Skip to main content

sqlite_plus_orm 0.1.9 copy "sqlite_plus_orm: ^0.1.9" to clipboard
sqlite_plus_orm: ^0.1.9 copied to clipboard

Typed queries, JSON documents, search, vectors, and migrations for sqlite_plus.

Changelog #

0.1.9 #

Depends on sqlite_plus ^0.1.9. The README opens on books.db. The library behavior is the 0.1.8 release.

0.1.8 #

Depends on sqlite_plus ^0.1.8. The library behavior is the 0.1.7 release.

0.1.7 #

Depends on sqlite_plus ^0.1.7.

  • A generated integer key named rowid, oid, or _rowid_ is the SQLite row id when the table has no PRIMARY KEY and table_xinfo does not list that name. plan() does not rebuild the table for that key.
  • A declared primary key, a WITHOUT ROWID table, and any other missing column still require a reviewed migration.
  • The README links the other packages on pub.dev, including a short example of each.
  • DbSet.openMemory and DbSet.openDbFile remain. A second openDbFile of the same path while the first owner is open throws StateError.

0.1.6 #

Depends on sqlite_plus ^0.1.6. Published number for the 0.1.4 library.

0.1.4 #

Depends on sqlite_plus 0.1.5.

  • @DbDocument() stores one object as JSON text. Codecs.document calls fromJson and toJson. row.card.path(r'$.title') binds json_extract. A returned row is decoded. The other rows stay text.
  • The model schema records "document": true for that column. The SQL type stays TEXT. A String column still uses @DbJson.
  • DbModules, DbTokenizers, and DbVectorMetrics are documented String constants. DbVectorLimits documents 1 and 4096. The annotation fields stay strings, so a missing name can still be written by hand.
  • @DbEntity(module:, tokenizer:) creates fts5, rtree, or vec0. An unknown module is rejected. A tokenizer belongs on fts5.
  • @DbVector(dimensions) stores List<double> as little-endian float32. The default metric is l2.
  • search queries an FTS5 table. nearest writes column MATCH ? AND k = ? and orders by distance.
  • sqlite_plus_arabic and sqlite_plus_vec are registered by the application. This package does not import them.

0.1.3 #

Depends on sqlite_plus 0.1.3. pub.dev already serves 0.1.2.

  • @DbScalar() on an instance method or getter reads mapped fields. selectWhere((row) => row.twice() > 20) runs that body for each row. The context registers table_method on the connection.
  • A column, a JSON path, or a call accepts the engine scalars: abs, upper, lower, length, trim, substr, replace, round, coalesce, ifNull, concat, sign, floor, ceil, sqrt, date, time, dateTime, unixEpoch, and julianDay.

0.1.2 #

Depends on sqlite_plus 0.1.2. The page uses the Linux release-build sample and the new logo.

0.1.1 #

The package page leads with a save, a revision, and a JSON path. Depends on sqlite_plus 0.1.1.

0.1.0 #

Release number. Depends on sqlite_plus 0.1.0.

0.0.3 #

Release number. Depends on sqlite_plus 0.0.3.

0.0.2 #

Published release. Depends on sqlite_plus 0.0.2.

0.0.1-dev.1 #

First pre-release.

  • Typed queries, change tracking, relationships, and reviewed migrations.
  • The reading-log example ships with the package.
1
likes
0
points
1.56k
downloads

Publisher

verified publisherquds.cc

Weekly Downloads

Typed queries, JSON documents, search, vectors, and migrations for sqlite_plus.

Repository (GitHub)
View/report issues

Topics

#sqlite #database #orm

License

unknown (license)

Dependencies

sqlite_plus

More

Packages that depend on sqlite_plus_orm