Virtual-table extensions

These expose data through CREATE VIRTUAL TABLE name USING module(...); some accept Go data via sqlite.Pointer(slice).

import (
  sqlite "gosqlite.org"
  _ "gosqlite.org/ext/array/auto"
)

// Bind a Go slice as a SQL table:
db.Where(`id IN (SELECT value FROM array(?))`,
  sqlite.Pointer([]int64{1, 5, 17})).Find(&users)

// Or DDL-based vtabs:
db.Exec(`CREATE VIRTUAL TABLE temp.log USING lines(data='INFO ok\nERROR …')`)
db.Exec(`CREATE VIRTUAL TABLE recent USING statement('SELECT name FROM users WHERE id >= ?')`)

Catalog

Package	What it exposes
`array`	a Go slice as a SQL table (transparent `sqlite.Pointer` path)
`lines`	split a text blob/file into rows (typed `lines.Table`)
`csv`	CSV files/strings as tables — delimiters, column affinity, `skip=N` (typed `csv.Table`)
`statement`	parametrized views with `?N` / `:name` HIDDEN bind columns
`closure`	`transitive_closure` graph walker with optional depth bounds (typed `closure.Graph`)
`pivot`	three-SELECT cross-tab — rows × columns × cell aggregate
`rtree`	R-Tree spatial index over the built-in `rtree`/`geopoly` vtabs + custom geometry/query callbacks + a `circle()` geometry (typed `rtree.Table`)
`series`	`generate_series(start, stop[, step])` integer sequence

Typed handles (csv.Table, lines.Table, closure.Graph, rtree.Table) hide the DDL + query SQL. Demos under examples/features/extensions/; end-to-end against gorm: gorm/examples/ext-vtabs/.

fs.FS-bound readers

csv, lines, fileio accept a sandbox fs.FS at registration time (which the /auto package can’t carry). Install a connect-hook on the singleton driver before gorm.Open so every pooled conn picks it up:

sqlite.DefaultDriver().RegisterConnectionHook(
  func(c sqlite.ExecQuerierContext, _ string) error {
    return csv.RegisterFS(c.(*sqlite.Conn), myFsys)
  })

The same shape works for lines.RegisterFS and fileio.RegisterFS (fileio provides sandboxed file read/write SQL functions). blobio is not filesystem-bound: it registers with plain blobio.Register(conn) and provides incremental BLOB I/O over database columns, not an fs.FS.