Make `generate_connect_constructor` the default

docs/pages/docs/Advanced Features/builder_options.md

@@ -48,13 +48,15 @@ At the moment, drift supports these options:
   is based on the table name (e.g. a `@DataClassName('Users') class UsersTable extends Table` would generate
   a `UsersTableCompanion`). With this option, the name is based on the data class (so `UsersCompanion` in
   this case).
-* `use_column_name_as_json_key_when_defined_in_moor_file` (defaults to `true`): When serializing columns declared inside a 
+* `use_column_name_as_json_key_when_defined_in_moor_file` (defaults to `true`): When serializing columns declared inside a
   `.moor` (or `.drift`) file from and to json, use their sql name instead of the generated Dart getter name
   (so a column named `user_name` would also use `user_name` as a json key instead of `userName`).
   You can always override the json key by using a `JSON KEY` column constraint
   (e.g. `user_name VARCHAR NOT NULL JSON KEY userName`).
-* `generate_connect_constructor`: Generate necessary code to support the [isolate runtime]({{ "isolates.md" | pageUrl }}).
-  This is a build option because isolates are still experimental. This will be the default option eventually.
+* `generate_connect_constructor`: Generates a named `connect()` constructor on database classes
+  that takes a `DatabaseConnection` instead of a `QueryExecutor` - this allows sharing stream queries
+  between two drift database instances, which can be helpful for some [isolate setups]({{ "isolates.md" | pageUrl }}).
+  The option is enabled by default.
 * `data_class_to_companions` (defaults to `true`): Controls whether drift will write the `toCompanion` method in generated
    data classes.
 * `mutable_classes` (defaults to `false`): The fields generated in generated data, companion and result set classes are final

docs/pages/docs/Advanced Features/isolates.md

@@ -45,7 +45,7 @@ replacement for the `NativeDatabase` you've been using before:
 {% include "blocks/snippet" snippets = snippets name = 'simple' %}
 
 In the common case where you only need a isolate for performance reasons, this
-is as simple as it gets.
+is everything you need to do to run queries in a background isolate.
 The rest of this article explains a more complex setup giving you full control
 over the internal components making up a drift isolate. This is useful for
 advanced use cases, including:
@@ -60,29 +60,26 @@ the complicated bits are hidden behind a simple method.
 
 ## Preparations
 
-To use the isolate api, first enable the appropriate [build option]({{ "builder_options.md" | pageUrl }}) by
-creating a file called `build.yaml` in your project root, next to your `pubspec.yaml`. It should have the following
-content:
-```yaml
-targets:
-  $default:
-    builders:
-      drift_dev:
-        options:
-          generate_connect_constructor: true
-```
+When a drift database is opened multiple times, the two instances need to
+synchronize query streams, so that updates made in one instance are reflected
+in queries watched by the other.
 
-Next, re-run the build. You can now add another constructor to the generated database class:
+By default, drift databases are created from a `QueryExecutor` - an interface
+responsible for running the SQL statements generated by higher-level drift APIs.
+To share both the underlying database and a drift-specific mechanism for
+streams, you can instead use a `DatabaseConnection` instance. For that, you need
+to use the generated `connect()` constructor in your database class:
 
 {% include "blocks/snippet" snippets = snippets name = 'database' %}
 
-This setup is unfortunately necessary for backwards compatibility. A
-`DatabaseConnection` and the `connect` constructor make it possible to share
-query streams between isolates, the default constructor can't do this. In a
-future drift reelase, this option will no longer be necessary.
+Having a second constructor taking a `DatabaseConnection` is necessary for
+backwards compatibility since the default constructor only takes a
+`QueryExecutor` which is insufficient to synchronize streams.
+If you only want to open your database with a `DatabaseConnection`, you can
+remove the default constructor though.
 
-After adding the `connect` constructor, you can launch a drift isolate to
-connect to:
+After adding the new constructor, you can create instances of your database
+that will transparently run queries on a background isolate:
 
 ## Using drift in a background isolate {#using-moor-in-a-background-isolate}
 
@@ -102,11 +99,12 @@ Internally, drift will connect when the first query is sent to the database.
 
 ### Initialization on the main thread
 
-At the moment, Flutter's platform channels are [not available on background isolates](https://github.com/flutter/flutter/issues/13937).
-If you want to use functions like `getApplicationDocumentsDirectory` from `path_provider` to
-construct the database's path, we'll have to use some tricks to avoid using platforms channels.
-Here, we're going to start the isolate running the database manually. This allows us to pass additional
-data that we calculated on the main thread.
+Before Flutter 3.7, platforms channels weren't [available on background isolates](https://github.com/flutter/flutter/issues/13937).
+So, if functions like `getApplicationDocumentsDirectory` from `path_provider`
+are used to construct the path to the database, some tricks were necessary.
+This section describes a workaround to start the isolate running the database
+manually. This allows passing additional data that can be computed on the main
+isolate, using platform channels.
 
 {% include "blocks/snippet" snippets = snippets name = 'initialization' %}
 

docs/pages/docs/Advanced Features/migrations.md

@@ -369,12 +369,12 @@ void main() {
 
 In general, a test looks like this:
 
-- Use `verifier.startAt()` to obtain a [connection](https://pub.dev/documentation/moor/latest/moor/DatabaseConnection-class.html)
+1. Use `verifier.startAt()` to obtain a [connection](https://drift.simonbinder.eu/api/drift/databaseconnection-class)
   to a database with an initial schema.
   This database contains all your tables, indices and triggers from that version, created by using `Migrator.createAll`.
-- Create your application database (you can enable the [`generate_connect_constructor`]({{ "builder_options.md" | pageUrl }}) to use
-  a `DatabaseConnection` directly)
-- Call `verifier.migrateAndValidate(db, version)`. This will initiate a migration towards the target version (here, `2`).
+2. Create your application database with that connection - you can forward the `DatabaseConnection` to the
+  `GeneratedDatabase.connect()` constructor on the parent class for this.
+3. Call `verifier.migrateAndValidate(db, version)`. This will initiate a migration towards the target version (here, `2`).
   Unlike the database created by `startAt`, this uses the migration logic you wrote for your database.
 
 `migrateAndValidate` will extract all `CREATE` statement from the `sqlite_schema` table and semantically compare them.

docs/pages/docs/Other engines/web.md

@@ -194,8 +194,7 @@ DatabaseConnection connectToWorker() {
 }
 ```
 
-You can pass that `DatabaseConnection` to your database by enabling the
-`generate_connect_constructor` build option.
+You can then open a drift database with that connection.
 For more information on the `DatabaseConnection` class, see the documentation on
 [isolates]({{ "../Advanced Features/isolates.md" | pageUrl }}).
 

drift/build.yaml

@@ -15,7 +15,6 @@ targets:
         options:
           override_hash_and_equals_in_result_sets: true
           use_column_name_as_json_key_when_defined_in_moor_file: true
-          generate_connect_constructor: true
           write_from_json_string_constructor: true
           raw_result_set_data: true
           named_parameters: true
@@ -38,7 +37,6 @@ targets:
           # Dart doesn't support YAML merge tags yet, https://github.com/dart-lang/yaml/issues/121
           override_hash_and_equals_in_result_sets: true
           use_column_name_as_json_key_when_defined_in_moor_file: true
-          generate_connect_constructor: true
           write_from_json_string_constructor: true
           raw_result_set_data: true
           named_parameters: true

drift_dev/CHANGELOG.md

@@ -4,6 +4,7 @@
 - Support `MAPPED BY` for individual columns in queries or in views defined with SQL.
 - Consistently interpret `CAST (x AS DATETIME)` and `CAST(x AS TEXT)` in drift files.
 - Support a `CAST` to an enum type in drift types.
+- The `generate_connect_constructor` option is now enabled by default.
 - Support two different queries using `LIST()` columns having the same result class name.
 - Fix table classes not extending defining Dart classes with modular generation.
 - Fix `@UseDataClass` with `extending` not working with modular generation.

drift_dev/lib/src/analysis/options.dart

@@ -45,9 +45,12 @@ class DriftOptions {
       defaultValue: true)
   final bool useColumnNameAsJsonKeyWhenDefinedInMoorFile;
 
-  /// Generate a `connect` constructor in database superclasses. This is
-  /// required to run databases in a background isolate.
-  @JsonKey(name: 'generate_connect_constructor', defaultValue: false)
+  /// Generate a `connect` constructor in database superclasses.
+  ///
+  /// This makes drift generate a constructor for database classes that takes a
+  /// `DatabaseConnection` instead of just a `QueryExecutor` - meaning that
+  /// stream queries can also be shared across multiple database instances.
+  @JsonKey(name: 'generate_connect_constructor', defaultValue: true)
   final bool generateConnectConstructor;
 
   @JsonKey(name: 'sqlite_modules', defaultValue: [])
@@ -104,7 +107,7 @@ class DriftOptions {
     this.skipVerificationCode = false,
     this.useDataClassNameForCompanions = false,
     this.useColumnNameAsJsonKeyWhenDefinedInMoorFile = true,
-    this.generateConnectConstructor = false,
+    this.generateConnectConstructor = true,
     this.dataClassToCompanions = true,
     this.generateMutableClasses = false,
     this.rawResultSetData = false,
@@ -180,9 +183,6 @@ class DriftOptions {
   bool hasModule(SqlModule module) => effectiveModules.contains(module);
 
   /// Checks whether a deprecated option is enabled.
-  ///
-  /// At this time, all deprecated options have been removed, meaning that this
-  /// getter always returns `false`.
   bool get enabledDeprecatedOption => false;
 
   SqlDialect get effectiveDialect => dialect?.dialect ?? SqlDialect.sqlite;

drift_dev/lib/src/generated/analysis/options.g.dart

@@ -49,7 +49,7 @@ DriftOptions _$DriftOptionsFromJson(Map json) => $checkedCreate(
               'use_column_name_as_json_key_when_defined_in_moor_file',
               (v) => v as bool? ?? true),
           generateConnectConstructor: $checkedConvert(
-              'generate_connect_constructor', (v) => v as bool? ?? false),
+              'generate_connect_constructor', (v) => v as bool? ?? true),
           dataClassToCompanions: $checkedConvert(
               'data_class_to_companions', (v) => v as bool? ?? true),
           generateMutableClasses:

examples/app/build.yaml

@@ -25,9 +25,6 @@ targets:
               version: "3.38"
               modules: [fts5]
 
-          # This allows us to share a drift database across isolates (or different tabs on the web)
-          generate_connect_constructor: true
-
       # Configuring this builder isn't required for most apps. In our case, we
       # want to compile the web worker in `web/worker.dart` to JS and we use the
       # build system for that.

examples/migrations_example/build.yaml

@@ -3,7 +3,6 @@ targets:
     builders:
       drift_dev:
         options:
-          generate_connect_constructor: true
           store_date_time_values_as_text: true
           sql:
             dialect: sqlite

examples/web_worker_example/build.yaml

@@ -31,7 +31,6 @@ targets:
     builders:
       drift_dev:
         options:
-          generate_connect_constructor: true
           named_parameters: true
       build_web_compilers:entrypoint:
         generate_for:

extras/integration_tests/drift_testcases/build.yaml

@@ -5,7 +5,6 @@ targets:
         options:
           override_hash_and_equals_in_result_sets: true
           use_column_name_as_json_key_when_defined_in_moor_file: true
-          generate_connect_constructor: true
           write_from_json_string_constructor: true
           raw_result_set_data: false
           named_parameters: false