crate::sql::pager::cell::Cell;
use crate::sql::pager::header::DbHeader;
use crate::sql::pager::index_cell::IndexCell;
use crate::sql::pager::interior_page::{InteriorCell, InteriorPage};
use crate::sql::pager::overflow::{
    OVERFLOW_THRESHOLD, OverflowRef, PagedEntry, read_overflow_chain, write_overflow_chain,
};
use crate::sql::pager::page::{PAGE_HEADER_SIZE, PAGE_SIZE, PAYLOAD_PER_PAGE, PageType};
use crate::sql::pager::pager::Pager;
use crate::sql::pager::table_page::TablePage;
use crate::sql::parser::create::CreateQuery;

// Re-export so callers can spell `sql::pager::AccessMode` without
// reaching into the `pager::pager::pager` submodule path.
pub use crate::sql::pager::pager::AccessMode;

/// Name of the internal catalog table. Reserved — user CREATEs of this
/// name must be rejected upstream.
pub const MASTER_TABLE_NAME: &str = "sqlrite_master";

/// Opens a database file in read-write mode. Shorthand for
/// [`open_database_with_mode`] with [`AccessMode::ReadWrite`].
pub fn open_database(path: &Path, db_name: String) -> Result<Database> {
    open_database_with_mode(path, db_name, AccessMode::ReadWrite)
}

/// Opens a database file in read-only mode. Acquires a shared OS-level
/// advisory lock, so other read-only openers coexist but any writer is
/// excluded. Attempts to mutate the returned `Database` (e.g. an
/// `INSERT`, or a `save_database` call against it) bottom out in a
/// `cannot commit: database is opened read-only` error from the Pager.
pub fn open_database_read_only(path: &Path, db_name: String) -> Result<Database> {
    open_database_with_mode(path, db_name, AccessMode::ReadOnly)
}

/// Opens a database file and reconstructs the in-memory `Database`,
/// leaving the long-lived `Pager` attached for subsequent auto-save
/// (read-write) or consistent-snapshot reads (read-only).
pub fn open_database_with_mode(path: &Path, db_name: String, mode: AccessMode) -> Result<Database> {
    let pager = Pager::open_with_mode(path, mode)?;

    // 1. Load sqlrite_master from the tree at header.schema_root_page.
    let mut master = build_empty_master_table();
    load_table_rows(&pager, &mut master, pager.header().schema_root_page)?;

    // 2. Two passes over master rows: first build every user table, then
    //    attach secondary indexes. Indexes need their base table to exist
    //    before we can populate them. Auto-indexes are created at table
    //    build time so we only have to load explicit indexes from disk
    //    (but we also reload the auto-index CONTENT because Table::new
    //    built it empty).
    let mut db = Database::new(db_name);
    let mut index_rows: Vec<IndexCatalogRow> = Vec::new();

    for rowid in master.rowids() {
        let ty = take_text(&master, "type", rowid)?;
        let name = take_text(&master, "name", rowid)?;
        let sql = take_text(&master, "sql", rowid)?;
        let rootpage = take_integer(&master, "rootpage", rowid)? as u32;
        let last_rowid = take_integer(&master, "last_rowid", rowid)?;

        match ty.as_str() {
            "table" => {
                let (parsed_name, columns) = parse_create_sql(&sql)?;
                if parsed_name != name {
                    return Err(SQLRiteError::Internal(format!(
                        "sqlrite_master row '{name}' carries SQL for '{parsed_name}' — corrupt catalog?"
                    )));
                }
                let mut table = build_empty_table(&name, columns, last_rowid);
                if rootpage != 0 {
                    load_table_rows(&pager, &mut table, rootpage)?;
                }
                if last_rowid > table.last_rowid {
                    table.last_rowid = last_rowid;
                }
                db.tables.insert(name, table);
            }
            "index" => {
                index_rows.push(IndexCatalogRow {
                    name,
                    sql,
                    rootpage,
                });
            }
            other => {
                return Err(SQLRiteError::Internal(format!(
                    "sqlrite_master row '{name}' has unknown type '{other}'"
                )));
            }
        }
    }

    // Second pass: attach each index to its table. HNSW indexes
    // (Phase 7d.2) take a different code path because their persisted
    // form is just the CREATE INDEX SQL — the graph itself isn't
    // persisted yet (Phase 7d.3). Detect HNSW via the SQL's USING clause
    // and route to a graph-rebuild instead of the B-Tree-cell load.
    //
    // Phase 8b — same shape for FTS indexes. The posting lists aren't
    // persisted yet (Phase 8c), so we replay the CREATE INDEX SQL on
    // open and let `execute_create_index` walk current rows.
    for row in index_rows {
        if create_index_sql_uses_hnsw(&row.sql) {
            rebuild_hnsw_index(&mut db, &pager, &row)?;
        } else if create_index_sql_uses_fts(&row.sql) {
            rebuild_fts_index(&mut db, &pager, &row)?;
        } else {
            attach_index(&mut db, &pager, row)?;
        }
    }

    // Phase 11.9 — replay any MVCC commit batches recovered from
    // the WAL into the freshly-built `MvStore`, and seed the
    // `MvccClock` past the highest persisted timestamp. Without
    // this step the in-memory MVCC state would always start blank
    // on reopen — fine for legacy single-session workloads, but a
    // correctness gap once `BEGIN CONCURRENT` is in play (a
    // second process could hand out a `begin_ts` below an
    // already-committed version's `end`, breaking the visibility
    // rule).
    //
    // The clock seed is the larger of (header.clock_high_water,
    // max(commit_ts among replayed batches)) so a crash between
    // commits and the next checkpoint — where the header's
    // high-water lags reality — still produces a clock that
    // doesn't regress.
    replay_mvcc_into_db(&mut db, &pager)?;

    db.source_path = Some(path.to_path_buf());
    db.pager = Some(pager);
    Ok(db)
}

/// Phase 11.9 — drains every MVCC commit batch the Pager recovered
/// from the WAL into `db.mv_store`, and advances `db.mvcc_clock`
/// to at least the highest observed timestamp.
///
/// Batches are replayed in WAL order, which matches commit order
/// (the WAL appends sequentially). Each record's `commit_ts`
/// becomes the version's `begin`, with the previous latest
/// version's `end` capped at the same timestamp — identical to
/// the live-commit path's `MvStore::push_committed`.
fn replay_mvcc_into_db(db: &mut Database, pager: &Pager) -> Result<()> {
    use crate::mvcc::RowVersion;

    let mut clock_seed = pager.clock_high_water();
    for batch in pager.recovered_mvcc_commits() {
        if batch.commit_ts > clock_seed {
            clock_seed = batch.commit_ts;
        }
        for rec in &batch.records {
            let version = RowVersion::committed(batch.commit_ts, rec.payload.clone());
            db.mv_store
                .push_committed(rec.row.clone(), version)
                .map_err(|e| {
                    SQLRiteError::Internal(format!(
                        "WAL MVCC replay: push_committed failed for {}/{}: {e}",
                        rec.row.table, rec.row.rowid,
                    ))
                })?;
        }
    }
    if clock_seed > 0 {
        db.mvcc_clock.observe(clock_seed);
    }
    Ok(())
}

/// Catalog row for a secondary index — deferred until after every table is
/// loaded so the index's base table exists by the time we populate it.
struct IndexCatalogRow {
    name: String,
    sql: String,
    rootpage: u32,
}

/// Persists `db` to disk. Diff-pager skips writing pages whose bytes
/// haven't changed; the [`PageAllocator`] preserves per-table page
/// numbers across saves so unchanged tables produce zero dirty frames.
///
/// Pages that were live before this save but aren't restaged this round
/// (e.g., the leaves of a dropped table) move onto a persisted free
/// list rooted at `header.freelist_head`; subsequent saves draw from
/// the freelist before extending the file. `VACUUM` (see
/// [`vacuum_database`]) compacts the file by ignoring the freelist and
/// allocating linearly from page 1.
///
/// [`PageAllocator`]: crate::sql::pager::allocator::PageAllocator
pub fn save_database(db: &mut Database, path: &Path) -> Result<()> {
    save_database_with_mode(db, path, /*compact=*/ false)
}

/// Reclaims space by rewriting every live B-Tree contiguously from
/// page 1, with no freelist. Equivalent to `save_database` but ignores
/// the existing freelist and per-table preferred pools — every page is
/// allocated by extending the high-water mark — so the resulting file
/// is tightly packed and the freelist is empty.
///
/// Used by the SQL-level `VACUUM;` statement.
pub fn vacuum_database(db: &mut Database, path: &Path) -> Result<()> {
    save_database_with_mode(db, path, /*compact=*/ true)
}

/// Shared save core. `compact = false` is the normal save path (uses
/// the existing freelist + per-table preferred pools). `compact = true`
/// is the VACUUM path (empty freelist, empty preferred pools, linear
/// allocation from page 1).
fn save_database_with_mode(db: &mut Database, path: &Path, compact: bool) -> Result<()> {
    // Phase 7d.3 — rebuild any HNSW index that DELETE / UPDATE-on-vector
    // marked dirty. Done up front under the &mut Database borrow we
    // already hold, before the immutable iteration loops below need
    // their own borrow.
    rebuild_dirty_hnsw_indexes(db);
    // Phase 8b — same drill for FTS indexes flagged by DELETE / UPDATE.
    rebuild_dirty_fts_indexes(db);

    let same_path = db.source_path.as_deref() == Some(path);
    let mut pager = if same_path {
        match db.pager.take() {
            Some(p) => p,
            None if path.exists() => Pager::open(path)?,
            None => Pager::create(path)?,
        }
    } else if path.exists() {
        Pager::open(path)?
    } else {
        Pager::create(path)?
    };

    // Snapshot what was live BEFORE we reset staged. Used to compute the
    // newly-freed set after staging completes. Page 0 (the header) is
    // never on the freelist — it's always live.
    let old_header = pager.header();
    let old_live: std::collections::HashSet<u32> = (1..old_header.page_count).collect();

    // Read the previously-persisted freelist so its leaf pages can be
    // reused as preferred allocations and its trunk pages don't leak.
    let (old_free_leaves, old_free_trunks) = if compact || old_header.freelist_head == 0 {
        (Vec::new(), Vec::new())
    } else {
        crate::sql::pager::freelist::read_freelist(&pager, old_header.freelist_head)?
    };

    // Snapshot the previous rootpages of each table/index so we can
    // seed per-table preferred pools (the unchanged-table case stages
    // byte-identical pages → diff pager skips every write for it).
    let old_rootpages = if compact {
        HashMap::new()
    } else {
        read_old_rootpages(&pager, old_header.schema_root_page)?
    };

    // SQLR-1 — snapshot every prior B-Tree's page set NOW, before any
    // staging starts. `Pager::read_page` shadows on-disk bytes with the
    // current `staged` buffer, so if we deferred these walks until each
    // object's turn in the staging loop, a *new* index added in this
    // save would extend past the old high-water and overwrite the
    // pages of any later-staged object whose old root sits in that
    // range — including `sqlrite_master`, which is always staged last.
    // The follow-up walk would then read the wrong B-Tree's bytes and
    // either hand the allocator a bogus preferred pool or panic
    // dispatching cells (a table-cell decoder vs. an index leaf, the
    // shape of the original SQLR-1 panic). Walking up front pins each
    // map to the committed bytes that were on disk before this save
    // touched anything.
    let old_preferred_pages: HashMap<(String, String), Vec<u32>> = if compact {
        HashMap::new()
    } else {
        let mut map: HashMap<(String, String), Vec<u32>> = HashMap::new();
        for ((kind, name), &root) in &old_rootpages {
            // Tables can carry overflow chains; index/HNSW/FTS leaves
            // never overflow in the current encoding, so the cheaper
            // walk suffices for them.
            let follow = kind == "table";
            let pages = collect_pages_for_btree(&pager, root, follow)?;
            map.insert((kind.clone(), name.clone()), pages);
        }
        map
    };
    let old_master_pages: Vec<u32> = if compact || old_header.schema_root_page == 0 {
        Vec::new()
    } else {
        collect_pages_for_btree(
            &pager,
            old_header.schema_root_page,
            /*follow_overflow=*/ true,
        )?
    };

    pager.clear_staged();

    // Allocator: in normal mode, seed with the old freelist; in compact
    // mode, start empty so allocation extends linearly from page 1.
    use std::collections::VecDeque;
    let initial_freelist: VecDeque<u32> = if compact {
        VecDeque::new()
    } else {
        crate::sql::pager::freelist::freelist_to_deque(old_free_leaves.clone())
    };
    let mut alloc = crate::sql::pager::allocator::PageAllocator::new(initial_freelist, 1);

    // 1. Stage each user table's B-Tree, collecting master-row info.
    //    `kind` is "table" or "index" — master has one row per each.
    let mut master_rows: Vec<CatalogEntry> = Vec::new();

    let mut table_names: Vec<&String> = db.tables.keys().collect();
    table_names.sort();
    for name in table_names {
        if name == MASTER_TABLE_NAME {
            return Err(SQLRiteError::Internal(format!(
                "user table cannot be named '{MASTER_TABLE_NAME}' (reserved)"
            )));
        }
        if !compact {
            if let Some(prev) = old_preferred_pages.get(&("table".to_string(), name.to_string())) {
                alloc.set_preferred(prev.clone());
            }
        }
        let table = &db.tables[name];
        let rootpage = stage_table_btree(&mut pager, table, &mut alloc)?;
        alloc.finish_preferred();
        master_rows.push(CatalogEntry {
            kind: "table".into(),
            name: name.clone(),
            sql: table_to_create_sql(table),
            rootpage,
            last_rowid: table.last_rowid,
        });
    }

    // 2. Stage each secondary index's B-Tree. Indexes persist in a
    //    deterministic order: sorted by (owning_table, index_name).
    let mut index_entries: Vec<(&Table, &SecondaryIndex)> = Vec::new();
    for table in db.tables.values() {
        for idx in &table.secondary_indexes {
            index_entries.push((table, idx));
        }
    }
    index_entries
        .sort_by(|(ta, ia), (tb, ib)| ta.tb_name.cmp(&tb.tb_name).then(ia.name.cmp(&ib.name)));
    for (_table, idx) in index_entries {
        if !compact {
            if let Some(prev) =
                old_preferred_pages.get(&("index".to_string(), idx.name.to_string()))
            {
                alloc.set_preferred(prev.clone());
            }
        }
        let rootpage = stage_index_btree(&mut pager, idx, &mut alloc)?;
        alloc.finish_preferred();
        master_rows.push(CatalogEntry {
            kind: "index".into(),
            name: idx.name.clone(),
            sql: idx.synthesized_sql(),
            rootpage,
            last_rowid: 0,
        });
    }

    // 2b. Phase 7d.3: persist HNSW indexes as their own cell-encoded
    //     page trees, with the rootpage recorded in sqlrite_master.
    //     Reopen loads the graph back from cells (fast, exact match)
    //     instead of rebuilding from rows.
    //
    //     Dirty indexes (set by DELETE / UPDATE-on-vector-col) are
    //     rebuilt from current rows BEFORE staging, so the on-disk
    //     graph reflects the current row set.
    let mut hnsw_entries: Vec<(&Table, &crate::sql::db::table::HnswIndexEntry)> = Vec::new();
    for table in db.tables.values() {
        for entry in &table.hnsw_indexes {
            hnsw_entries.push((table, entry));
        }
    }
    hnsw_entries
        .sort_by(|(ta, ea), (tb, eb)| ta.tb_name.cmp(&tb.tb_name).then(ea.name.cmp(&eb.name)));
    for (table, entry) in hnsw_entries {
        if !compact {
            if let Some(prev) =
                old_preferred_pages.get(&("index".to_string(), entry.name.to_string()))
            {
                alloc.set_preferred(prev.clone());
            }
        }
        let rootpage = stage_hnsw_btree(&mut pager, &entry.index, &mut alloc)?;
        alloc.finish_preferred();
        master_rows.push(CatalogEntry {
            kind: "index".into(),
            name: entry.name.clone(),
            sql: synthesize_hnsw_create_index_sql(
                &entry.name,
                &table.tb_name,
                &entry.column_name,
                entry.metric,
            ),
            rootpage,
            last_rowid: 0,
        });
    }

    // 2c. Phase 8c — persist FTS posting lists as their own
    //     cell-encoded page trees, with the rootpage recorded in
    //     sqlrite_master. Reopen loads the postings back from cells
    //     (fast, exact match) instead of re-tokenizing rows.
    //
    //     Dirty indexes (set by DELETE / UPDATE-on-text-col) are
    //     rebuilt from current rows BEFORE staging by
    //     `rebuild_dirty_fts_indexes`, so the on-disk tree reflects
    //     the current row set.
    let mut fts_entries: Vec<(&Table, &crate::sql::db::table::FtsIndexEntry)> = Vec::new();
    for table in db.tables.values() {
        for entry in &table.fts_indexes {
            fts_entries.push((table, entry));
        }
    }
    fts_entries
        .sort_by(|(ta, ea), (tb, eb)| ta.tb_name.cmp(&tb.tb_name).then(ea.name.cmp(&eb.name)));
    let any_fts = !fts_entries.is_empty();
    for (table, entry) in fts_entries {
        if !compact {
            if let Some(prev) =
                old_preferred_pages.get(&("index".to_string(), entry.name.to_string()))
            {
                alloc.set_preferred(prev.clone());
            }
        }
        let rootpage = stage_fts_btree(&mut pager, &entry.index, &mut alloc)?;
        alloc.finish_preferred();
        master_rows.push(CatalogEntry {
            kind: "index".into(),
            name: entry.name.clone(),
            sql: format!(
                "CREATE INDEX {} ON {} USING fts ({})",
                entry.name, table.tb_name, entry.column_name
            ),
            rootpage,
            last_rowid: 0,
        });
    }

    // 3. Build an in-memory sqlrite_master with one row per table or index,
    //    then stage it via the same tree-build path. Seed master's
    //    preferred pool with the previous master tree's pages so the
    //    catalog page numbers stay stable across saves whenever the
    //    catalog content didn't change.
    let mut master = build_empty_master_table();
    for (i, entry) in master_rows.into_iter().enumerate() {
        let rowid = (i as i64) + 1;
        master.restore_row(
            rowid,
            vec![
                Some(Value::Text(entry.kind)),
                Some(Value::Text(entry.name)),
                Some(Value::Text(entry.sql)),
                Some(Value::Integer(entry.rootpage as i64)),
                Some(Value::Integer(entry.last_rowid)),
            ],
        )?;
    }
    if !compact && !old_master_pages.is_empty() {
        // Use the page list snapshotted before any staging touched
        // disk; re-walking here would read whatever a new index
        // already restaged on top of master's old root (SQLR-1).
        alloc.set_preferred(old_master_pages.clone());
    }
    let master_root = stage_table_btree(&mut pager, &master, &mut alloc)?;
    alloc.finish_preferred();

    // 4. Compute newly-freed pages: the previously-live set minus what
    //    we just restaged. The previous freelist's trunk pages get
    //    re-encoded too — they're in `old_live`, weren't restaged, so
    //    the filter naturally moves them to the new freelist.
    //
    // In `compact` mode (VACUUM), we *discard* newly_freed instead of
    // routing it onto the new freelist. The whole point of VACUUM is
    // to let the file truncate to the new high-water mark, so any page
    // past it gets dropped at the next checkpoint.
    if !compact {
        let used = alloc.used().clone();
        let mut newly_freed: Vec<u32> = old_live
            .iter()
            .copied()
            .filter(|p| !used.contains(p))
            .collect();
        let _ = &old_free_trunks; // silenced — handled by the old_live filter
        alloc.add_to_freelist(newly_freed.drain(..));
    }

    // 5. Encode the new freelist into trunk pages. `stage_freelist`
    //    consumes some of the free pages AS the trunk pages themselves —
    //    a trunk is just a free page borrowed for metadata. Pages that
    //    were on the freelist but become trunks no longer need to be
    //    "extension" pages; the high-water mark from the staging loop
    //    above is already correct.
    let new_free_pages = alloc.drain_freelist();
    let new_freelist_head =
        crate::sql::pager::freelist::stage_freelist(&mut pager, new_free_pages)?;

    // 6. Pick the format version. v6 is on demand: only bumps when the
    //    new freelist is non-empty. FTS-bearing files keep their v5
    //    promotion; v6 is a strict superset (v6 readers handle v4/v5/v6).
    use crate::sql::pager::header::{FORMAT_VERSION_V5, FORMAT_VERSION_V6};
    let format_version = if new_freelist_head != 0 {
        FORMAT_VERSION_V6
    } else if any_fts {
        // Preserve a v6 file at v6 (don't downgrade) but otherwise
        // bump v4 → v5 for FTS like Phase 8c does.
        std::cmp::max(FORMAT_VERSION_V5, old_header.format_version)
    } else {
        // Preserve whatever the file already was.
        old_header.format_version
    };

    pager.commit(DbHeader {
        page_count: alloc.high_water(),
        schema_root_page: master_root,
        format_version,
        freelist_head: new_freelist_head,
    })?;

    if same_path {
        db.pager = Some(pager);
    }
    Ok(())
}

/// Build material for a single row in sqlrite_master.
struct CatalogEntry {
    kind: String, // "table" or "index"
    name: String,
    sql: String,
    rootpage: u32,
    last_rowid: i64,
}

// -------------------------------------------------------------------------
// sqlrite_master — hardcoded catalog table schema

fn build_empty_master_table() -> Table {
    // Phase 3e: `type` is the first column, matching SQLite's convention.
    // It distinguishes `'table'` rows from `'index'` rows.
    let columns = vec![
        Column::new("type".into(), "text".into(), false, true, false),
        Column::new("name".into(), "text".into(), true, true, true),
        Column::new("sql".into(), "text".into(), false, true, false),
        Column::new("rootpage".into(), "integer".into(), false, true, false),
        Column::new("last_rowid".into(), "integer".into(), false, true, false),
    ];
    build_empty_table(MASTER_TABLE_NAME, columns, 0)
}

/// Reads a required Text column from a known-good catalog row.
fn take_text(table: &Table, col: &str, rowid: i64) -> Result<String> {
    match table.get_value(col, rowid) {
        Some(Value::Text(s)) => Ok(s),
        other => Err(SQLRiteError::Internal(format!(
            "sqlrite_master column '{col}' at rowid {rowid}: expected Text, got {other:?}"
        ))),
    }
}

/// Reads a required Integer column from a known-good catalog row.
fn take_integer(table: &Table, col: &str, rowid: i64) -> Result<i64> {
    match table.get_value(col, rowid) {
        Some(Value::Integer(v)) => Ok(v),
        other => Err(SQLRiteError::Internal(format!(
            "sqlrite_master column '{col}' at rowid {rowid}: expected Integer, got {other:?}"
        ))),
    }
}

// -------------------------------------------------------------------------
// CREATE-TABLE SQL synthesis and re-parsing

/// Synthesizes a CREATE TABLE SQL string that recreates the table's schema.
/// Deterministic: same schema → same SQL, so diffing commits stay stable.
fn table_to_create_sql(table: &Table) -> String {
    let mut parts = Vec::with_capacity(table.columns.len());
    for c in &table.columns {
        // Render the SQL type literally so the round-trip through
        // CREATE TABLE re-parsing recreates the same schema. Vector
        // carries its dimension inline.
        let ty: String = match &c.datatype {
            DataType::Integer => "INTEGER".to_string(),
            DataType::Text => "TEXT".to_string(),
            DataType::Real => "REAL".to_string(),
            DataType::Bool => "BOOLEAN".to_string(),
            DataType::Vector(dim) => format!("VECTOR({dim})"),
            DataType::Json => "JSON".to_string(),
            DataType::None | DataType::Invalid => "TEXT".to_string(),
        };
        let mut piece = format!("{} {}", c.column_name, ty);
        if c.is_pk {
            piece.push_str(" PRIMARY KEY");
        } else {
            if c.is_unique {
                piece.push_str(" UNIQUE");
            }
            if c.not_null {
                piece.push_str(" NOT NULL");
            }
        }
        if let Some(default) = &c.default {
            piece.push_str(" DEFAULT ");
            piece.push_str(&render_default_literal(default));
        }
        parts.push(piece);
    }
    format!("CREATE TABLE {} ({});", table.tb_name, parts.join(", "))
}

/// Renders a DEFAULT value back to SQL-literal form so the synthesized
/// CREATE TABLE round-trips through `parse_create_sql`. Text values get
/// single-quoted with single-quote doubling for escaping. Vector defaults
/// are not currently expressible at CREATE TABLE time, so we render them
/// as their bracket-array form (matches the INSERT literal grammar).
fn render_default_literal(value: &Value) -> String {
    match value {
        Value::Integer(i) => i.to_string(),
        Value::Real(f) => f.to_string(),
        Value::Bool(b) => {
            if *b {
                "TRUE".to_string()
            } else {
                "FALSE".to_string()
            }
        }
        Value::Text(s) => format!("'{}'", s.replace('\'', "''")),
        Value::Null => "NULL".to_string(),
        Value::Vector(_) => value.to_display_string(),
    }
}

/// Reverses `table_to_create_sql`: feeds the SQL back through `sqlparser`
/// and produces our internal column list. Returns `(table_name, columns)`.
fn parse_create_sql(sql: &str) -> Result<(String, Vec<Column>)> {
    let dialect = SqlriteDialect::new();
    let mut ast = Parser::parse_sql(&dialect, sql).map_err(SQLRiteError::from)?;
    let stmt = ast.pop().ok_or_else(|| {
        SQLRiteError::Internal("sqlrite_master row held an empty SQL string".to_string())
    })?;
    let create = CreateQuery::new(&stmt)?;
    let columns = create
        .columns
        .into_iter()
        .map(|pc| {
            Column::with_default(
                pc.name,
                pc.datatype,
                pc.is_pk,
                pc.not_null,
                pc.is_unique,
                pc.default,
            )
        })
        .collect();
    Ok((create.table_name, columns))
}

// -------------------------------------------------------------------------
// In-memory table (re)construction

/// Builds an empty in-memory `Table` given the declared columns.
fn build_empty_table(name: &str, columns: Vec<Column>, last_rowid: i64) -> Table {
    let rows: Arc<Mutex<HashMap<String, Row>>> = Arc::new(Mutex::new(HashMap::new()));
    let mut secondary_indexes: Vec<SecondaryIndex> = Vec::new();
    {
        let mut map = rows.lock().expect("rows mutex poisoned");
        for col in &columns {
            // Mirror the dispatch in `Table::new` so the reconstructed
            // table has the same shape it'd have if it were built fresh
            // from SQL. Phase 7a adds the Vector arm — without it,
            // VECTOR columns silently restore as Row::None and every
            // restore_row hits a "storage None vs value Some(Vector(...))"
            // type mismatch.
            let row = match &col.datatype {
                DataType::Integer => Row::Integer(BTreeMap::new()),
                DataType::Text => Row::Text(BTreeMap::new()),
                DataType::Real => Row::Real(BTreeMap::new()),
                DataType::Bool => Row::Bool(BTreeMap::new()),
                DataType::Vector(_dim) => Row::Vector(BTreeMap::new()),
                // JSON columns reuse Text storage — see Table::new and
                // Phase 7e's scope-correction note.
                DataType::Json => Row::Text(BTreeMap::new()),
                DataType::None | DataType::Invalid => Row::None,
            };
            map.insert(col.column_name.clone(), row);

            // Auto-create UNIQUE/PK indexes so the restored table has the
            // same shape Table::new would have built from fresh SQL.
            if (col.is_pk || col.is_unique)
                && matches!(col.datatype, DataType::Integer | DataType::Text)
            {
                if let Ok(idx) = SecondaryIndex::new(
                    SecondaryIndex::auto_name(name, &col.column_name),
                    name.to_string(),
                    col.column_name.clone(),
                    &col.datatype,
                    true,
                    IndexOrigin::Auto,
                ) {
                    secondary_indexes.push(idx);
                }
            }
        }
    }

    let primary_key = columns
        .iter()
        .find(|c| c.is_pk)
        .map(|c| c.column_name.clone())
        .unwrap_or_else(|| "-1".to_string());

    Table {
        tb_name: name.to_string(),
        columns,
        rows,
        secondary_indexes,
        // HNSW indexes (Phase 7d.2) are reconstructed on open by re-
        // executing each `CREATE INDEX … USING hnsw` SQL stored in
        // `sqlrite_master`. This builder produces the empty shell;
        // `replay_create_index_for_hnsw` (in this same module) walks
        // sqlrite_master after every table is loaded and rebuilds the
        // graph from current row data. Persistence of the graph itself
        // (avoiding the on-open rebuild cost) is Phase 7d.3.
        hnsw_indexes: Vec::new(),
        // FTS indexes (Phase 8b) follow the same pattern — the
        // CREATE INDEX … USING fts SQL is the source of truth on open
        // and the in-memory posting list gets rebuilt from current
        // rows. Cell-encoded persistence of the postings is Phase 8c.
        fts_indexes: Vec::new(),
        last_rowid,
        primary_key,
    }
}

// -------------------------------------------------------------------------
// Leaf-chain read / write

/// Walks a table's B-Tree from `root_page`, following the leftmost-child
/// chain down to the first leaf, then iterating leaves via their sibling
/// `next_page` pointers. Every cell is decoded and replayed into `table`.
///
/// Open-path note: we eagerly materialize the entire table into `Table`'s
/// in-memory maps. Phase 5 will introduce a `Cursor` that hits the pager
/// on demand so queries can stream through the tree without a full upfront
/// load.
/// Re-parses `CREATE INDEX` SQL from sqlrite_master and restores the
/// index on its base table by walking the tree of index cells at
/// `rootpage`. The base table is expected to already be in `db.tables`.
fn attach_index(db: &mut Database, pager: &Pager, row: IndexCatalogRow) -> Result<()> {
    let (table_name, column_name, is_unique) = parse_create_index_sql(&row.sql)?;

    let table = db.get_table_mut(table_name.clone()).map_err(|_| {
        SQLRiteError::Internal(format!(
            "index '{}' references unknown table '{table_name}' (sqlrite_master out of sync?)",
            row.name
        ))
    })?;
    let datatype = table
        .columns
        .iter()
        .find(|c| c.column_name == column_name)
        .map(|c| clone_datatype(&c.datatype))
        .ok_or_else(|| {
            SQLRiteError::Internal(format!(
                "index '{}' references unknown column '{column_name}' on '{table_name}'",
                row.name
            ))
        })?;

    // An auto-index on this column may already exist (built by
    // build_empty_table for UNIQUE/PK columns). If the names match, reuse
    // the slot instead of adding a duplicate entry.
    let existing_slot = table
        .secondary_indexes
        .iter()
        .position(|i| i.name == row.name);
    let idx = match existing_slot {
        Some(i) => {
            // Drain any entries that may have been populated during table
            // restore_row calls — we're about to repopulate from the
            // persisted tree.
            table.secondary_indexes.remove(i)
        }
        None => SecondaryIndex::new(
            row.name.clone(),
            table_name.clone(),
            column_name.clone(),
            &datatype,
            is_unique,
            IndexOrigin::Explicit,
        )?,
    };
    let mut idx = idx;
    // Wipe any stale entries from the auto path so the load is idempotent.
    let is_unique_flag = idx.is_unique;
    let origin = idx.origin;
    idx = SecondaryIndex::new(
        idx.name,
        idx.table_name,
        idx.column_name,
        &datatype,
        is_unique_flag,
        origin,
    )?;

    // Populate from the index tree's cells.
    load_index_rows(pager, &mut idx, row.rootpage)?;

    table.secondary_indexes.push(idx);
    Ok(())
}

/// Walks the leaves of an index B-Tree rooted at `root_page` and inserts
/// every `(value, rowid)` pair into `idx`.
fn load_index_rows(pager: &Pager, idx: &mut SecondaryIndex, root_page: u32) -> Result<()> {
    if root_page == 0 {
        return Ok(());
    }
    let first_leaf = find_leftmost_leaf(pager, root_page)?;
    let mut current = first_leaf;
    while current != 0 {
        let page_buf = pager
            .read_page(current)
            .ok_or_else(|| SQLRiteError::Internal(format!("missing index leaf page {current}")))?;
        if page_buf[0] != PageType::TableLeaf as u8 {
            return Err(SQLRiteError::Internal(format!(
                "page {current} tagged {} but expected TableLeaf (index)",
                page_buf[0]
            )));
        }
        let next_leaf = u32::from_le_bytes(page_buf[1..5].try_into().unwrap());
        let payload: &[u8; PAYLOAD_PER_PAGE] = (&page_buf[PAGE_HEADER_SIZE..])
            .try_into()
            .map_err(|_| SQLRiteError::Internal("index leaf payload size".to_string()))?;
        let leaf = TablePage::from_bytes(payload);

        for slot in 0..leaf.slot_count() {
            // Slots on an index page hold KIND_INDEX cells; decode directly.
            let offset = leaf.slot_offset_raw(slot)?;
            let (ic, _) = IndexCell::decode(leaf.as_bytes(), offset)?;
            idx.insert(&ic.value, ic.rowid)?;
        }
        current = next_leaf;
    }
    Ok(())
}

/// Minimal recognizer for the synthesized-or-user `CREATE INDEX` SQL we
/// store in sqlrite_master. Returns `(table_name, column_name, is_unique)`.
///
/// Uses sqlparser so user-supplied SQL with extra whitespace, case, etc.
/// still works; the only shape we accept is single-column indexes.
fn parse_create_index_sql(sql: &str) -> Result<(String, String, bool)> {
    use sqlparser::ast::{CreateIndex, Expr, Statement};

    let dialect = SqlriteDialect::new();
    let mut ast = Parser::parse_sql(&dialect, sql).map_err(SQLRiteError::from)?;
    let Some(Statement::CreateIndex(CreateIndex {
        table_name,
        columns,
        unique,
        ..
    })) = ast.pop()
    else {
        return Err(SQLRiteError::Internal(format!(
            "sqlrite_master index row's SQL isn't a CREATE INDEX: {sql}"
        )));
    };
    if columns.len() != 1 {
        return Err(SQLRiteError::NotImplemented(
            "multi-column indexes aren't supported yet".to_string(),
        ));
    }
    let col = match &columns[0].column.expr {
        Expr::Identifier(ident) => ident.value.clone(),
        Expr::CompoundIdentifier(parts) => {
            parts.last().map(|p| p.value.clone()).unwrap_or_default()
        }
        other => {
            return Err(SQLRiteError::Internal(format!(
                "unsupported indexed column expression: {other:?}"
            )));
        }
    };
    Ok((table_name.to_string(), col, unique))
}

/// True iff a CREATE INDEX SQL string uses `USING hnsw` (case-insensitive).
/// Used by the open path to route HNSW indexes to the graph-rebuild path
/// instead of the standard B-Tree cell-load. Pre-Phase-7d.2 indexes
/// don't have a USING clause, so they all return false and continue
/// taking the existing path.
fn create_index_sql_uses_hnsw(sql: &str) -> bool {
    use sqlparser::ast::{CreateIndex, IndexType, Statement};

    let dialect = SqlriteDialect::new();
    let Ok(mut ast) = Parser::parse_sql(&dialect, sql) else {
        return false;
    };
    let Some(Statement::CreateIndex(CreateIndex { using, .. })) = ast.pop() else {
        return false;
    };
    matches!(using, Some(IndexType::Custom(ident)) if ident.value.eq_ignore_ascii_case("hnsw"))
}

/// Phase 8b — peeks at a CREATE INDEX SQL to detect `USING fts(...)`.
/// Mirrors [`create_index_sql_uses_hnsw`].
fn create_index_sql_uses_fts(sql: &str) -> bool {
    use sqlparser::ast::{CreateIndex, IndexType, Statement};

    let dialect = SqlriteDialect::new();
    let Ok(mut ast) = Parser::parse_sql(&dialect, sql) else {
        return false;
    };
    let Some(Statement::CreateIndex(CreateIndex { using, .. })) = ast.pop() else {
        return false;
    };
    matches!(using, Some(IndexType::Custom(ident)) if ident.value.eq_ignore_ascii_case("fts"))
}

/// Phase 8c — loads (or rebuilds) an FTS index on database open. Two
/// paths mirror [`rebuild_hnsw_index`]:
///
///   - **rootpage != 0** (Phase 8c default): the posting list is
///     persisted as cell-encoded pages. Read every cell directly via
///     [`load_fts_postings`] and reconstruct the index — no
///     re-tokenization, exact bit-for-bit reproduction.
///
///   - **rootpage == 0** (compatibility): no on-disk postings, e.g.
///     for files saved by Phase 8b before persistence landed. Replay
///     the CREATE INDEX SQL through `execute_create_index`, which
///     walks the table's current rows and tokenizes them fresh.
fn rebuild_fts_index(db: &mut Database, pager: &Pager, row: &IndexCatalogRow) -> Result<()> {
    use crate::sql::db::table::FtsIndexEntry;
    use crate::sql::executor::execute_create_index;
    use crate::sql::fts::PostingList;
    use sqlparser::ast::Statement;

    let dialect = SqlriteDialect::new();
    let mut ast = Parser::parse_sql(&dialect, &row.sql).map_err(SQLRiteError::from)?;
    let Some(stmt @ Statement::CreateIndex(_)) = ast.pop() else {
        return Err(SQLRiteError::Internal(format!(
            "sqlrite_master FTS row's SQL isn't a CREATE INDEX: {}",
            row.sql
        )));
    };

    if row.rootpage == 0 {
        // Compatibility path — no persisted postings; replay rows.
        execute_create_index(&stmt, db)?;
        return Ok(());
    }

    let (doc_lengths, postings) = load_fts_postings(pager, row.rootpage)?;
    let index = PostingList::from_persisted_postings(doc_lengths, postings);
    let (tbl_name, col_name) = parse_fts_create_index_sql(&row.sql)?;
    let table_mut = db.get_table_mut(tbl_name.clone()).map_err(|_| {
        SQLRiteError::Internal(format!(
            "FTS index '{}' references unknown table '{tbl_name}'",
            row.name
        ))
    })?;
    table_mut.fts_indexes.push(FtsIndexEntry {
        name: row.name.clone(),
        column_name: col_name,
        index,
        needs_rebuild: false,
    });
    Ok(())
}

/// Pulls (table_name, column_name) out of a `CREATE INDEX … USING fts(col)`
/// SQL string. Same shape as `parse_hnsw_create_index_sql`.
fn parse_fts_create_index_sql(sql: &str) -> Result<(String, String)> {
    use sqlparser::ast::{CreateIndex, Expr, Statement};

    let dialect = SqlriteDialect::new();
    let mut ast = Parser::parse_sql(&dialect, sql).map_err(SQLRiteError::from)?;
    let Some(Statement::CreateIndex(CreateIndex {
        table_name,
        columns,
        ..
    })) = ast.pop()
    else {
        return Err(SQLRiteError::Internal(format!(
            "sqlrite_master FTS row's SQL isn't a CREATE INDEX: {sql}"
        )));
    };
    if columns.len() != 1 {
        return Err(SQLRiteError::NotImplemented(
            "multi-column FTS indexes aren't supported yet".to_string(),
        ));
    }
    let col = match &columns[0].column.expr {
        Expr::Identifier(ident) => ident.value.clone(),
        Expr::CompoundIdentifier(parts) => {
            parts.last().map(|p| p.value.clone()).unwrap_or_default()
        }
        other => {
            return Err(SQLRiteError::Internal(format!(
                "FTS CREATE INDEX has unexpected column expr: {other:?}"
            )));
        }
    };
    Ok((table_name.to_string(), col))
}

/// Loads (or rebuilds) an HNSW index on database open. Two paths:
///
///   - **rootpage != 0** (Phase 7d.3 default): the graph is persisted
///     as cell-encoded pages. Read every node directly via
///     `load_hnsw_nodes` and reconstruct the index — fast, zero
///     algorithm runs, exact bit-for-bit reproduction of what was saved.
///
///   - **rootpage == 0** (compatibility): no on-disk graph, e.g. for
///     files saved by Phase 7d.2 before persistence landed. Replay the
///     CREATE INDEX SQL through `execute_create_index`, which walks the
///     table's current rows and populates a fresh graph. Slower but
///     correctness-equivalent on the first save with the new code.
fn rebuild_hnsw_index(db: &mut Database, pager: &Pager, row: &IndexCatalogRow) -> Result<()> {
    use crate::sql::db::table::HnswIndexEntry;
    use crate::sql::executor::execute_create_index;
    use crate::sql::hnsw::HnswIndex;
    use sqlparser::ast::Statement;

    let dialect = SqlriteDialect::new();
    let mut ast = Parser::parse_sql(&dialect, &row.sql).map_err(SQLRiteError::from)?;
    let Some(stmt @ Statement::CreateIndex(_)) = ast.pop() else {
        return Err(SQLRiteError::Internal(format!(
            "sqlrite_master HNSW row's SQL isn't a CREATE INDEX: {}",
            row.sql
        )));
    };

    if row.rootpage == 0 {
        // Compatibility path — no persisted graph; walk current rows.
        execute_create_index(&stmt, db)?;
        return Ok(());
    }

    // Persistence path — read the cell tree, deserialize. The metric
    // travels through the synthesized CREATE INDEX SQL stored in
    // `sqlrite_master`; pre-SQLR-28 rows omit the WITH clause and
    // decode as L2, which matches what those graphs were built with.
    let (tbl_name, col_name, metric) = parse_hnsw_create_index_sql(&row.sql)?;
    let nodes = load_hnsw_nodes(pager, row.rootpage)?;
    let index = HnswIndex::from_persisted_nodes(metric, 0xC0FFEE, nodes);

    // Parse the CREATE INDEX to know which table + column to attach to
    // — same shape as the row-walk path; we just don't execute it.
    let table_mut = db.get_table_mut(tbl_name.clone()).map_err(|_| {
        SQLRiteError::Internal(format!(
            "HNSW index '{}' references unknown table '{tbl_name}'",
            row.name
        ))
    })?;
    table_mut.hnsw_indexes.push(HnswIndexEntry {
        name: row.name.clone(),
        column_name: col_name,
        metric,
        index,
        needs_rebuild: false,
    });
    Ok(())
}

/// Phase 7d.3 — Phase-7d.3-side helper: walk every leaf in the HNSW
/// page tree at `root_page` and decode each cell as a node. Returns
/// the (node_id, layers) tuples in slot-order (already ascending by
/// node_id since they were staged that way). The caller hands them to
/// `HnswIndex::from_persisted_nodes`.
fn load_hnsw_nodes(pager: &Pager, root_page: u32) -> Result<Vec<(i64, Vec<Vec<i64>>)>> {
    use crate::sql::pager::hnsw_cell::HnswNodeCell;

    let mut nodes: Vec<(i64, Vec<Vec<i64>>)> = Vec::new();
    let first_leaf = find_leftmost_leaf(pager, root_page)?;
    let mut current = first_leaf;
    while current != 0 {
        let page_buf = pager
            .read_page(current)
            .ok_or_else(|| SQLRiteError::Internal(format!("missing HNSW leaf page {current}")))?;
        if page_buf[0] != PageType::TableLeaf as u8 {
            return Err(SQLRiteError::Internal(format!(
                "page {current} tagged {} but expected TableLeaf (HNSW)",
                page_buf[0]
            )));
        }
        let next_leaf = u32::from_le_bytes(page_buf[1..5].try_into().unwrap());
        let payload: &[u8; PAYLOAD_PER_PAGE] = (&page_buf[PAGE_HEADER_SIZE..])
            .try_into()
            .map_err(|_| SQLRiteError::Internal("HNSW leaf payload size".to_string()))?;
        let leaf = TablePage::from_bytes(payload);
        for slot in 0..leaf.slot_count() {
            let offset = leaf.slot_offset_raw(slot)?;
            let (cell, _) = HnswNodeCell::decode(leaf.as_bytes(), offset)?;
            nodes.push((cell.node_id, cell.layers));
        }
        current = next_leaf;
    }
    Ok(nodes)
}

/// Pulls `(table_name, column_name, metric)` out of a CREATE INDEX
/// SQL string of the form `CREATE INDEX … USING hnsw (col) [WITH
/// (metric = '<m>')]`. Used by the persistence path on open to know
/// where to attach the loaded graph and which distance metric to
/// rebuild it under. Pre-SQLR-28 rows omit the WITH clause and
/// default to L2.
fn parse_hnsw_create_index_sql(sql: &str) -> Result<(String, String, DistanceMetric)> {
    use crate::sql::hnsw::DistanceMetric;
    use sqlparser::ast::{BinaryOperator, CreateIndex, Expr, Statement, Value as AstValue};

    let dialect = SqlriteDialect::new();
    let mut ast = Parser::parse_sql(&dialect, sql).map_err(SQLRiteError::from)?;
    let Some(Statement::CreateIndex(CreateIndex {
        table_name,
        columns,
        with,
        ..
    })) = ast.pop()
    else {
        return Err(SQLRiteError::Internal(format!(
            "sqlrite_master HNSW row's SQL isn't a CREATE INDEX: {sql}"
        )));
    };
    if columns.len() != 1 {
        return Err(SQLRiteError::NotImplemented(
            "multi-column HNSW indexes aren't supported yet".to_string(),
        ));
    }
    let col = match &columns[0].column.expr {
        Expr::Identifier(ident) => ident.value.clone(),
        Expr::CompoundIdentifier(parts) => {
            parts.last().map(|p| p.value.clone()).unwrap_or_default()
        }
        other => {
            return Err(SQLRiteError::Internal(format!(
                "unsupported HNSW indexed column expression: {other:?}"
            )));
        }
    };

    // Pull the metric off the parsed WITH (...) bag. The user-facing
    // CREATE INDEX path validates this in detail (typo'd metric names,
    // unknown keys, etc.); here on the persistence read-path we trust
    // what we previously wrote and surface a clean Internal error if
    // it ever doesn't match.
    let mut metric = DistanceMetric::L2;
    for opt in &with {
        if let Expr::BinaryOp { left, op, right } = opt {
            if matches!(op, BinaryOperator::Eq) {
                if let (Expr::Identifier(key), Expr::Value(v)) = (left.as_ref(), right.as_ref())
                    && key.value.eq_ignore_ascii_case("metric")
                {
                    if let AstValue::SingleQuotedString(s) | AstValue::DoubleQuotedString(s) =
                        &v.value
                    {
                        metric = DistanceMetric::from_sql_name(s).ok_or_else(|| {
                            SQLRiteError::Internal(format!(
                                "sqlrite_master HNSW row carries unknown metric '{s}'"
                            ))
                        })?;
                    }
                }
            }
        }
    }

    Ok((table_name.to_string(), col, metric))
}

/// Phase 7d.3 — rebuilds in-place any HnswIndexEntry whose
/// `needs_rebuild` flag is set (DELETE / UPDATE-on-vector marked it).
/// Walks the table's current Vec<f32> column storage and runs the
/// HNSW algorithm fresh. Called at the top of `save_database` before
/// any immutable borrows of `db` start.
///
/// Cost: O(N · ef_construction · log N) per dirty index. Fine for
/// small tables, expensive for ≥100k-row tables — matches the
/// trade-off SQLite makes for FTS5: dirtying-and-rebuilding is the
/// MVP, more sophisticated incremental delete strategies (soft-delete
/// + tombstones, neighbor reconnection) are future polish.
fn rebuild_dirty_hnsw_indexes(db: &mut Database) {
    use crate::sql::hnsw::HnswIndex;

    for table in db.tables.values_mut() {
        // Snapshot which (index_name, column, metric) triples need
        // rebuilding, before we go grabbing column data — keeps the
        // borrow structure simple. The per-entry metric matters here:
        // rebuilding a cosine-built graph as L2 (or vice versa) would
        // silently corrupt the topology and break the SQLR-28 probe.
        let dirty: Vec<(String, String, DistanceMetric)> = table
            .hnsw_indexes
            .iter()
            .filter(|e| e.needs_rebuild)
            .map(|e| (e.name.clone(), e.column_name.clone(), e.metric))
            .collect();
        if dirty.is_empty() {
            continue;
        }

        for (idx_name, col_name, metric) in dirty {
            // Snapshot every (rowid, vec) for this column.
            let mut vectors: Vec<(i64, Vec<f32>)> = Vec::new();
            {
                let row_data = table.rows.lock().expect("rows mutex poisoned");
                if let Some(Row::Vector(map)) = row_data.get(&col_name) {
                    for (id, v) in map.iter() {
                        vectors.push((*id, v.clone()));
                    }
                }
            }
            // Pre-build a HashMap for the get_vec closure so we don't
            // pay O(N) lookup per insert call.
            let snapshot: std::collections::HashMap<i64, Vec<f32>> =
                vectors.iter().cloned().collect();

            let mut new_idx = HnswIndex::new(metric, 0xC0FFEE);
            // Sort by id so the rebuild is deterministic across runs.
            vectors.sort_by_key(|(id, _)| *id);
            for (id, v) in &vectors {
                new_idx.insert(*id, v, |q| snapshot.get(&q).cloned().unwrap_or_default());
            }

            // Replace the entry's index + clear the dirty flag.
            if let Some(entry) = table.hnsw_indexes.iter_mut().find(|e| e.name == idx_name) {
                entry.index = new_idx;
                entry.needs_rebuild = false;
            }
        }
    }
}

/// Synthesises the CREATE INDEX SQL stored back into `sqlrite_master`
/// for an HNSW index. The metric travels through the SQL via an
/// optional `WITH (metric = '<m>')` clause; L2 indexes omit the clause
/// for byte-identical round-trip with pre-SQLR-28 catalogs.
fn synthesize_hnsw_create_index_sql(
    index_name: &str,
    table_name: &str,
    column_name: &str,
    metric: DistanceMetric,
) -> String {
    if matches!(metric, DistanceMetric::L2) {
        format!("CREATE INDEX {index_name} ON {table_name} USING hnsw ({column_name})")
    } else {
        format!(
            "CREATE INDEX {index_name} ON {table_name} USING hnsw ({column_name}) WITH (metric = '{}')",
            metric.sql_name()
        )
    }
}

/// Phase 8b — rebuild every FTS index a DELETE / UPDATE-on-text-col
/// marked dirty. Mirrors [`rebuild_dirty_hnsw_indexes`]; runs at save
/// time under `&mut Database`. Cheap on a clean DB (the `dirty` snapshot
/// is empty so the per-table loop short-circuits).
fn rebuild_dirty_fts_indexes(db: &mut Database) {
    use crate::sql::fts::PostingList;

    for table in db.tables.values_mut() {
        let dirty: Vec<(String, String)> = table
            .fts_indexes
            .iter()
            .filter(|e| e.needs_rebuild)
            .map(|e| (e.name.clone(), e.column_name.clone()))
            .collect();
        if dirty.is_empty() {
            continue;
        }

        for (idx_name, col_name) in dirty {
            // Snapshot every (rowid, text) pair for this column under
            // the row mutex, then drop the lock before re-tokenizing.
            let mut docs: Vec<(i64, String)> = Vec::new();
            {
                let row_data = table.rows.lock().expect("rows mutex poisoned");
                if let Some(Row::Text(map)) = row_data.get(&col_name) {
                    for (id, v) in map.iter() {
                        // "Null" sentinel is the parser's
                        // null-marker for TEXT cells; skip those —
                        // they'd round-trip as the literal string
                        // "Null" otherwise. Aligns with insert_row's
                        // typed_value gate.
                        if v != "Null" {
                            docs.push((*id, v.clone()));
                        }
                    }
                }
            }

            let mut new_idx = PostingList::new();
            // Sort by id so the rebuild is deterministic across runs
            // (the BTreeMap inside PostingList is order-stable, but
            // doc-length aggregation order doesn't matter — sorting
            // here is purely for reproducibility on inspection).
            docs.sort_by_key(|(id, _)| *id);
            for (id, text) in &docs {
                new_idx.insert(*id, text);
            }

            if let Some(entry) = table.fts_indexes.iter_mut().find(|e| e.name == idx_name) {
                entry.index = new_idx;
                entry.needs_rebuild = false;
            }
        }
    }
}

/// Cheap clone helper — `DataType` doesn't derive `Clone` elsewhere.
fn clone_datatype(dt: &DataType) -> DataType {
    match dt {
        DataType::Integer => DataType::Integer,
        DataType::Text => DataType::Text,
        DataType::Real => DataType::Real,
        DataType::Bool => DataType::Bool,
        DataType::Vector(dim) => DataType::Vector(*dim),
        DataType::Json => DataType::Json,
        DataType::None => DataType::None,
        DataType::Invalid => DataType::Invalid,
    }
}

/// Stages an index's B-Tree at `start_page`. Each leaf cell is a
/// `KIND_INDEX` entry carrying `(original_rowid, value)`. Returns
/// `(root_page, next_free_page)`.
///
/// The tree's shape matches a regular table's — leaves chained via
/// `next_page`, optional interior layer above. `Cell::peek_rowid` works
/// uniformly for index cells (same prefix as local cells), so the
/// existing slot directory and binary search carry over.
fn stage_index_btree(
    pager: &mut Pager,
    idx: &SecondaryIndex,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<u32> {
    // Build the leaves.
    let leaves = stage_index_leaves(pager, idx, alloc)?;
    if leaves.len() == 1 {
        return Ok(leaves[0].0);
    }
    let mut level: Vec<(u32, i64)> = leaves;
    while level.len() > 1 {
        level = stage_interior_level(pager, &level, alloc)?;
    }
    Ok(level[0].0)
}

/// Packs the index's (value, rowid) entries into a sibling-chained run
/// of `TableLeaf` pages. Iteration order matches `SecondaryIndex::iter_entries`
/// (ascending value; rowids in insertion order within a value), which is
/// also ascending by the "cell rowid" carried in each IndexCell (the
/// original row's rowid) — so Cell::peek_rowid + the slot directory's
/// rowid ordering stays consistent.
fn stage_index_leaves(
    pager: &mut Pager,
    idx: &SecondaryIndex,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<Vec<(u32, i64)>> {
    let mut leaves: Vec<(u32, i64)> = Vec::new();
    let mut current_leaf = TablePage::empty();
    let mut current_leaf_page = alloc.allocate();
    let mut current_max_rowid: Option<i64> = None;

    // Sort the entries by original rowid so the in-page slot directory,
    // which binary-searches by rowid, stays valid. (iter_entries orders by
    // value; we reorder here for B-Tree correctness.)
    let mut entries: Vec<(Value, i64)> = idx.iter_entries().collect();
    entries.sort_by_key(|(_, r)| *r);

    for (value, rowid) in entries {
        let cell = IndexCell::new(rowid, value);
        let entry_bytes = cell.encode()?;

        if !current_leaf.would_fit(entry_bytes.len()) {
            let next_leaf_page_num = alloc.allocate();
            emit_leaf(pager, current_leaf_page, &current_leaf, next_leaf_page_num);
            leaves.push((current_leaf_page, current_max_rowid.unwrap_or(i64::MIN)));
            current_leaf = TablePage::empty();
            current_leaf_page = next_leaf_page_num;

            if !current_leaf.would_fit(entry_bytes.len()) {
                return Err(SQLRiteError::Internal(format!(
                    "index entry of {} bytes exceeds empty-page capacity {}",
                    entry_bytes.len(),
                    current_leaf.free_space()
                )));
            }
        }
        current_leaf.insert_entry(rowid, &entry_bytes)?;
        current_max_rowid = Some(rowid);
    }

    emit_leaf(pager, current_leaf_page, &current_leaf, 0);
    leaves.push((current_leaf_page, current_max_rowid.unwrap_or(i64::MIN)));
    Ok(leaves)
}

/// Phase 7d.3 — stages an HNSW index's page tree at `start_page`.
/// Each leaf cell is a `KIND_HNSW` entry carrying one node's
/// (node_id, layers). Returns `(root_page, next_free_page)`.
///
/// Tree shape is identical to `stage_index_btree` — chained leaves +
/// optional interior layers. The slot directory binary-searches by
/// node_id (which is the cell's "rowid" in `Cell::peek_rowid` terms),
/// so reads can locate any node in O(log N) once 7d.4-or-later
/// optimizes the load path to lazy-fetch instead of read-all.
/// Today, `load_hnsw_nodes` reads the entire tree on open.
fn stage_hnsw_btree(
    pager: &mut Pager,
    idx: &crate::sql::hnsw::HnswIndex,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<u32> {
    let leaves = stage_hnsw_leaves(pager, idx, alloc)?;
    if leaves.len() == 1 {
        return Ok(leaves[0].0);
    }
    let mut level: Vec<(u32, i64)> = leaves;
    while level.len() > 1 {
        level = stage_interior_level(pager, &level, alloc)?;
    }
    Ok(level[0].0)
}

/// Phase 8c — stage one FTS index as a `TableLeaf`-shaped B-Tree.
/// Mirrors `stage_hnsw_btree` (sibling-chained leaves, optional interior
/// levels). Returns `(root_page, next_free_page)`. Each leaf is filled
/// with `KIND_FTS_POSTING` cells: one sidecar cell holding the
/// doc-lengths map, then one cell per term in lexicographic order.
fn stage_fts_btree(
    pager: &mut Pager,
    idx: &crate::sql::fts::PostingList,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<u32> {
    let leaves = stage_fts_leaves(pager, idx, alloc)?;
    if leaves.len() == 1 {
        return Ok(leaves[0].0);
    }
    let mut level: Vec<(u32, i64)> = leaves;
    while level.len() > 1 {
        level = stage_interior_level(pager, &level, alloc)?;
    }
    Ok(level[0].0)
}

/// Packs FTS posting cells into a sibling-chained run of `TableLeaf`
/// pages. Cell layout: a single doc-lengths sidecar at `cell_id = 1`,
/// followed by one cell per term in lexicographic order with
/// `cell_id = 2..=N + 1`. Sequential ids keep the slot directory's
/// rowid ordering valid (the `cell_id` field is what `peek_rowid`
/// returns).
fn stage_fts_leaves(
    pager: &mut Pager,
    idx: &crate::sql::fts::PostingList,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<Vec<(u32, i64)>> {
    use crate::sql::pager::fts_cell::FtsPostingCell;

    let mut leaves: Vec<(u32, i64)> = Vec::new();
    let mut current_leaf = TablePage::empty();
    let mut current_leaf_page = alloc.allocate();
    let mut current_max_rowid: Option<i64> = None;

    // Build the cell sequence: sidecar first, then per-term cells. The
    // sidecar always exists (even on an empty index) so reload sees a
    // canonical "this index was persisted" marker in slot 0.
    let mut cell_id: i64 = 1;
    let mut cells: Vec<FtsPostingCell> = Vec::new();
    cells.push(FtsPostingCell::doc_lengths(
        cell_id,
        idx.serialize_doc_lengths(),
    ));
    for (term, entries) in idx.serialize_postings() {
        cell_id += 1;
        cells.push(FtsPostingCell::posting(cell_id, term, entries));
    }

    for cell in cells {
        let entry_bytes = cell.encode()?;

        if !current_leaf.would_fit(entry_bytes.len()) {
            let next_leaf_page_num = alloc.allocate();
            emit_leaf(pager, current_leaf_page, &current_leaf, next_leaf_page_num);
            leaves.push((current_leaf_page, current_max_rowid.unwrap_or(i64::MIN)));
            current_leaf = TablePage::empty();
            current_leaf_page = next_leaf_page_num;

            if !current_leaf.would_fit(entry_bytes.len()) {
                // A single posting cell exceeds page capacity. Phase
                // 8c MVP doesn't chain via overflow cells (the plan
                // notes this as a stretch goal); surface a clear
                // error so users know which term tripped it.
                return Err(SQLRiteError::Internal(format!(
                    "FTS posting cell {} of {} bytes exceeds empty-page capacity {} \
                     (term too long or too many postings; overflow chaining is Phase 8.1)",
                    cell.cell_id,
                    entry_bytes.len(),
                    current_leaf.free_space()
                )));
            }
        }
        current_leaf.insert_entry(cell.cell_id, &entry_bytes)?;
        current_max_rowid = Some(cell.cell_id);
    }

    emit_leaf(pager, current_leaf_page, &current_leaf, 0);
    leaves.push((current_leaf_page, current_max_rowid.unwrap_or(i64::MIN)));
    Ok(leaves)
}

/// (rowid, value) pairs as decoded from a single FTS cell — value is
/// either term frequency (posting cell) or doc length (sidecar cell).
type FtsEntries = Vec<(i64, u32)>;
/// (term, posting list) pairs as decoded from non-sidecar FTS cells.
type FtsPostings = Vec<(String, FtsEntries)>;

/// Phase 8c — read every cell of an FTS index from `root_page` back
/// into the `(doc_lengths, postings)` shape `PostingList::from_persisted_postings`
/// expects. Mirrors `load_hnsw_nodes`: leftmost-leaf descent, walk the
/// sibling chain, decode each slot.
fn load_fts_postings(pager: &Pager, root_page: u32) -> Result<(FtsEntries, FtsPostings)> {
    use crate::sql::pager::fts_cell::FtsPostingCell;

    let mut doc_lengths: Vec<(i64, u32)> = Vec::new();
    let mut postings: Vec<(String, Vec<(i64, u32)>)> = Vec::new();
    let mut saw_sidecar = false;

    let first_leaf = find_leftmost_leaf(pager, root_page)?;
    let mut current = first_leaf;
    while current != 0 {
        let page_buf = pager
            .read_page(current)
            .ok_or_else(|| SQLRiteError::Internal(format!("missing FTS leaf page {current}")))?;
        if page_buf[0] != PageType::TableLeaf as u8 {
            return Err(SQLRiteError::Internal(format!(
                "page {current} tagged {} but expected TableLeaf (FTS)",
                page_buf[0]
            )));
        }
        let next_leaf = u32::from_le_bytes(page_buf[1..5].try_into().unwrap());
        let payload: &[u8; PAYLOAD_PER_PAGE] = (&page_buf[PAGE_HEADER_SIZE..])
            .try_into()
            .map_err(|_| SQLRiteError::Internal("FTS leaf payload size".to_string()))?;
        let leaf = TablePage::from_bytes(payload);
        for slot in 0..leaf.slot_count() {
            let offset = leaf.slot_offset_raw(slot)?;
            let (cell, _) = FtsPostingCell::decode(leaf.as_bytes(), offset)?;
            if cell.is_doc_lengths() {
                if saw_sidecar {
                    return Err(SQLRiteError::Internal(
                        "FTS index has more than one doc-lengths sidecar cell".to_string(),
                    ));
                }
                saw_sidecar = true;
                doc_lengths = cell.entries;
            } else {
                postings.push((cell.term, cell.entries));
            }
        }
        current = next_leaf;
    }

    if !saw_sidecar {
        return Err(SQLRiteError::Internal(
            "FTS index missing doc-lengths sidecar cell — corrupt or truncated tree".to_string(),
        ));
    }
    Ok((doc_lengths, postings))
}

/// Packs HNSW nodes into a sibling-chained run of `TableLeaf` pages.
/// `serialize_nodes` already returns nodes in ascending node_id order,
/// so the slot directory's rowid ordering stays valid.
fn stage_hnsw_leaves(
    pager: &mut Pager,
    idx: &crate::sql::hnsw::HnswIndex,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<Vec<(u32, i64)>> {
    use crate::sql::pager::hnsw_cell::HnswNodeCell;

    let mut leaves: Vec<(u32, i64)> = Vec::new();
    let mut current_leaf = TablePage::empty();
    let mut current_leaf_page = alloc.allocate();
    let mut current_max_rowid: Option<i64> = None;

    let serialized = idx.serialize_nodes();

    // Empty index → emit a single empty leaf page so the rootpage
    // pointer in sqlrite_master stays nonzero (== "graph is persisted,
    // it just happens to be empty"). load_hnsw_nodes is fine with an
    // empty leaf — slot_count() returns 0.
    for (node_id, layers) in serialized {
        let cell = HnswNodeCell::new(node_id, layers);
        let entry_bytes = cell.encode()?;

        if !current_leaf.would_fit(entry_bytes.len()) {
            let next_leaf_page_num = alloc.allocate();
            emit_leaf(pager, current_leaf_page, &current_leaf, next_leaf_page_num);
            leaves.push((current_leaf_page, current_max_rowid.unwrap_or(i64::MIN)));
            current_leaf = TablePage::empty();
            current_leaf_page = next_leaf_page_num;

            if !current_leaf.would_fit(entry_bytes.len()) {
                return Err(SQLRiteError::Internal(format!(
                    "HNSW node {node_id} cell of {} bytes exceeds empty-page capacity {}",
                    entry_bytes.len(),
                    current_leaf.free_space()
                )));
            }
        }
        current_leaf.insert_entry(node_id, &entry_bytes)?;
        current_max_rowid = Some(node_id);
    }

    emit_leaf(pager, current_leaf_page, &current_leaf, 0);
    leaves.push((current_leaf_page, current_max_rowid.unwrap_or(i64::MIN)));
    Ok(leaves)
}

fn load_table_rows(pager: &Pager, table: &mut Table, root_page: u32) -> Result<()> {
    let first_leaf = find_leftmost_leaf(pager, root_page)?;
    let mut current = first_leaf;
    while current != 0 {
        let page_buf = pager
            .read_page(current)
            .ok_or_else(|| SQLRiteError::Internal(format!("missing leaf page {current}")))?;
        if page_buf[0] != PageType::TableLeaf as u8 {
            return Err(SQLRiteError::Internal(format!(
                "page {current} tagged {} but expected TableLeaf",
                page_buf[0]
            )));
        }
        let next_leaf = u32::from_le_bytes(page_buf[1..5].try_into().unwrap());
        let payload: &[u8; PAYLOAD_PER_PAGE] = (&page_buf[PAGE_HEADER_SIZE..])
            .try_into()
            .map_err(|_| SQLRiteError::Internal("leaf payload slice size".to_string()))?;
        let leaf = TablePage::from_bytes(payload);

        for slot in 0..leaf.slot_count() {
            let entry = leaf.entry_at(slot)?;
            let cell = match entry {
                PagedEntry::Local(c) => c,
                PagedEntry::Overflow(r) => {
                    let body_bytes =
                        read_overflow_chain(pager, r.first_overflow_page, r.total_body_len)?;
                    let (c, _) = Cell::decode(&body_bytes, 0)?;
                    c
                }
            };
            table.restore_row(cell.rowid, cell.values)?;
        }
        current = next_leaf;
    }
    Ok(())
}

/// Walks every page reachable from `root_page` and returns their page
/// numbers. Includes `root_page`, every interior page, every leaf, and
/// — when `follow_overflow` is true — every overflow page chained off
/// table-leaf cells. Used by `save_database` to seed each table's
/// per-table preferred pool and to compute the newly-freed set.
///
/// `follow_overflow = true` for table B-Trees (cells may carry
/// `OverflowRef`s pointing at chained overflow pages); `false` for
/// secondary-index, HNSW, and FTS B-Trees, which never overflow in the
/// current encoding.
fn collect_pages_for_btree(
    pager: &Pager,
    root_page: u32,
    follow_overflow: bool,
) -> Result<Vec<u32>> {
    if root_page == 0 {
        return Ok(Vec::new());
    }
    let mut pages: Vec<u32> = Vec::new();
    let mut stack: Vec<u32> = vec![root_page];

    while let Some(p) = stack.pop() {
        let buf = pager.read_page(p).ok_or_else(|| {
            SQLRiteError::Internal(format!(
                "collect_pages: missing page {p} (rooted at {root_page})"
            ))
        })?;
        pages.push(p);
        match buf[0] {
            t if t == PageType::InteriorNode as u8 => {
                let payload: &[u8; PAYLOAD_PER_PAGE] =
                    (&buf[PAGE_HEADER_SIZE..]).try_into().map_err(|_| {
                        SQLRiteError::Internal("interior payload slice size".to_string())
                    })?;
                let interior = InteriorPage::from_bytes(payload);
                // Push every divider's child + the rightmost child.
                for slot in 0..interior.slot_count() {
                    let cell = interior.cell_at(slot)?;
                    stack.push(cell.child_page);
                }
                stack.push(interior.rightmost_child());
            }
            t if t == PageType::TableLeaf as u8 => {
                if follow_overflow {
                    let payload: &[u8; PAYLOAD_PER_PAGE] =
                        (&buf[PAGE_HEADER_SIZE..]).try_into().map_err(|_| {
                            SQLRiteError::Internal("leaf payload slice size".to_string())
                        })?;
                    let leaf = TablePage::from_bytes(payload);
                    for slot in 0..leaf.slot_count() {
                        match leaf.entry_at(slot)? {
                            PagedEntry::Local(_) => {}
                            PagedEntry::Overflow(r) => {
                                let mut cur = r.first_overflow_page;
                                while cur != 0 {
                                    pages.push(cur);
                                    let ob = pager.read_page(cur).ok_or_else(|| {
                                        SQLRiteError::Internal(format!(
                                            "collect_pages: missing overflow page {cur}"
                                        ))
                                    })?;
                                    if ob[0] != PageType::Overflow as u8 {
                                        return Err(SQLRiteError::Internal(format!(
                                            "collect_pages: page {cur} expected Overflow, got tag {}",
                                            ob[0]
                                        )));
                                    }
                                    cur = u32::from_le_bytes(ob[1..5].try_into().unwrap());
                                }
                            }
                        }
                    }
                }
            }
            other => {
                return Err(SQLRiteError::Internal(format!(
                    "collect_pages: unexpected page type {other} at page {p}"
                )));
            }
        }
    }
    Ok(pages)
}

/// Reads the previously-persisted `sqlrite_master` and returns a map from
/// `(kind, name)` to that object's rootpage. Used by `save_database` to
/// seed each table/index's per-table preferred pool with the pages it
/// occupied last time round.
///
/// `kind` is `"table"` or `"index"` (the catalog already disambiguates
/// the three index families via the SQL string, but for page-collection
/// purposes a "table" tree must follow overflow refs while an "index"
/// tree never does — that's the only distinction we need here).
fn read_old_rootpages(pager: &Pager, schema_root: u32) -> Result<HashMap<(String, String), u32>> {
    let mut out: HashMap<(String, String), u32> = HashMap::new();
    if schema_root == 0 {
        return Ok(out);
    }
    let mut master = build_empty_master_table();
    load_table_rows(pager, &mut master, schema_root)?;
    for rowid in master.rowids() {
        let kind = take_text(&master, "type", rowid)?;
        let name = take_text(&master, "name", rowid)?;
        let rootpage = take_integer(&master, "rootpage", rowid)? as u32;
        out.insert((kind, name), rootpage);
    }
    Ok(out)
}

/// Descends from `root_page` through `InteriorNode` pages, always taking
/// the leftmost child, until a `TableLeaf` is reached. Returns that leaf's
/// page number. A root that's already a leaf is returned as-is.
fn find_leftmost_leaf(pager: &Pager, root_page: u32) -> Result<u32> {
    let mut current = root_page;
    loop {
        let page_buf = pager.read_page(current).ok_or_else(|| {
            SQLRiteError::Internal(format!("missing page {current} during tree descent"))
        })?;
        match page_buf[0] {
            t if t == PageType::TableLeaf as u8 => return Ok(current),
            t if t == PageType::InteriorNode as u8 => {
                let payload: &[u8; PAYLOAD_PER_PAGE] =
                    (&page_buf[PAGE_HEADER_SIZE..]).try_into().map_err(|_| {
                        SQLRiteError::Internal("interior payload slice size".to_string())
                    })?;
                let interior = InteriorPage::from_bytes(payload);
                current = interior.leftmost_child()?;
            }
            other => {
                return Err(SQLRiteError::Internal(format!(
                    "unexpected page type {other} during tree descent at page {current}"
                )));
            }
        }
    }
}

/// Stages a table's B-Tree, drawing every page number from `alloc`.
/// Returns the root page (the topmost interior page, or the single leaf
/// when the table fits in one page).
///
/// Builds bottom-up: pack rows into `TableLeaf` pages chained via
/// `next_page`, then if more than one leaf, recursively wrap them in
/// `InteriorNode` levels until one root remains.
///
/// Deterministic: same rows + same allocator handouts → byte-identical
/// pages at the same numbers, so the diff pager skips unchanged tables.
fn stage_table_btree(
    pager: &mut Pager,
    table: &Table,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<u32> {
    let leaves = stage_leaves(pager, table, alloc)?;
    if leaves.len() == 1 {
        return Ok(leaves[0].0);
    }
    let mut level: Vec<(u32, i64)> = leaves;
    while level.len() > 1 {
        level = stage_interior_level(pager, &level, alloc)?;
    }
    Ok(level[0].0)
}

/// Packs the table's rows into a sibling-linked chain of `TableLeaf` pages.
/// Returns each leaf's `(page_number, max_rowid)` for use by the next
/// interior level. Allocates leaf and overflow pages from `alloc`.
fn stage_leaves(
    pager: &mut Pager,
    table: &Table,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<Vec<(u32, i64)>> {
    let mut leaves: Vec<(u32, i64)> = Vec::new();
    let mut current_leaf = TablePage::empty();
    let mut current_leaf_page = alloc.allocate();
    let mut current_max_rowid: Option<i64> = None;

    for rowid in table.rowids() {
        let entry_bytes = build_row_entry(pager, table, rowid, alloc)?;

        if !current_leaf.would_fit(entry_bytes.len()) {
            // The new leaf goes at whatever the allocator hands out
            // next. Commit the current leaf with that as its sibling
            // pointer.
            let next_leaf_page_num = alloc.allocate();
            emit_leaf(pager, current_leaf_page, &current_leaf, next_leaf_page_num);
            leaves.push((current_leaf_page, current_max_rowid.unwrap_or(i64::MIN)));
            current_leaf = TablePage::empty();
            current_leaf_page = next_leaf_page_num;
            // current_max_rowid is reassigned by the insert below; no need
            // to zero it out here.

            if !current_leaf.would_fit(entry_bytes.len()) {
                return Err(SQLRiteError::Internal(format!(
                    "entry of {} bytes exceeds empty-page capacity {}",
                    entry_bytes.len(),
                    current_leaf.free_space()
                )));
            }
        }
        current_leaf.insert_entry(rowid, &entry_bytes)?;
        current_max_rowid = Some(rowid);
    }

    // Final leaf: sibling next_page = 0 (end of chain).
    emit_leaf(pager, current_leaf_page, &current_leaf, 0);
    leaves.push((current_leaf_page, current_max_rowid.unwrap_or(i64::MIN)));
    Ok(leaves)
}

/// Encodes a single row's on-leaf entry — either the local cell bytes, or
/// an `OverflowRef` pointing at a freshly-allocated overflow chain if the
/// encoded cell exceeded the inline threshold. Allocates any overflow
/// pages from `alloc`.
fn build_row_entry(
    pager: &mut Pager,
    table: &Table,
    rowid: i64,
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<Vec<u8>> {
    let values = table.extract_row(rowid);
    let local_cell = Cell::new(rowid, values);
    let local_bytes = local_cell.encode()?;
    if local_bytes.len() > OVERFLOW_THRESHOLD {
        let overflow_start = write_overflow_chain(pager, &local_bytes, alloc)?;
        Ok(OverflowRef {
            rowid,
            total_body_len: local_bytes.len() as u64,
            first_overflow_page: overflow_start,
        }
        .encode())
    } else {
        Ok(local_bytes)
    }
}

/// Builds one level of `InteriorNode` pages above the given children.
/// Each interior packs as many dividers as will fit; the last child
/// assigned to an interior becomes its `rightmost_child`. Returns the
/// emitted interior pages as `(page_number, max_rowid_in_subtree)`.
fn stage_interior_level(
    pager: &mut Pager,
    children: &[(u32, i64)],
    alloc: &mut crate::sql::pager::allocator::PageAllocator,
) -> Result<Vec<(u32, i64)>> {
    let mut next_level: Vec<(u32, i64)> = Vec::new();
    let mut idx = 0usize;

    while idx < children.len() {
        let interior_page_num = alloc.allocate();

        // Seed the interior with the first unassigned child as its
        // rightmost. As we add more children, the previous rightmost
        // graduates to being a divider and the new arrival takes over
        // as rightmost.
        let (mut rightmost_child_page, mut rightmost_child_max) = children[idx];
        idx += 1;
        let mut interior = InteriorPage::empty(rightmost_child_page);

        while idx < children.len() {
            let new_divider_cell = InteriorCell {
                divider_rowid: rightmost_child_max,
                child_page: rightmost_child_page,
            };
            let new_divider_bytes = new_divider_cell.encode();
            if !interior.would_fit(new_divider_bytes.len()) {
                break;
            }
            interior.insert_divider(rightmost_child_max, rightmost_child_page)?;
            let (next_child_page, next_child_max) = children[idx];
            interior.set_rightmost_child(next_child_page);
            rightmost_child_page = next_child_page;
            rightmost_child_max = next_child_max;
            idx += 1;
        }

        emit_interior(pager, interior_page_num, &interior);
        next_level.push((interior_page_num, rightmost_child_max));
    }

    Ok(next_level)
}

/// Wraps a `TablePage` in the 7-byte page header and hands it to the pager.
fn emit_leaf(pager: &mut Pager, page_num: u32, leaf: &TablePage, next_leaf: u32) {
    let mut buf = [0u8; PAGE_SIZE];
    buf[0] = PageType::TableLeaf as u8;
    buf[1..5].copy_from_slice(&next_leaf.to_le_bytes());
    // For leaf pages the legacy `payload_len` field isn't used — the slot
    // directory self-describes. Zero it by convention.
    buf[5..7].copy_from_slice(&0u16.to_le_bytes());
    buf[PAGE_HEADER_SIZE..].copy_from_slice(leaf.as_bytes());
    pager.stage_page(page_num, buf);
}

/// Wraps an `InteriorPage` in the 7-byte page header. Interior pages
/// don't use `next_page` (there's no sibling chain between interiors);
/// `payload_len` is also unused (the slot directory self-describes).
fn emit_interior(pager: &mut Pager, page_num: u32, interior: &InteriorPage) {
    let mut buf = [0u8; PAGE_SIZE];
    buf[0] = PageType::InteriorNode as u8;
    buf[1..5].copy_from_slice(&0u32.to_le_bytes());
    buf[5..7].copy_from_slice(&0u16.to_le_bytes());
    buf[PAGE_HEADER_SIZE..].copy_from_slice(interior.as_bytes());
    pager.stage_page(page_num, buf);
}

#[cfg(test)]
mod tests {
    use super::*;
    use crate::sql::pager::freelist::MIN_PAGES_FOR_AUTO_VACUUM;
    use crate::sql::process_command;

    fn seed_db() -> Database {
        let mut db = Database::new("test".to_string());
        process_command(
            "CREATE TABLE users (id INTEGER PRIMARY KEY, name TEXT NOT NULL UNIQUE, age INTEGER);",
            &mut db,
        )
        .unwrap();
        process_command(
            "INSERT INTO users (name, age) VALUES ('alice', 30);",
            &mut db,
        )
        .unwrap();
        process_command("INSERT INTO users (name, age) VALUES ('bob', 25);", &mut db).unwrap();
        process_command(
            "CREATE TABLE notes (id INTEGER PRIMARY KEY, body TEXT);",
            &mut db,
        )
        .unwrap();
        process_command("INSERT INTO notes (body) VALUES ('hello');", &mut db).unwrap();
        db
    }

    fn tmp_path(name: &str) -> std::path::PathBuf {
        let mut p = std::env::temp_dir();
        let pid = std::process::id();
        let nanos = std::time::SystemTime::now()
            .duration_since(std::time::UNIX_EPOCH)
            .map(|d| d.as_nanos())
            .unwrap_or(0);
        p.push(format!("sqlrite-{pid}-{nanos}-{name}.sqlrite"));
        p
    }

    /// Phase 4c: every .sqlrite has a `-wal` sidecar now. Delete both so
    /// `/tmp` doesn't accumulate orphan WALs across test runs.
    fn cleanup(path: &std::path::Path) {
        let _ = std::fs::remove_file(path);
        let mut wal = path.as_os_str().to_owned();
        wal.push("-wal");
        let _ = std::fs::remove_file(std::path::PathBuf::from(wal));
    }

    #[test]
    fn round_trip_preserves_schema_and_data() {
        let path = tmp_path("roundtrip");
        let mut db = seed_db();
        save_database(&mut db, &path).expect("save");

        let loaded = open_database(&path, "test".to_string()).expect("open");
        assert_eq!(loaded.tables.len(), 2);

        let users = loaded.get_table("users".to_string()).expect("users table");
        assert_eq!(users.columns.len(), 3);
        let rowids = users.rowids();
        assert_eq!(rowids.len(), 2);
        let names: Vec<String> = rowids
            .iter()
            .filter_map(|r| match users.get_value("name", *r) {
                Some(Value::Text(s)) => Some(s),
                _ => None,
            })
            .collect();
        assert!(names.contains(&"alice".to_string()));
        assert!(names.contains(&"bob".to_string()));

        let notes = loaded.get_table("notes".to_string()).expect("notes table");
        assert_eq!(notes.rowids().len(), 1);

        cleanup(&path);
    }

    // -----------------------------------------------------------------
    // Phase 7a — VECTOR(N) save / reopen round-trip
    // -----------------------------------------------------------------

    #[test]
    fn round_trip_preserves_vector_column() {
        let path = tmp_path("vec_roundtrip");

        // Build, populate, save.
        {
            let mut db = Database::new("test".to_string());
            process_command(
                "CREATE TABLE docs (id INTEGER PRIMARY KEY, embedding VECTOR(3));",
                &mut db,
            )
            .unwrap();
            process_command(
                "INSERT INTO docs (embedding) VALUES ([0.1, 0.2, 0.3]);",
                &mut db,
            )
            .unwrap();
            process_command(
                "INSERT INTO docs (embedding) VALUES ([1.5, -2.0, 3.5]);",
                &mut db,
            )
            .unwrap();
            save_database(&mut db, &path).expect("save");
        } // db drops → its exclusive lock releases before reopen.

        // Reopen and verify schema + data both round-tripped.
        let loaded = open_database(&path, "test".to_string()).expect("open");
        let docs = loaded.get_table("docs".to_string()).expect("docs table");

        // Schema preserved: column is still VECTOR(3).
        let embedding_col = docs
            .columns
            .iter()
            .find(|c| c.column_name == "embedding")
            .expect("embedding column");
        assert!(
            matches!(embedding_col.datatype, DataType::Vector(3)),
            "expected DataType::Vector(3) after round-trip, got {:?}",
            embedding_col.datatype
        );

        // Data preserved: both vectors still readable bit-for-bit.
        let mut rows: Vec<Vec<f32>> = docs
            .rowids()
            .iter()
            .filter_map(|r| match docs.get_value("embedding", *r) {
                Some(Value::Vector(v)) => Some(v),
                _ => None,
            })
            .collect();
        rows.sort_by(|a, b| a[0].partial_cmp(&b[0]).unwrap());
        assert_eq!(rows.len(), 2);
        assert_eq!(rows[0], vec![0.1f32, 0.2, 0.3]);
        assert_eq!(rows[1], vec![1.5f32, -2.0, 3.5]);

        cleanup(&path);
    }

    #[test]
    fn round_trip_preserves_json_column() {
        // Phase 7e — JSON columns are stored as Text under the hood with
        // INSERT-time validation. Save + reopen should preserve the
        // schema (DataType::Json) and the underlying text bytes; a
        // post-reopen json_extract should still resolve paths correctly.
        let path = tmp_path("json_roundtrip");

        {
            let mut db = Database::new("test".to_string());
            process_command(
                "CREATE TABLE docs (id INTEGER PRIMARY KEY, payload JSON);",
                &mut db,
            )
            .unwrap();
            process_command(
                r#"INSERT INTO docs (payload) VALUES ('{"name": "alice", "tags": ["rust","sql"]}');"#,
                &mut db,
            )
            .unwrap();
            save_database(&mut db, &path).expect("save");
        }

        let mut loaded = open_database(&path, "test".to_string()).expect("open");
        let docs = loaded.get_table("docs".to_string()).expect("docs");

        // Schema: column declared as JSON, restored with the same type.
        let payload_col = docs
            .columns
            .iter()
            .find(|c| c.column_name == "payload")
            .unwrap();
        assert!(
            matches!(payload_col.datatype, DataType::Json),
            "expected DataType::Json, got {:?}",
            payload_col.datatype
        );

        // json_extract works against the reopened data — exercises the
        // full Text-storage + serde_json::from_str path post-reopen.
        let resp = process_command(
            r#"SELECT id FROM docs WHERE json_extract(payload, '$.name') = 'alice';"#,
            &mut loaded,
        )
        .expect("select via json_extract after reopen");
        assert!(resp.contains("1 row returned"), "got: {resp}");

        cleanup(&path);
    }

    #[test]
    fn round_trip_rebuilds_hnsw_index_from_create_sql() {
        // Phase 7d.3: HNSW indexes now persist their graph as cell-encoded
        // pages. After save+reopen the index entry reattaches with the
        // same column + same node count, loaded directly from disk
        // instead of re-walking rows.
        let path = tmp_path("hnsw_roundtrip");

        // Build, populate, index, save.
        {
            let mut db = Database::new("test".to_string());
            process_command(
                "CREATE TABLE docs (id INTEGER PRIMARY KEY, e VECTOR(2));",
                &mut db,
            )
            .unwrap();
            for v in &[
                "[1.0, 0.0]",
                "[2.0, 0.0]",
                "[0.0, 3.0]",
                "[1.0, 4.0]",
                "[10.0, 10.0]",
            ] {
                process_command(&format!("INSERT INTO docs (e) VALUES ({v});"), &mut db).unwrap();
            }
            process_command("CREATE INDEX ix_e ON docs USING hnsw (e);", &mut db).unwrap();
            save_database(&mut db, &path).expect("save");
        } // db drops → exclusive lock releases.

        // Reopen and verify the index reattached, with the same name +
        // column + populated graph.
        let mut loaded = open_database(&path, "test".to_string()).expect("open");
        {
            let table = loaded.get_table("docs".to_string()).expect("docs");
            assert_eq!(table.hnsw_indexes.len(), 1, "HNSW index should reattach");
            let entry = &table.hnsw_indexes[0];
            assert_eq!(entry.name, "ix_e");
            assert_eq!(entry.column_name, "e");
            assert_eq!(entry.index.len(), 5, "loaded graph should hold all 5 rows");
            assert!(
                !entry.needs_rebuild,
                "fresh load should not be marked dirty"
            );
        }

        // Quick functional check: KNN query through the loaded index
        // returns results.
        let resp = process_command(
            "SELECT id FROM docs ORDER BY vec_distance_l2(e, [1.0, 0.0]) ASC LIMIT 3;",
            &mut loaded,
        )
        .unwrap();
        assert!(resp.contains("3 rows returned"), "got: {resp}");

        cleanup(&path);
    }

    /// SQLR-28 — the HNSW metric must round-trip across save+reopen.
    /// Without this, the SQL re-synthesised into `sqlrite_master`
    /// would drop the metric and a cosine-built graph would reload
    /// as L2, silently breaking subsequent cosine probes.
    #[test]
    fn round_trip_preserves_hnsw_cosine_metric() {
        use crate::sql::hnsw::DistanceMetric;
        let path = tmp_path("hnsw_metric_roundtrip");

        {
            let mut db = Database::new("test".to_string());
            process_command(
                "CREATE TABLE docs (id INTEGER PRIMARY KEY, e VECTOR(2));",
                &mut db,
            )
            .unwrap();
            for v in &["[1.0, 0.0]", "[0.0, 1.0]", "[0.7071, 0.7071]"] {
                process_command(&format!("INSERT INTO docs (e) VALUES ({v});"), &mut db).unwrap();
            }
            process_command(
                "CREATE INDEX ix_cos ON docs USING hnsw (e) WITH (metric = 'cosine');",
                &mut db,
            )
            .unwrap();
            save_database(&mut db, &path).expect("save");
        }

        let mut loaded = open_database(&path, "test".to_string()).expect("open");
        {
            let table = loaded.get_table("docs".to_string()).expect("docs");
            assert_eq!(table.hnsw_indexes.len(), 1);
            assert_eq!(
                table.hnsw_indexes[0].metric,
                DistanceMetric::Cosine,
                "metric should round-trip through CREATE INDEX SQL"
            );
            assert_eq!(table.hnsw_indexes[0].index.distance, DistanceMetric::Cosine);
        }

        // Cosine probe still finds the self-vector after reopen — the
        // optimizer's metric gate should match the loaded entry's
        // metric, so this should hit the graph shortcut.
        let resp = process_command(
            "SELECT id FROM docs ORDER BY vec_distance_cosine(e, [1.0, 0.0]) ASC LIMIT 1;",
            &mut loaded,
        )
        .unwrap();
        assert!(resp.contains("1 row returned"), "got: {resp}");

        cleanup(&path);
    }

    #[test]
    fn round_trip_rebuilds_fts_index_from_create_sql() {
        // Phase 8c: FTS indexes now persist their posting lists as
        // cell-encoded pages. After save+reopen the index entry
        // reattaches with the same column + same posting count, loaded
        // directly from disk (no re-tokenization).
        let path = tmp_path("fts_roundtrip");

        {
            let mut db = Database::new("test".to_string());
            process_command(
                "CREATE TABLE docs (id INTEGER PRIMARY KEY, body TEXT);",
                &mut db,
            )
            .unwrap();
            for body in &[
                "rust embedded database",
                "rust web framework",
                "go embedded systems",
                "python web framework",
                "rust rust embedded power",
            ] {
                process_command(
                    &format!("INSERT INTO docs (body) VALUES ('{body}');"),
                    &mut db,
                )
                .unwrap();
            }
            process_command("CREATE INDEX ix_body ON docs USING fts (body);", &mut db).unwrap();
            save_database(&mut db, &path).expect("save");
        } // db drops → exclusive lock releases.

        let mut loaded = open_database(&path, "test".to_string()).expect("open");
        {
            let table = loaded.get_table("docs".to_string()).expect("docs");
            assert_eq!(table.fts_indexes.len(), 1, "FTS index should reattach");
            let entry = &table.fts_indexes[0];
            assert_eq!(entry.name, "ix_body");
            assert_eq!(entry.column_name, "body");
            assert_eq!(
                entry.index.len(),
                5,
                "rebuilt posting list should hold all 5 rows"
            );
            assert!(!entry.needs_rebuild);
        }

        // Functional smoke: an FTS query through the reloaded index
        // returns the expected hit count.
        let resp = process_command(
            "SELECT id FROM docs WHERE fts_match(body, 'rust');",
            &mut loaded,
        )
        .unwrap();
        assert!(resp.contains("3 rows returned"), "got: {resp}");

        cleanup(&path);
    }

    #[test]
    fn delete_then_save_then_reopen_excludes_deleted_node_from_fts() {
        // Phase 8b — DELETE marks the FTS index dirty; save rebuilds it
        // from current rows; reopen replays the CREATE INDEX SQL against
        // the post-delete row set. The deleted rowid must not surface
        // in `fts_match` results post-reopen.
        let path = tmp_path("fts_delete_rebuild");
        let mut db = Database::new("test".to_string());
        process_command(
            "CREATE TABLE docs (id INTEGER PRIMARY KEY, body TEXT);",
            &mut db,
        )
        .unwrap();
        for body in &[
            "rust embedded",
            "rust framework",
            "go embedded",
            "python web",
        ] {
            process_command(
                &format!("INSERT INTO docs (body) VALUES ('{body}');"),
                &mut db,
            )
            .unwrap();
        }
        process_command("CREATE INDEX ix_body ON docs USING fts (body);", &mut db).unwrap();

        // Delete row 1 ('rust embedded'); save (rebuild fires); reopen.
        process_command("DELETE FROM docs WHERE id = 1;", &mut db).unwrap();
        save_database(&mut db, &path).expect("save");
        drop(db);

        let mut loaded = open_database(&path, "test".to_string()).expect("open");
        let resp = process_command(
            "SELECT id FROM docs WHERE fts_match(body, 'rust');",
            &mut loaded,
        )
        .unwrap();
        // Pre-delete: 2 rows ('rust embedded', 'rust framework') had
        // 'rust'. Post-delete: only id=2 remains.
        assert!(resp.contains("1 row returned"), "got: {resp}");

        cleanup(&path);
    }

    #[test]
    fn fts_roundtrip_uses_persistence_path_not_replay() {
        // Phase 8c — assert the reload didn't go through the
        // rootpage=0 replay shortcut. We do this by reading the
        // sqlrite_master row for the FTS index and confirming its
        // rootpage field is non-zero.
        let path = tmp_path("fts_persistence_path");

        {
            let mut db = Database::new("test".to_string());
            process_command(
                "CREATE TABLE docs (id INTEGER PRIMARY KEY, body TEXT);",
                &mut db,
            )
            .unwrap();
            process_command(
                "INSERT INTO docs (body) VALUES ('rust embedded database');",
                &mut db,
            )
            .unwrap();
            process_command("CREATE INDEX ix_body ON docs USING fts (body);", &mut db).unwrap();
            save_database(&mut db, &path).expect("save");
        }

        // Read raw sqlrite_master to find the FTS index row.
        let pager = Pager::open(&path).expect("open pager");
        let mut master = build_empty_master_table();
        load_table_rows(&pager, &mut master, pager.header().schema_root_page).unwrap();
        let mut found_rootpage: Option<u32> = None;
        for rowid in master.rowids() {
            let name = take_text(&master, "name", rowid).unwrap();
            if name == "ix_body" {
                let rp = take_integer(&master, "rootpage", rowid).unwrap();
                found_rootpage = Some(rp as u32);
            }
        }
        let rootpage = found_rootpage.expect("ix_body row in sqlrite_master");
        assert!(
            rootpage != 0,
            "Phase 8c FTS save should set rootpage != 0; got {rootpage}"
        );

        cleanup(&path);
    }

    #[test]
    fn save_without_fts_keeps_format_v4() {
        // Phase 8c on-demand bump — a database with zero FTS indexes
        // continues writing the v4 header. Existing v4 users must not
        // see their files silently promoted to v5 by an upgrade.
        use crate::sql::pager::header::FORMAT_VERSION_V4;

        let path = tmp_path("fts_no_bump");
        let mut db = Database::new("test".to_string());
        process_command(
            "CREATE TABLE t (id INTEGER PRIMARY KEY, n INTEGER);",
            &mut db,
        )
        .unwrap();
        process_command("INSERT INTO t (n) VALUES (1);", &mut db).unwrap();
        save_database(&mut db, &path).unwrap();
        drop(db);

        let pager = Pager::open(&path).expect("open");
        assert_eq!(
            pager.header().format_version,
            FORMAT_VERSION_V4,
            "no-FTS save should keep v4"
        );
        cleanup(&path);
    }

    #[test]
    fn save_with_fts_bumps_to_v5() {
        // Phase 8c on-demand bump — first FTS-bearing save promotes
        // the file to v5. v5 readers handle both v4 and v5; v4
        // readers correctly refuse a v5 file.
        use crate::sql::pager::header::FORMAT_VERSION_V5;

        let path = tmp_path("fts_bump_v5");
        let mut db = Database::new("test".to_string());
        process_command(
            "CREATE TABLE docs (id INTEGER PRIMARY KEY, body TEXT);",
            &mut db,
        )
        .unwrap();
        process_command("INSERT INTO docs (body) VALUES ('hello');", &mut db).unwrap();
        process_command("CREATE INDEX ix_body ON docs USING fts (body);", &mut db).unwrap();
        save_database(&mut db, &path).unwrap();
        drop(db);

        let pager = Pager::open(&path).expect("open");
        assert_eq!(
            pager.header().format_version,
            FORMAT_VERSION_V5,
            "FTS save should promote to v5"
        );
        cleanup(&path);
    }

    #[test]
    fn fts_persistence_handles_empty_and_zero_token_docs() {
        // Phase 8c — sidecar cell carries doc-lengths for every doc
        // including any with zero tokens (so total_docs is honest
        // post-reopen). Empty index also round-trips: a CREATE INDEX
        // on an empty table emits a single empty leaf with just the
        // (empty) sidecar.
        let path = tmp_path("fts_edges");

        {
            let mut db = Database::new("test".to_string());
            process_command(
                "CREATE TABLE docs (id INTEGER PRIMARY KEY, body TEXT);",
                &mut db,
            )
            .unwrap();
            process_command("CREATE INDEX ix_body ON docs USING fts (body);", &mut db).unwrap();
            // Mix: real text, then a row that tokenizes to zero tokens
            // (only punctuation), then real again.
            process_command("INSERT INTO docs (body) VALUES ('rust embedded');", &mut db).unwrap();
            process_command("INSERT INTO docs (body) VALUES ('!!!---???');", &mut db).unwrap();
            process_command("INSERT INTO docs (body) VALUES ('go embedded');", &mut db).unwrap();
            save_database(&mut db, &path).unwrap();
        }

        let loaded = open_database(&path, "test".to_string()).expect("open");
        let table = loaded.get_table("docs".to_string()).unwrap();
        let entry = &table.fts_indexes[0];
        // All three rows present — including the zero-token row,
        // which is critical for total_docs honesty in BM25.
        assert_eq!(entry.index.len(), 3);
        // 'embedded' appears in 2 rows after reload.
        let res = entry
            .index
            .query("embedded", &crate::sql::fts::Bm25Params::default());
        assert_eq!(res.len(), 2);

        cleanup(&path);
    }

    #[test]
    fn fts_persistence_round_trips_large_corpus() {
        // Phase 8c — exercise multi-leaf staging. ~500 docs with
        // single-token bodies generates enough cells to overflow a
        // single 4 KiB leaf (each posting cell averages ~8 bytes).
        let path = tmp_path("fts_large_corpus");

        let mut expected_terms: std::collections::BTreeSet<String> =
            std::collections::BTreeSet::new();
        {
            let mut db = Database::new("test".to_string());
            process_command(
                "CREATE TABLE docs (id INTEGER PRIMARY KEY, body TEXT);",
                &mut db,
            )
            .unwrap();
            process_command("CREATE INDEX ix_body ON docs USING fts (body);", &mut db).unwrap();
            // 500 docs, each one a unique term — drives unique-term
            // count up so multiple leaves are required.
            for i in 0..500 {
                let term = format!("term{i:04}");
                process_command(
                    &format!("INSERT INTO docs (body) VALUES ('{term}');"),
                    &mut db,
                )
                .unwrap();
                expected_terms.insert(term);
            }
            save_database(&mut db, &path).unwrap();
        }

        let loaded = open_database(&path, "test".to_string()).expect("open");
        let table = loaded.get_table("docs".to_string()).unwrap();
        let entry = &table.fts_indexes[0];
        assert_eq!(entry.index.len(), 500);

        // Spot-check a handful of terms come back with their original
        // single-row posting list.
        for &i in &[0_i64, 137, 248, 391, 499] {
            let term = format!("term{i:04}");
            let res = entry
                .index
                .query(&term, &crate::sql::fts::Bm25Params::default());
            assert_eq!(res.len(), 1, "term {term} should match exactly 1 row");
            // PrimaryKey rowids start at 1; doc i was inserted at
            // rowid i+1.
            assert_eq!(res[0].0, i + 1);
        }

        cleanup(&path);
    }

    #[test]
    fn delete_then_save_then_reopen_excludes_deleted_node_from_hnsw() {
        // Phase 7d.3 — DELETE marks HNSW dirty; save rebuilds it from
        // current rows + serializes; reopen loads the post-delete graph.
        // After all that, the deleted rowid must NOT come back from a
        // KNN query.
        let path = tmp_path("hnsw_delete_rebuild");
        let mut db = Database::new("test".to_string());
        process_command(
            "CREATE TABLE docs (id INTEGER PRIMARY KEY, e VECTOR(2));",
            &mut db,
        )
        .unwrap();
        for v in &["[1.0, 0.0]", "[2.0, 0.0]", "[3.0, 0.0]", "[4.0, 0.0]"] {
            process_command(&format!("INSERT INTO docs (e) VALUES ({v});"), &mut db).unwrap();
        }
        process_command("CREATE INDEX ix_e ON docs USING hnsw (e);", &mut db).unwrap();

        // Delete row 1 (the closest match to [0.5, 0.0]).
        process_command("DELETE FROM docs WHERE id = 1;", &mut db).unwrap();
        // Confirm it marked dirty.
        let dirty_before_save = db.tables["docs"].hnsw_indexes[0].needs_rebuild;
        assert!(dirty_before_save, "DELETE should mark dirty");

        save_database(&mut db, &path).expect("save");
        // Confirm save cleared the dirty flag.
        let dirty_after_save = db.tables["docs"].hnsw_indexes[0].needs_rebuild;
        assert!(!dirty_after_save, "save should clear dirty");
        drop(db);

        // Reopen, query for the closest match. Row 1 is gone; row 2
        // (id=2, vector [2.0, 0.0]) should now be the nearest.
        let loaded = open_database(&path, "test".to_string()).expect("open");
        let docs = loaded.get_table("docs".to_string()).expect("docs");

        // Row 1 must not appear in any storage anymore.
        assert!(
            !docs.rowids().contains(&1),
            "deleted row 1 should not be in row storage"
        );
        assert_eq!(docs.rowids().len(), 3, "should have 3 surviving rows");

        // The HNSW index must also have shed the deleted node.
        assert_eq!(
            docs.hnsw_indexes[0].index.len(),
            3,
            "HNSW graph should have shed the deleted node"
        );

        cleanup(&path);
    }

    #[test]
    fn round_trip_survives_writes_after_load() {
        let path = tmp_path("after_load");
        save_database(&mut seed_db(), &path).unwrap();

        {
            let mut db = open_database(&path, "test".to_string()).unwrap();
            process_command(
                "INSERT INTO users (name, age) VALUES ('carol', 40);",
                &mut db,
            )
            .unwrap();
            save_database(&mut db, &path).unwrap();
        } // db drops → its exclusive lock releases before we reopen below.

        let db2 = open_database(&path, "test".to_string()).unwrap();
        let users = db2.get_table("users".to_string()).unwrap();
        assert_eq!(users.rowids().len(), 3);

        cleanup(&path);
    }

    #[test]
    fn open_rejects_garbage_file() {
        let path = tmp_path("bad");
        std::fs::write(&path, b"not a sqlrite database, just bytes").unwrap();
        let result = open_database(&path, "x".to_string());
        assert!(result.is_err());
        cleanup(&path);
    }

    #[test]
    fn many_small_rows_spread_across_leaves() {
        let path = tmp_path("many_rows");
        let mut db = Database::new("big".to_string());
        process_command(
            "CREATE TABLE things (id INTEGER PRIMARY KEY, data TEXT);",
            &mut db,
        )
        .unwrap();
        for i in 0..200 {
            let body = "x".repeat(200);
            let q = format!("INSERT INTO things (data) VALUES ('row-{i}-{body}');");
            process_command(&q, &mut db).unwrap();
        }
        save_database(&mut db, &path).unwrap();
        let loaded = open_database(&path, "big".to_string()).unwrap();
        let things = loaded.get_table("things".to_string()).unwrap();
        assert_eq!(things.rowids().len(), 200);
        cleanup(&path);
    }

    #[test]
    fn huge_row_goes_through_overflow() {
        let path = tmp_path("overflow_row");
        let mut db = Database::new("big".to_string());
        process_command(
            "CREATE TABLE docs (id INTEGER PRIMARY KEY, body TEXT);",
            &mut db,
        )
        .unwrap();
        let body = "A".repeat(10_000);
        process_command(
            &format!("INSERT INTO docs (body) VALUES ('{body}');"),
            &mut db,
        )
        .unwrap();
        save_database(&mut db, &path).unwrap();

        let loaded = open_database(&path, "big".to_string()).unwrap();
        let docs = loaded.get_table("docs".to_string()).unwrap();
        let rowids = docs.rowids();
        assert_eq!(rowids.len(), 1);
        let stored = docs.get_value("body", rowids[0]);
        match stored {
            Some(Value::Text(s)) => assert_eq!(s.len(), 10_000),
            other => panic!("expected Text, got {other:?}"),
        }
        cleanup(&path);
    }

    #[test]
    fn create_sql_synthesis_round_trips() {
        // Build a table via CREATE, then verify table_to_create_sql +
        // parse_create_sql reproduce an equivalent column list.
        let mut db = Database::new("x".to_string());
        process_command(
            "CREATE TABLE t (id INTEGER PRIMARY KEY, tag TEXT UNIQUE, note TEXT NOT NULL);",
            &mut db,
        )
        .unwrap();
        let t = db.get_table("t".to_string()).unwrap();
        let sql = table_to_create_sql(t);
        let (name, cols) = parse_create_sql(&sql).unwrap();
        assert_eq!(name, "t");
        assert_eq!(cols.len(), 3);
        assert!(cols[0].is_pk);
        assert!(cols[1].is_unique);
        assert!(cols[2].not_null);
    }

    #[test]
    fn sqlrite_master_is_not_exposed_as_a_user_table() {
        // After open, the public db.tables map should not list the master.
        let path = tmp_path("no_master");
        save_database(&mut seed_db(), &path).unwrap();
        let loaded = open_database(&path, "x".to_string()).unwrap();
        assert!(!loaded.tables.contains_key(MASTER_TABLE_NAME));
        cleanup(&path);
    }

    #[test]
    fn multi_leaf_table_produces_an_interior_root() {
        // 200 fat rows force the table into multiple leaves, which means
        // save_database must build at least one InteriorNode above them.
        // The test verifies the round-trip works and confirms the root is
        // indeed an interior page (not a leaf) by reading the page type
        // directly out of the open pager.
        let path = tmp_path("multi_leaf_interior");
        let mut db = Database::new("big".to_string());
        process_command(
            "CREATE TABLE things (id INTEGER PRIMARY KEY, data TEXT);",
            &mut db,
        )
        .unwrap();
        for i in 0..200 {
            let body = "x".repeat(200);
            let q = format!("INSERT INTO things (data) VALUES ('row-{i}-{body}');");
            process_command(&q, &mut db).unwrap();
        }
        save_database(&mut db, &path).unwrap();

        // Confirm the round-trip preserved all 200 rows.
        let loaded = open_database(&path, "big".to_string()).unwrap();
        let things = loaded.get_table("things".to_string()).unwrap();
        assert_eq!(things.rowids().len(), 200);

        // Peek at `things`'s root page via the pager attached to the
        // loaded DB and check it's an InteriorNode, not a leaf.
        let pager = loaded
            .pager
            .as_ref()
            .expect("loaded DB should have a pager");
        // sqlrite_master's row for `things` holds its root page. Easiest
        // way to find it: walk the leaf chain by using find_leftmost_leaf
        // and then hop one level up. Simpler: read the master, scan for
        // the "things" row, look up rootpage.
        let mut master = build_empty_master_table();
        load_table_rows(pager, &mut master, pager.header().schema_root_page).unwrap();
        let things_root = master
            .rowids()
            .into_iter()
            .find_map(|r| match master.get_value("name", r) {
                Some(Value::Text(s)) if s == "things" => match master.get_value("rootpage", r) {
                    Some(Value::Integer(p)) => Some(p as u32),
                    _ => None,
                },
                _ => None,
            })
            .expect("things should appear in sqlrite_master");
        let root_buf = pager.read_page(things_root).unwrap();
        assert_eq!(
            root_buf[0],
            PageType::InteriorNode as u8,
            "expected a multi-leaf table to have an interior root, got tag {}",
            root_buf[0]
        );

        cleanup(&path);
    }

    #[test]
    fn explicit_index_persists_across_save_and_open() {
        let path = tmp_path("idx_persist");
        let mut db = Database::new("idx".to_string());
        process_command(
            "CREATE TABLE users (id INTEGER PRIMARY KEY, tag TEXT);",
            &mut db,
        )
        .unwrap();
        for i in 1..=5 {
            let tag = if i % 2 == 0 { "odd" } else { "even" };
            process_command(
                &format!("INSERT INTO users (tag) VALUES ('{tag}');"),
                &mut db,
            )
            .unwrap();
        }
        process_command("CREATE INDEX users_tag_idx ON users (tag);", &mut db).unwrap();
        save_database(&mut db, &path).unwrap();

        let loaded = open_database(&path, "idx".to_string()).unwrap();
        let users = loaded.get_table("users".to_string()).unwrap();
        let idx = users
            .index_by_name("users_tag_idx")
            .expect("explicit index should survive save/open");
        assert_eq!(idx.column_name, "tag");
        assert!(!idx.is_unique);
        // 5 rows: rowids 2, 4 are "odd" (i % 2 == 0 when i is 2 or 4) — 2 entries;
        // rowids 1, 3, 5 are "even" (i % 2 != 0) — 3 entries.
        let even_rowids = idx.lookup(&Value::Text("even".into()));
        let odd_rowids = idx.lookup(&Value::Text("odd".into()));
        assert_eq!(even_rowids.len(), 3);
        assert_eq!(odd_rowids.len(), 2);

        cleanup(&path);
    }

    #[test]
    fn auto_indexes_for_unique_columns_survive_save_open() {
        let path = tmp_path("auto_idx_persist");
        let mut db = Database::new("a".to_string());
        process_command(
            "CREATE TABLE users (id INTEGER PRIMARY KEY, email TEXT NOT NULL UNIQUE);",
            &mut db,
        )
        .unwrap();
        process_command("INSERT INTO users (email) VALUES ('a@x');", &mut db).unwrap();
        process_command("INSERT INTO users (email) VALUES ('b@x');", &mut db).unwrap();
        save_database(&mut db, &path).unwrap();

        let loaded = open_database(&path, "a".to_string()).unwrap();
        let users = loaded.get_table("users".to_string()).unwrap();
        // Every UNIQUE column auto-creates an index; the load path populated
        // it from the persisted entries.
        let auto_name = SecondaryIndex::auto_name("users", "email");
        let idx = users
            .index_by_name(&auto_name)
            .expect("auto index should be restored");
        assert!(idx.is_unique);
        assert_eq!(idx.lookup(&Value::Text("a@x".into())).len(), 1);
        assert_eq!(idx.lookup(&Value::Text("b@x".into())).len(), 1);

        cleanup(&path);
    }

    /// SQLR-1 — `CREATE INDEX` on a wide table must round-trip when the
    /// index B-tree grows past one leaf and needs an interior level.
    /// Before the fix, the post-DDL auto-save panicked with
    /// `Internal("unknown paged-entry kind tag 0x4 …")` because a
    /// table-cell decoder was being run against an index leaf
    /// (`KIND_INDEX = 0x04`).
    ///
    /// 5 000 rows mirror the original repro from the issue and exceed
    /// every leaf-fanout cliff for the small `(rowid, value)` cells in
    /// a TEXT-keyed secondary index.
    #[test]
    fn secondary_index_with_interior_level_round_trips() {
        let path = tmp_path("sqlr1_wide_index");
        let mut db = Database::new("idx".to_string());
        db.source_path = Some(path.clone());

        process_command(
            "CREATE TABLE bloat (id INTEGER PRIMARY KEY, payload TEXT);",
            &mut db,
        )
        .unwrap();
        // BEGIN/COMMIT collapses 5 000 inserts into one save (matches
        // `auto_vacuum_setup` and the issue's repro shape).
        process_command("BEGIN;", &mut db).unwrap();
        for i in 0..5000 {
            process_command(
                &format!("INSERT INTO bloat (payload) VALUES ('p-{i:08}');"),
                &mut db,
            )
            .unwrap();
        }
        process_command("COMMIT;", &mut db).unwrap();

        // The DDL that used to panic.
        process_command("CREATE INDEX idx_p ON bloat (payload);", &mut db).unwrap();

        // Reopen and verify lookups, plus that the index tree actually
        // grew an interior layer (otherwise this test wouldn't cover the
        // regression).
        drop(db);
        let loaded = open_database(&path, "idx".to_string()).unwrap();
        let bloat = loaded.get_table("bloat".to_string()).unwrap();
        let idx = bloat
            .index_by_name("idx_p")
            .expect("idx_p should survive close/reopen");
        assert!(!idx.is_unique);

        // Spot-check the keyspace: first, middle, last value each map
        // back to exactly the row that carried them.
        for &(probe_i, expected_rowid) in &[(0i64, 1i64), (2500, 2501), (4999, 5000)] {
            let value = Value::Text(format!("p-{probe_i:08}"));
            let hits = idx.lookup(&value);
            assert_eq!(
                hits,
                vec![expected_rowid],
                "lookup({value:?}) should yield rowid {expected_rowid}",
            );
        }

        // Confirm the index tree is multi-level (the regression's
        // necessary condition) — root must be an `InteriorNode` and
        // `find_leftmost_leaf` must reach a `TableLeaf` through it.
        let pager = loaded.pager.as_ref().unwrap();
        let mut master = build_empty_master_table();
        load_table_rows(pager, &mut master, pager.header().schema_root_page).unwrap();
        let idx_root = master
            .rowids()
            .into_iter()
            .find_map(
                |r| match (master.get_value("name", r), master.get_value("type", r)) {
                    (Some(Value::Text(name)), Some(Value::Text(kind)))
                        if name == "idx_p" && kind == "index" =>
                    {
                        match master.get_value("rootpage", r) {
                            Some(Value::Integer(p)) => Some(p as u32),
                            _ => None,
                        }
                    }
                    _ => None,
                },
            )
            .expect("idx_p should appear in sqlrite_master");
        let root_buf = pager.read_page(idx_root).unwrap();
        assert_eq!(
            root_buf[0],
            PageType::InteriorNode as u8,
            "5 000-entry index must have an interior root — without one this test wouldn't cover SQLR-1",
        );
        let leaf = find_leftmost_leaf(pager, idx_root).unwrap();
        let leaf_buf = pager.read_page(leaf).unwrap();
        assert_eq!(leaf_buf[0], PageType::TableLeaf as u8);

        cleanup(&path);
    }

    /// SQLR-1 follow-on — the page-recycling path between two large
    /// versions of the same index name must not corrupt cell decoding.
    /// `DROP INDEX` returns its pages to the freelist; the next
    /// `CREATE INDEX` is free to reuse them. If the allocator hands an
    /// old index leaf to a *table* without zeroing it, an upstream
    /// table walk would see KIND_INDEX cells and panic.
    #[test]
    fn drop_then_recreate_wide_index_does_not_panic() {
        let path = tmp_path("sqlr1_drop_recreate");
        let mut db = Database::new("idx".to_string());
        db.source_path = Some(path.clone());

        process_command(
            "CREATE TABLE bloat (id INTEGER PRIMARY KEY, payload TEXT);",
            &mut db,
        )
        .unwrap();
        process_command("BEGIN;", &mut db).unwrap();
        for i in 0..5000 {
            process_command(
                &format!("INSERT INTO bloat (payload) VALUES ('p-{i:08}');"),
                &mut db,
            )
            .unwrap();
        }
        process_command("COMMIT;", &mut db).unwrap();

        process_command("CREATE INDEX idx_p ON bloat (payload);", &mut db).unwrap();
        process_command("DROP INDEX idx_p;", &mut db).unwrap();
        // Recreate from scratch — exercises the recycle path.
        process_command("CREATE INDEX idx_p ON bloat (payload);", &mut db).unwrap();

        drop(db);
        let loaded = open_database(&path, "idx".to_string()).unwrap();
        let bloat = loaded.get_table("bloat".to_string()).unwrap();
        let idx = bloat
            .index_by_name("idx_p")
            .expect("idx_p should survive drop+recreate+reopen");
        assert_eq!(
            idx.lookup(&Value::Text("p-00002500".into())),
            vec![2501],
            "post-recycle lookup must still resolve correctly",
        );

        cleanup(&path);
    }

    #[test]
    fn deep_tree_round_trips() {
        // Force a 3-level tree by bypassing process_command (which prints
        // the full table on every INSERT, making large bulk loads O(N^2)
        // in I/O). We build the Table directly via restore_row.
        use crate::sql::db::table::Column as TableColumn;

        let path = tmp_path("deep_tree");
        let mut db = Database::new("deep".to_string());
        let columns = vec![
            TableColumn::new("id".into(), "integer".into(), true, true, true),
            TableColumn::new("s".into(), "text".into(), false, true, false),
        ];
        let mut table = build_empty_table("t", columns, 0);
        // ~900-byte rows → ~4 rows per leaf. 6000 rows → ~1500 leaves,
        // which with interior fanout ~400 needs 2 interior levels (3-level
        // tree total, counting leaves).
        for i in 1..=6_000i64 {
            let body = "q".repeat(900);
            table
                .restore_row(
                    i,
                    vec![
                        Some(Value::Integer(i)),
                        Some(Value::Text(format!("r-{i}-{body}"))),
                    ],
                )
                .unwrap();
        }
        db.tables.insert("t".to_string(), table);
        save_database(&mut db, &path).unwrap();

        let loaded = open_database(&path, "deep".to_string()).unwrap();
        let t = loaded.get_table("t".to_string()).unwrap();
        assert_eq!(t.rowids().len(), 6_000);

        // Confirm the tree actually grew past 2 levels — i.e., the root's
        // leftmost child is itself an interior page, not a leaf.
        let pager = loaded.pager.as_ref().unwrap();
        let mut master = build_empty_master_table();
        load_table_rows(pager, &mut master, pager.header().schema_root_page).unwrap();
        let t_root = master
            .rowids()
            .into_iter()
            .find_map(|r| match master.get_value("name", r) {
                Some(Value::Text(s)) if s == "t" => match master.get_value("rootpage", r) {
                    Some(Value::Integer(p)) => Some(p as u32),
                    _ => None,
                },
                _ => None,
            })
            .expect("t in sqlrite_master");
        let root_buf = pager.read_page(t_root).unwrap();
        assert_eq!(root_buf[0], PageType::InteriorNode as u8);
        let root_payload: &[u8; PAYLOAD_PER_PAGE] =
            (&root_buf[PAGE_HEADER_SIZE..]).try_into().unwrap();
        let root_interior = InteriorPage::from_bytes(root_payload);
        let child = root_interior.leftmost_child().unwrap();
        let child_buf = pager.read_page(child).unwrap();
        assert_eq!(
            child_buf[0],
            PageType::InteriorNode as u8,
            "expected 3-level tree: root's leftmost child should also be InteriorNode",
        );

        cleanup(&path);
    }

    #[test]
    fn alter_rename_table_survives_save_and_reopen() {
        let path = tmp_path("alter_rename_table_roundtrip");
        let mut db = seed_db();
        save_database(&mut db, &path).expect("save");

        process_command("ALTER TABLE users RENAME TO members;", &mut db).expect("rename");
        save_database(&mut db, &path).expect("save after rename");

        let loaded = open_database(&path, "t".to_string()).expect("reopen");
        assert!(!loaded.contains_table("users".to_string()));
        assert!(loaded.contains_table("members".to_string()));
        let members = loaded.get_table("members".to_string()).unwrap();
        assert_eq!(members.rowids().len(), 2, "rows should survive");
        // Auto-indexes followed the rename.
        assert!(
            members
                .index_by_name("sqlrite_autoindex_members_id")
                .is_some()
        );
        assert!(
            members
                .index_by_name("sqlrite_autoindex_members_name")
                .is_some()
        );

        cleanup(&path);
    }

    #[test]
    fn alter_rename_column_survives_save_and_reopen() {
        let path = tmp_path("alter_rename_col_roundtrip");
        let mut db = seed_db();
        save_database(&mut db, &path).expect("save");

        process_command(
            "ALTER TABLE users RENAME COLUMN name TO full_name;",
            &mut db,
        )
        .expect("rename column");
        save_database(&mut db, &path).expect("save after rename");

        let loaded = open_database(&path, "t".to_string()).expect("reopen");
        let users = loaded.get_table("users".to_string()).unwrap();
        assert!(users.contains_column("full_name".to_string()));
        assert!(!users.contains_column("name".to_string()));
        // Verify a row's value survived the rename round-trip.
        let alice_rowid = users
            .rowids()
            .into_iter()
            .find(|r| users.get_value("full_name", *r) == Some(Value::Text("alice".to_string())))
            .expect("alice row should be findable under renamed column");
        assert_eq!(
            users.get_value("full_name", alice_rowid),
            Some(Value::Text("alice".to_string()))
        );

        cleanup(&path);
    }

    #[test]
    fn alter_add_column_with_default_survives_save_and_reopen() {
        let path = tmp_path("alter_add_default_roundtrip");
        let mut db = seed_db();
        save_database(&mut db, &path).expect("save");

        process_command(
            "ALTER TABLE users ADD COLUMN status TEXT DEFAULT 'active';",
            &mut db,
        )
        .expect("add column");
        save_database(&mut db, &path).expect("save after add");

        let loaded = open_database(&path, "t".to_string()).expect("reopen");
        let users = loaded.get_table("users".to_string()).unwrap();
        assert!(users.contains_column("status".to_string()));
        for rowid in users.rowids() {
            assert_eq!(
                users.get_value("status", rowid),
                Some(Value::Text("active".to_string())),
                "backfilled default should round-trip for rowid {rowid}"
            );
        }
        // The DEFAULT clause itself should still be on the column metadata
        // so a subsequent INSERT picks it up.
        let status_col = users
            .columns
            .iter()
            .find(|c| c.column_name == "status")
            .unwrap();
        assert_eq!(status_col.default, Some(Value::Text("active".to_string())));

        cleanup(&path);
    }

    #[test]
    fn alter_drop_column_survives_save_and_reopen() {
        let path = tmp_path("alter_drop_col_roundtrip");
        let mut db = seed_db();
        save_database(&mut db, &path).expect("save");

        process_command("ALTER TABLE users DROP COLUMN age;", &mut db).expect("drop column");
        save_database(&mut db, &path).expect("save after drop");

        let loaded = open_database(&path, "t".to_string()).expect("reopen");
        let users = loaded.get_table("users".to_string()).unwrap();
        assert!(!users.contains_column("age".to_string()));
        assert!(users.contains_column("name".to_string()));

        cleanup(&path);
    }

    #[test]
    fn drop_table_survives_save_and_reopen() {
        let path = tmp_path("drop_table_roundtrip");
        let mut db = seed_db();
        save_database(&mut db, &path).expect("save");

        // Verify both tables landed.
        {
            let loaded = open_database(&path, "t".to_string()).expect("open");
            assert!(loaded.contains_table("users".to_string()));
            assert!(loaded.contains_table("notes".to_string()));
        }

        process_command("DROP TABLE users;", &mut db).expect("drop users");
        save_database(&mut db, &path).expect("save after drop");

        let loaded = open_database(&path, "t".to_string()).expect("reopen");
        assert!(
            !loaded.contains_table("users".to_string()),
            "dropped table should not resurface on reopen"
        );
        assert!(
            loaded.contains_table("notes".to_string()),
            "untouched table should survive"
        );

        cleanup(&path);
    }

    #[test]
    fn drop_index_survives_save_and_reopen() {
        let path = tmp_path("drop_index_roundtrip");
        let mut db = Database::new("t".to_string());
        process_command(
            "CREATE TABLE notes (id INTEGER PRIMARY KEY, body TEXT);",
            &mut db,
        )
        .unwrap();
        process_command("CREATE INDEX notes_body_idx ON notes (body);", &mut db).unwrap();
        save_database(&mut db, &path).expect("save");

        process_command("DROP INDEX notes_body_idx;", &mut db).unwrap();
        save_database(&mut db, &path).expect("save after drop");

        let loaded = open_database(&path, "t".to_string()).expect("reopen");
        let notes = loaded.get_table("notes".to_string()).unwrap();
        assert!(
            notes.index_by_name("notes_body_idx").is_none(),
            "dropped index should not resurface on reopen"
        );
        // The auto-index for the PK should still be there.
        assert!(notes.index_by_name("sqlrite_autoindex_notes_id").is_some());

        cleanup(&path);
    }

    #[test]
    fn default_clause_survives_save_and_reopen() {
        let path = tmp_path("default_roundtrip");
        let mut db = Database::new("t".to_string());

        process_command(
            "CREATE TABLE users (id INTEGER PRIMARY KEY, status TEXT DEFAULT 'active', score INTEGER DEFAULT 0);",
            &mut db,
        )
        .unwrap();
        save_database(&mut db, &path).expect("save");

        let mut loaded = open_database(&path, "t".to_string()).expect("open");

        // The reloaded column metadata should still carry the DEFAULT.
        let users = loaded.get_table("users".to_string()).expect("users table");
        let status_col = users
            .columns
            .iter()
            .find(|c| c.column_name == "status")
            .expect("status column");
        assert_eq!(
            status_col.default,
            Some(Value::Text("active".to_string())),
            "DEFAULT 'active' should round-trip"
        );
        let score_col = users
            .columns
            .iter()
            .find(|c| c.column_name == "score")
            .expect("score column");
        assert_eq!(
            score_col.default,
            Some(Value::Integer(0)),
            "DEFAULT 0 should round-trip"
        );

        // Now exercise the runtime path: an INSERT that omits both DEFAULT
        // columns should pick them up from the reloaded schema.
        process_command("INSERT INTO users (id) VALUES (1);", &mut loaded).unwrap();
        let users = loaded.get_table("users".to_string()).unwrap();
        assert_eq!(
            users.get_value("status", 1),
            Some(Value::Text("active".to_string()))
        );
        assert_eq!(users.get_value("score", 1), Some(Value::Integer(0)));

        cleanup(&path);
    }

    // ---------------------------------------------------------------------
    // SQLR-6 — free-list + VACUUM tests
    // ---------------------------------------------------------------------

    /// Drop a table; subsequent CREATE TABLE should reuse the freed pages
    /// rather than extending the file. The page_count after drop+create
    /// should be at most what it was after the original two tables —
    /// proving the new table landed on freelist pages.
    #[test]
    fn drop_table_freelist_persists_pages_for_reuse() {
        let path = tmp_path("freelist_reuse");
        let mut db = seed_db();
        db.source_path = Some(path.clone());
        save_database(&mut db, &path).expect("save");
        let pages_two_tables = db.pager.as_ref().unwrap().header().page_count;

        // Drop one table; its pages go on the freelist.
        process_command("DROP TABLE users;", &mut db).expect("drop users");
        let pages_after_drop = db.pager.as_ref().unwrap().header().page_count;
        assert_eq!(
            pages_after_drop, pages_two_tables,
            "page_count should not shrink on drop — the freed pages persist on the freelist"
        );
        let head_after_drop = db.pager.as_ref().unwrap().header().freelist_head;
        assert!(
            head_after_drop != 0,
            "freelist_head must be non-zero after drop"
        );

        // Re-create a similar-shaped table; should reuse freelist pages.
        process_command(
            "CREATE TABLE accounts (id INTEGER PRIMARY KEY, label TEXT NOT NULL UNIQUE);",
            &mut db,
        )
        .expect("create accounts");
        process_command("INSERT INTO accounts (label) VALUES ('a');", &mut db).unwrap();
        process_command("INSERT INTO accounts (label) VALUES ('b');", &mut db).unwrap();
        let pages_after_create = db.pager.as_ref().unwrap().header().page_count;
        assert!(
            pages_after_create <= pages_two_tables + 2,
            "creating a similar-sized table after a drop should mostly draw from the \
             freelist, not extend the file (got {pages_after_create} > {pages_two_tables} + 2)"
        );

        cleanup(&path);
    }

    /// `VACUUM;` after a drop must shrink the file and clear the freelist.
    #[test]
    fn drop_then_vacuum_shrinks_file() {
        let path = tmp_path("vacuum_shrinks");
        let mut db = seed_db();
        db.source_path = Some(path.clone());
        // Add a few more rows to make the dropped table bigger.
        for i in 0..20 {
            process_command(
                &format!("INSERT INTO users (name, age) VALUES ('user{i}', {i});"),
                &mut db,
            )
            .unwrap();
        }
        save_database(&mut db, &path).expect("save");

        process_command("DROP TABLE users;", &mut db).expect("drop");
        let size_before_vacuum = std::fs::metadata(&path).unwrap().len();
        let pages_before_vacuum = db.pager.as_ref().unwrap().header().page_count;
        let head_before = db.pager.as_ref().unwrap().header().freelist_head;
        assert!(head_before != 0, "drop should populate the freelist");

        // VACUUM (via process_command) checkpoints internally so the
        // file actually shrinks on disk before we observe its size.
        process_command("VACUUM;", &mut db).expect("vacuum");

        let size_after = std::fs::metadata(&path).unwrap().len();
        let pages_after = db.pager.as_ref().unwrap().header().page_count;
        let head_after = db.pager.as_ref().unwrap().header().freelist_head;
        assert!(
            pages_after < pages_before_vacuum,
            "VACUUM must reduce page_count: was {pages_before_vacuum}, now {pages_after}"
        );
        assert_eq!(head_after, 0, "VACUUM must clear the freelist");
        assert!(
            size_after < size_before_vacuum,
            "VACUUM must shrink the file on disk: was {size_before_vacuum} bytes, now {size_after}"
        );

        cleanup(&path);
    }

    /// VACUUM on a non-empty multi-table DB must not lose any rows.
    #[test]
    fn vacuum_round_trips_data() {
        let path = tmp_path("vacuum_round_trip");
        let mut db = seed_db();
        db.source_path = Some(path.clone());
        save_database(&mut db, &path).expect("save");
        process_command("VACUUM;", &mut db).expect("vacuum");

        // Re-open from disk to make sure the on-disk catalog round-trips.
        drop(db);
        let loaded = open_database(&path, "t".to_string()).expect("reopen after vacuum");
        assert!(loaded.contains_table("users".to_string()));
        assert!(loaded.contains_table("notes".to_string()));
        let users = loaded.get_table("users".to_string()).unwrap();
        // seed_db inserts two users.
        assert_eq!(users.rowids().len(), 2);

        cleanup(&path);
    }

    /// Format version is bumped to v6 only after a save that creates a
    /// non-empty freelist. VACUUM clears the freelist but doesn't
    /// downgrade — v6 is a strict superset, so once at v6 we stay.
    #[test]
    fn freelist_format_version_promotion() {
        use crate::sql::pager::header::{FORMAT_VERSION_BASELINE, FORMAT_VERSION_V6};
        let path = tmp_path("v6_promotion");
        let mut db = seed_db();
        db.source_path = Some(path.clone());
        save_database(&mut db, &path).expect("save");
        let v_after_save = db.pager.as_ref().unwrap().header().format_version;
        assert_eq!(
            v_after_save, FORMAT_VERSION_BASELINE,
            "fresh DB without drops should stay at the baseline version"
        );

        process_command("DROP TABLE users;", &mut db).expect("drop");
        let v_after_drop = db.pager.as_ref().unwrap().header().format_version;
        assert_eq!(
            v_after_drop, FORMAT_VERSION_V6,
            "first save with a non-empty freelist must promote to V6"
        );

        process_command("VACUUM;", &mut db).expect("vacuum");
        let v_after_vacuum = db.pager.as_ref().unwrap().header().format_version;
        assert_eq!(
            v_after_vacuum, FORMAT_VERSION_V6,
            "VACUUM must not downgrade — V6 is a strict superset"
        );

        cleanup(&path);
    }

    /// Freelist persists across reopen: drop, save, close, reopen,
    /// confirm the next CREATE TABLE re-uses pages from the persisted
    /// freelist (rather than extending the file).
    #[test]
    fn freelist_round_trip_through_reopen() {
        let path = tmp_path("freelist_reopen");
        let pages_two_tables;
        {
            let mut db = seed_db();
            db.source_path = Some(path.clone());
            save_database(&mut db, &path).expect("save");
            pages_two_tables = db.pager.as_ref().unwrap().header().page_count;
            process_command("DROP TABLE users;", &mut db).expect("drop");
            let head = db.pager.as_ref().unwrap().header().freelist_head;
            assert!(head != 0, "drop must populate the freelist");
        }

        // Reopen from disk — the freelist must come back.
        let mut db = open_database(&path, "t".to_string()).expect("reopen");
        assert!(
            db.pager.as_ref().unwrap().header().freelist_head != 0,
            "freelist_head must survive close/reopen"
        );

        process_command(
            "CREATE TABLE accounts (id INTEGER PRIMARY KEY, label TEXT NOT NULL UNIQUE);",
            &mut db,
        )
        .expect("create accounts");
        process_command("INSERT INTO accounts (label) VALUES ('reopened');", &mut db).unwrap();
        let pages_after_create = db.pager.as_ref().unwrap().header().page_count;
        assert!(
            pages_after_create <= pages_two_tables + 2,
            "post-reopen create should reuse freelist (got {pages_after_create} > \
             {pages_two_tables} + 2 — file extended instead of reusing)"
        );

        cleanup(&path);
    }

    /// VACUUM inside an explicit transaction must error before touching the
    /// disk. `BEGIN; VACUUM;` is the documented rejection path.
    #[test]
    fn vacuum_inside_transaction_is_rejected() {
        let path = tmp_path("vacuum_txn");
        let mut db = seed_db();
        db.source_path = Some(path.clone());
        save_database(&mut db, &path).expect("save");

        process_command("BEGIN;", &mut db).expect("begin");
        let err = process_command("VACUUM;", &mut db).unwrap_err();
        assert!(
            format!("{err}").contains("VACUUM cannot run inside a transaction"),
            "expected in-transaction rejection, got: {err}"
        );
        // Roll back to leave the DB in a clean state.
        process_command("ROLLBACK;", &mut db).unwrap();
        cleanup(&path);
    }

    /// VACUUM on an in-memory database is a documented no-op.
    #[test]
    fn vacuum_on_in_memory_database_is_noop() {
        let mut db = Database::new("mem".to_string());
        process_command("CREATE TABLE t (id INTEGER PRIMARY KEY);", &mut db).unwrap();
        let out = process_command("VACUUM;", &mut db).expect("vacuum no-op");
        assert!(
            out.to_lowercase().contains("no-op") || out.to_lowercase().contains("in-memory"),
            "expected no-op message for in-memory VACUUM, got: {out}"
        );
    }

    /// Untouched tables shouldn't write any pages on the save that
    /// follows a DROP of an unrelated table. Confirms the per-table
    /// preferred pool keeps page numbers stable so the diff pager skips
    /// every byte-identical leaf.
    #[test]
    fn unchanged_table_pages_skip_diff_after_unrelated_drop() {
        // Need three tables so dropping one in the middle still leaves
        // an "unrelated" alphabetical neighbour. Layout pre-drop (sorted):
        //   accounts, notes, users
        // Drop `notes`. `accounts` and `users` should keep their pages.
        let path = tmp_path("diff_after_drop");
        let mut db = Database::new("t".to_string());
        db.source_path = Some(path.clone());
        process_command(
            "CREATE TABLE accounts (id INTEGER PRIMARY KEY, label TEXT);",
            &mut db,
        )
        .unwrap();
        process_command(
            "CREATE TABLE notes (id INTEGER PRIMARY KEY, body TEXT);",
            &mut db,
        )
        .unwrap();
        process_command(
            "CREATE TABLE users (id INTEGER PRIMARY KEY, name TEXT);",
            &mut db,
        )
        .unwrap();
        for i in 0..5 {
            process_command(
                &format!("INSERT INTO accounts (label) VALUES ('a{i}');"),
                &mut db,
            )
            .unwrap();
            process_command(
                &format!("INSERT INTO notes (body) VALUES ('n{i}');"),
                &mut db,
            )
            .unwrap();
            process_command(
                &format!("INSERT INTO users (name) VALUES ('u{i}');"),
                &mut db,
            )
            .unwrap();
        }
        save_database(&mut db, &path).expect("baseline save");

        // Capture page bytes for `accounts` and `users` so we can
        // verify they don't change.
        let pager = db.pager.as_ref().unwrap();
        let acc_root = read_old_rootpages(pager, pager.header().schema_root_page)
            .unwrap()
            .get(&("table".to_string(), "accounts".to_string()))
            .copied()
            .unwrap();
        let users_root = read_old_rootpages(pager, pager.header().schema_root_page)
            .unwrap()
            .get(&("table".to_string(), "users".to_string()))
            .copied()
            .unwrap();
        let acc_bytes_before: Vec<u8> = pager.read_page(acc_root).unwrap().to_vec();
        let users_bytes_before: Vec<u8> = pager.read_page(users_root).unwrap().to_vec();

        // Drop the middle table.
        process_command("DROP TABLE notes;", &mut db).expect("drop notes");

        let pager = db.pager.as_ref().unwrap();
        // `accounts` and `users` should still live at the same pages
        // with byte-identical content.
        let acc_after = pager.read_page(acc_root).unwrap();
        let users_after = pager.read_page(users_root).unwrap();
        assert_eq!(
            &acc_after[..],
            &acc_bytes_before[..],
            "accounts root page must not be rewritten when an unrelated table is dropped"
        );
        assert_eq!(
            &users_after[..],
            &users_bytes_before[..],
            "users root page must not be rewritten when an unrelated table is dropped"
        );

        cleanup(&path);
    }

    // ---- SQLR-10: auto-VACUUM trigger after page-releasing DDL ----

    /// Builds a file-backed DB with one small "keep" table and one
    /// large "bloat" table, sized so the post-drop freelist will
    /// comfortably cross the default 25% threshold and the
    /// `MIN_PAGES_FOR_AUTO_VACUUM` floor (16 pages). Used by the
    /// auto-VACUUM happy-path tests.
    fn auto_vacuum_setup(path: &std::path::Path) -> Database {
        let mut db = Database::new("av".to_string());
        db.source_path = Some(path.to_path_buf());
        process_command(
            "CREATE TABLE keep (id INTEGER PRIMARY KEY, n INTEGER);",
            &mut db,
        )
        .unwrap();
        process_command("INSERT INTO keep (n) VALUES (1);", &mut db).unwrap();
        process_command(
            "CREATE TABLE bloat (id INTEGER PRIMARY KEY, payload TEXT);",
            &mut db,
        )
        .unwrap();
        // Wrap the bulk insert in a transaction so we pay one save at
        // COMMIT instead of 5000 round-trips through auto-save.
        process_command("BEGIN;", &mut db).unwrap();
        for i in 0..5000 {
            process_command(
                &format!("INSERT INTO bloat (payload) VALUES ('p-{i:08}');"),
                &mut db,
            )
            .unwrap();
        }
        process_command("COMMIT;", &mut db).unwrap();
        db
    }

    /// Default threshold (0.25) is engaged for fresh `Database`s and
    /// fires when a `DROP TABLE` orphans enough pages — file shrinks
    /// without anyone calling `VACUUM;`.
    #[test]
    fn auto_vacuum_default_threshold_triggers_on_drop_table() {
        let path = tmp_path("av_default_drop_table");
        let mut db = auto_vacuum_setup(&path);
        // Sanity: setup respects the shipped default.
        assert_eq!(db.auto_vacuum_threshold(), Some(0.25));

        // Checkpoint before measuring `size_before` so the bloat actually
        // lives in the main file and not just the WAL — otherwise
        // `size_before` is the bare 2-page header and any post-vacuum
        // checkpoint will look like the file *grew*.
        if let Some(p) = db.pager.as_mut() {
            let _ = p.checkpoint();
        }
        let pages_before = db.pager.as_ref().unwrap().header().page_count;
        let size_before = std::fs::metadata(&path).unwrap().len();
        assert!(
            pages_before >= MIN_PAGES_FOR_AUTO_VACUUM,
            "setup should produce >= MIN_PAGES_FOR_AUTO_VACUUM ({MIN_PAGES_FOR_AUTO_VACUUM}) \
             pages so the floor doesn't suppress the trigger; got {pages_before}"
        );

        // Drop the bloat table — freelist should pass 25% of page_count
        // and the auto-VACUUM hook should compact in place. Note: no
        // explicit `VACUUM;` statement is issued.
        process_command("DROP TABLE bloat;", &mut db).expect("drop");

        let pages_after = db.pager.as_ref().unwrap().header().page_count;
        let head_after = db.pager.as_ref().unwrap().header().freelist_head;
        // Second checkpoint so the post-vacuum file shrinks on disk
        // (auto-VACUUM stages the compact through WAL just like manual
        // VACUUM does).
        if let Some(p) = db.pager.as_mut() {
            let _ = p.checkpoint();
        }
        let size_after = std::fs::metadata(&path).unwrap().len();

        assert!(
            pages_after < pages_before,
            "auto-VACUUM must reduce page_count: was {pages_before}, now {pages_after}"
        );
        assert_eq!(head_after, 0, "auto-VACUUM must clear the freelist");
        assert!(
            size_after < size_before,
            "auto-VACUUM must shrink the file on disk: was {size_before}, now {size_after}"
        );

        cleanup(&path);
    }

    /// Setting the threshold to `None` disables the trigger entirely:
    /// the same workload that shrinks under the default leaves the file
    /// at its high-water mark.
    #[test]
    fn auto_vacuum_disabled_keeps_file_at_hwm() {
        let path = tmp_path("av_disabled");
        let mut db = auto_vacuum_setup(&path);
        db.set_auto_vacuum_threshold(None).expect("disable");
        assert_eq!(db.auto_vacuum_threshold(), None);

        let pages_before = db.pager.as_ref().unwrap().header().page_count;

        process_command("DROP TABLE bloat;", &mut db).expect("drop");

        let pages_after = db.pager.as_ref().unwrap().header().page_count;
        let head_after = db.pager.as_ref().unwrap().header().freelist_head;
        assert_eq!(
            pages_after, pages_before,
            "with auto-VACUUM disabled, drop must keep page_count at the HWM"
        );
        assert!(
            head_after != 0,
            "drop must still populate the freelist (manual VACUUM would be needed to reclaim)"
        );

        cleanup(&path);
    }

    /// `DROP INDEX` is the second of three page-releasing DDL paths
    /// covered by SQLR-10. We bloat the freelist via a separate
    /// `DROP TABLE` first (with auto-VACUUM disabled so it doesn't
    /// compact early), then re-arm the trigger and drop a small index
    /// — the cumulative freelist crosses 25% on the index drop and
    /// auto-VACUUM fires.
    ///
    /// The detour around bloat is necessary because building a
    /// secondary index on a 5000-row column would need multi-level
    /// interior nodes, and the cell-decoder's interior-page support
    /// is a separate work item from SQLR-10.
    #[test]
    fn auto_vacuum_triggers_on_drop_index() {
        let path = tmp_path("av_drop_index");
        let mut db = auto_vacuum_setup(&path);

        // Phase 1: drop the bloat table with auto-VACUUM disabled so
        // its pages land on the freelist without being reclaimed.
        db.set_auto_vacuum_threshold(None).expect("disable");
        process_command("DROP TABLE bloat;", &mut db).expect("drop bloat");
        let pages_after_bloat_drop = db.pager.as_ref().unwrap().header().page_count;
        let head_after_bloat_drop = db.pager.as_ref().unwrap().header().freelist_head;
        assert!(
            head_after_bloat_drop != 0,
            "bloat drop must populate the freelist (else later index drop won't trip the threshold)"
        );

        // Phase 2: a small index on the surviving `keep` table. The
        // index reuses one page from the freelist (which is fine —
        // freelist still holds plenty more).
        process_command("CREATE INDEX idx_keep_n ON keep (n);", &mut db).expect("create idx");

        // Phase 3: re-arm the trigger and drop the index. The freelist
        // is already heavily populated from phase 1; this drop just
        // adds the index page on top, keeping the ratio well above
        // 25%, so auto-VACUUM should fire.
        db.set_auto_vacuum_threshold(Some(0.25)).expect("re-arm");
        process_command("DROP INDEX idx_keep_n;", &mut db).expect("drop index");

        let pages_after = db.pager.as_ref().unwrap().header().page_count;
        let head_after = db.pager.as_ref().unwrap().header().freelist_head;
        assert!(
            pages_after < pages_after_bloat_drop,
            "DROP INDEX should fire auto-VACUUM and reduce page_count: \
             was {pages_after_bloat_drop}, now {pages_after}"
        );
        assert_eq!(
            head_after, 0,
            "auto-VACUUM after DROP INDEX must clear the freelist"
        );

        cleanup(&path);
    }

    /// `ALTER TABLE … DROP COLUMN` releases pages too — the third path
    /// the SQLR-10 trigger covers.
    #[test]
    fn auto_vacuum_triggers_on_alter_drop_column() {
        let path = tmp_path("av_alter_drop_col");
        let mut db = auto_vacuum_setup(&path);
        let pages_before = db.pager.as_ref().unwrap().header().page_count;

        // Drop the wide `payload` column — this rewrites every row in
        // `bloat` without the column, so the old leaf pages get freed.
        process_command("ALTER TABLE bloat DROP COLUMN payload;", &mut db).expect("alter drop");

        let pages_after = db.pager.as_ref().unwrap().header().page_count;
        assert!(
            pages_after < pages_before,
            "ALTER TABLE DROP COLUMN should fire auto-VACUUM and reduce page_count: \
             was {pages_before}, now {pages_after}"
        );
        assert_eq!(db.pager.as_ref().unwrap().header().freelist_head, 0);

        cleanup(&path);
    }

    /// A high threshold (0.99) suppresses the trigger when the freelist
    /// ratio is well below it — the file stays at HWM.
    #[test]
    fn auto_vacuum_skips_below_threshold() {
        let path = tmp_path("av_below_threshold");
        let mut db = auto_vacuum_setup(&path);
        db.set_auto_vacuum_threshold(Some(0.99)).expect("set");

        let pages_before = db.pager.as_ref().unwrap().header().page_count;

        process_command("DROP TABLE bloat;", &mut db).expect("drop");

        let pages_after = db.pager.as_ref().unwrap().header().page_count;
        assert_eq!(
            pages_after, pages_before,
            "freelist ratio after a single drop is far below 0.99 — \
             page_count must stay at the HWM"
        );
        assert!(
            db.pager.as_ref().unwrap().header().freelist_head != 0,
            "drop must still populate the freelist"
        );

        cleanup(&path);
    }

    /// Inside an explicit transaction, the page-releasing DDL doesn't
    /// flush to disk yet — the freelist isn't accurate, so the trigger
    /// must skip. The compact would also publish in-flight work out of
    /// band, which is exactly what the manual `VACUUM;` rejection
    /// inside a txn already prevents.
    #[test]
    fn auto_vacuum_skips_inside_transaction() {
        let path = tmp_path("av_in_txn");
        let mut db = auto_vacuum_setup(&path);
        let pages_before = db.pager.as_ref().unwrap().header().page_count;

        process_command("BEGIN;", &mut db).expect("begin");
        process_command("DROP TABLE bloat;", &mut db).expect("drop in txn");
        // Mid-transaction: no save has occurred, so the on-disk
        // freelist_head must be unchanged and page_count must not have
        // shifted from a sneaky compact.
        let pages_mid = db.pager.as_ref().unwrap().header().page_count;
        assert_eq!(
            pages_mid, pages_before,
            "auto-VACUUM must not fire mid-transaction"
        );

        process_command("ROLLBACK;", &mut db).expect("rollback");
        cleanup(&path);
    }

    /// Tiny databases (under `MIN_PAGES_FOR_AUTO_VACUUM`) skip the
    /// trigger even if the ratio would otherwise qualify — the cost of
    /// rewriting a 64 KiB file isn't worth the few bytes reclaimed.
    #[test]
    fn auto_vacuum_skips_under_min_pages_floor() {
        let path = tmp_path("av_under_floor");
        let mut db = seed_db(); // small: just users + notes, ~5 pages
        db.source_path = Some(path.clone());
        save_database(&mut db, &path).expect("save");
        // Confirm we're below the floor so the test is meaningful.
        let pages_before = db.pager.as_ref().unwrap().header().page_count;
        assert!(
            pages_before < MIN_PAGES_FOR_AUTO_VACUUM,
            "test setup is too large: floor would not apply (got {pages_before} pages, \
             floor is {MIN_PAGES_FOR_AUTO_VACUUM})"
        );

        process_command("DROP TABLE users;", &mut db).expect("drop");

        let pages_after = db.pager.as_ref().unwrap().header().page_count;
        assert_eq!(
            pages_after, pages_before,
            "below MIN_PAGES_FOR_AUTO_VACUUM, drop must not trigger compaction"
        );
        assert!(
            db.pager.as_ref().unwrap().header().freelist_head != 0,
            "drop must still populate the freelist normally"
        );

        cleanup(&path);
    }

    /// Setter rejects NaN, infinities, and values outside `0.0..=1.0`
    /// rather than silently saturating.
    #[test]
    fn set_auto_vacuum_threshold_rejects_out_of_range() {
        let mut db = Database::new("t".to_string());
        for bad in [-0.01_f32, 1.01, f32::NAN, f32::INFINITY, f32::NEG_INFINITY] {
            let err = db.set_auto_vacuum_threshold(Some(bad)).unwrap_err();
            assert!(
                format!("{err}").contains("auto_vacuum_threshold"),
                "expected a typed range error for {bad}, got: {err}"
            );
        }
        // The default survives the rejected sets unchanged.
        assert_eq!(db.auto_vacuum_threshold(), Some(0.25));
        // And valid values land.
        db.set_auto_vacuum_threshold(Some(0.0)).unwrap();
        assert_eq!(db.auto_vacuum_threshold(), Some(0.0));
        db.set_auto_vacuum_threshold(Some(1.0)).unwrap();
        assert_eq!(db.auto_vacuum_threshold(), Some(1.0));
        db.set_auto_vacuum_threshold(None).unwrap();
        assert_eq!(db.auto_vacuum_threshold(), None);
    }

    // ---------------------------------------------------------------
    // SQLR-13 — `PRAGMA auto_vacuum` SQL-level coverage. Mirrors the
    // SQLR-10 setter tests above, but routed through SQL so SDK / FFI
    // / MCP consumers (which can't reach the Rust setter directly)
    // get the same guarantees.
    // ---------------------------------------------------------------

    /// `PRAGMA auto_vacuum = N;` set + `PRAGMA auto_vacuum;` read
    /// round-trip the threshold, observable via `auto_vacuum_threshold`.
    #[test]
    fn pragma_auto_vacuum_set_and_read_via_sql() {
        let mut db = Database::new("t".to_string());

        let resp = process_command("PRAGMA auto_vacuum = 0.5;", &mut db).expect("set");
        assert!(
            resp.contains("PRAGMA"),
            "set form should produce a PRAGMA status, got: {resp}"
        );
        assert_eq!(db.auto_vacuum_threshold(), Some(0.5));

        // Read form — status mentions a returned row.
        let resp = process_command("PRAGMA auto_vacuum;", &mut db).expect("read");
        assert!(resp.contains("1 row"), "expected a 1-row read, got: {resp}");
    }

    /// `PRAGMA auto_vacuum = OFF;` (bare identifier — sqlparser's own
    /// pragma-value parser would reject this, the SQLR-13 dispatcher
    /// must accept it) and `= NONE;` both disable the trigger. So does
    /// the quoted form `'OFF'`.
    #[test]
    fn pragma_auto_vacuum_off_disables_trigger() {
        for raw in ["OFF", "off", "NONE", "none", "'OFF'", "'NONE'"] {
            let mut db = Database::new("t".to_string());
            assert_eq!(db.auto_vacuum_threshold(), Some(0.25));

            let stmt = format!("PRAGMA auto_vacuum = {raw};");
            process_command(&stmt, &mut db)
                .unwrap_or_else(|e| panic!("`{stmt}` should disable: {e}"));
            assert_eq!(
                db.auto_vacuum_threshold(),
                None,
                "`{stmt}` should clear the threshold"
            );
        }
    }

    /// Out-of-range numeric values surface as a typed error via the
    /// shared `set_auto_vacuum_threshold` validator — no silent
    /// saturation. Mirrors the SQLR-10 setter coverage.
    #[test]
    fn pragma_auto_vacuum_rejects_out_of_range_via_sql() {
        let mut db = Database::new("t".to_string());
        for bad in ["-0.01", "1.01", "1.5"] {
            let stmt = format!("PRAGMA auto_vacuum = {bad};");
            let err = process_command(&stmt, &mut db).unwrap_err();
            assert!(
                format!("{err}").contains("auto_vacuum_threshold"),
                "expected range error for `{stmt}`, got: {err}"
            );
        }
        // Default survives all the rejected sets.
        assert_eq!(db.auto_vacuum_threshold(), Some(0.25));
    }

    /// Junk strings (anything that isn't a number or `OFF`/`NONE`) are
    /// rejected at parse time with a typed error, not silently treated
    /// as "disable".
    #[test]
    fn pragma_auto_vacuum_rejects_unknown_strings_via_sql() {
        let mut db = Database::new("t".to_string());
        let err = process_command("PRAGMA auto_vacuum = WAL;", &mut db).unwrap_err();
        assert!(
            format!("{err}").contains("OFF/NONE"),
            "expected OFF/NONE-style error, got: {err}"
        );
        // Default unaffected.
        assert_eq!(db.auto_vacuum_threshold(), Some(0.25));
    }

    /// Pragmas SQLRite doesn't know about return `NotImplemented` —
    /// not a generic parser error. Future pragmas plug in here.
    /// (Phase 11.3 made `journal_mode` a recognised pragma; this
    /// test uses a name that's still unsupported.)
    #[test]
    fn pragma_unknown_returns_not_implemented() {
        let mut db = Database::new("t".to_string());
        let err = process_command("PRAGMA synchronous = NORMAL;", &mut db).unwrap_err();
        assert!(
            matches!(err, SQLRiteError::NotImplemented(_)),
            "unknown pragma must surface NotImplemented, got: {err:?}"
        );
    }

    /// Setting the threshold via SQL must produce identical behavior to
    /// the Rust setter on the actual auto-VACUUM trigger: `= 0.99`
    /// suppresses, `= OFF` disables, default fires. Sanity-checks that
    /// `process_command_with_render`'s pre-parse step doesn't desync
    /// the in-memory state from the file.
    #[test]
    fn pragma_auto_vacuum_drives_real_trigger() {
        // Sub-case A — `PRAGMA auto_vacuum = OFF;` keeps file at HWM.
        {
            let path = tmp_path("av_pragma_off");
            let mut db = auto_vacuum_setup(&path);
            process_command("PRAGMA auto_vacuum = OFF;", &mut db).expect("disable via PRAGMA");
            assert_eq!(db.auto_vacuum_threshold(), None);

            let pages_before = db.pager.as_ref().unwrap().header().page_count;
            process_command("DROP TABLE bloat;", &mut db).expect("drop");
            let pages_after = db.pager.as_ref().unwrap().header().page_count;
            assert_eq!(
                pages_after, pages_before,
                "PRAGMA-driven OFF must keep page_count at the HWM"
            );
            cleanup(&path);
        }

        // Sub-case B — high threshold via PRAGMA suppresses the
        // trigger on a single drop.
        {
            let path = tmp_path("av_pragma_high");
            let mut db = auto_vacuum_setup(&path);
            process_command("PRAGMA auto_vacuum = 0.99;", &mut db).expect("set high");
            assert_eq!(db.auto_vacuum_threshold(), Some(0.99));

            let pages_before = db.pager.as_ref().unwrap().header().page_count;
            process_command("DROP TABLE bloat;", &mut db).expect("drop");
            let pages_after = db.pager.as_ref().unwrap().header().page_count;
            assert_eq!(
                pages_after, pages_before,
                "high PRAGMA threshold must suppress the trigger"
            );
            cleanup(&path);
        }

        // Sub-case C — re-arm via PRAGMA after disable: the trigger
        // fires again on the next page-releasing DDL.
        {
            let path = tmp_path("av_pragma_rearm");
            let mut db = auto_vacuum_setup(&path);
            process_command("PRAGMA auto_vacuum = OFF;", &mut db).unwrap();
            // Drop with the trigger off — pages land on the freelist
            // but the file stays at HWM.
            process_command("DROP TABLE bloat;", &mut db).unwrap();
            let pages_after_off_drop = db.pager.as_ref().unwrap().header().page_count;
            assert!(db.pager.as_ref().unwrap().header().freelist_head != 0);

            // Re-arm via PRAGMA, then drop one more thing — the
            // accumulated freelist still exceeds 25%, so auto-VACUUM
            // fires.
            process_command("PRAGMA auto_vacuum = 0.25;", &mut db).expect("re-arm");
            process_command("CREATE INDEX idx_keep_n ON keep (n);", &mut db).unwrap();
            process_command("DROP INDEX idx_keep_n;", &mut db).expect("drop index");

            let pages_after_rearm = db.pager.as_ref().unwrap().header().page_count;
            assert!(
                pages_after_rearm < pages_after_off_drop,
                "re-armed PRAGMA must let auto-VACUUM fire: was {pages_after_off_drop}, \
                 now {pages_after_rearm}"
            );
            assert_eq!(db.pager.as_ref().unwrap().header().freelist_head, 0);
            cleanup(&path);
        }
    }

    /// VACUUM modifiers (FULL, REINDEX, table targets, …) are rejected
    /// with NotImplemented — only bare `VACUUM;` is supported.
    #[test]
    fn vacuum_modifiers_are_rejected() {
        let path = tmp_path("vacuum_modifiers");
        let mut db = seed_db();
        db.source_path = Some(path.clone());
        save_database(&mut db, &path).expect("save");
        for stmt in ["VACUUM FULL;", "VACUUM users;"] {
            let err = process_command(stmt, &mut db).unwrap_err();
            assert!(
                format!("{err}").contains("VACUUM modifiers"),
                "expected modifier rejection for `{stmt}`, got: {err}"
            );
        }
        cleanup(&path);
    }
}