sequel 5.19.0 → 5.24.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/CHANGELOG +102 -0
- data/doc/dataset_filtering.rdoc +15 -0
- data/doc/opening_databases.rdoc +5 -1
- data/doc/release_notes/5.20.0.txt +89 -0
- data/doc/release_notes/5.21.0.txt +87 -0
- data/doc/release_notes/5.22.0.txt +48 -0
- data/doc/release_notes/5.23.0.txt +56 -0
- data/doc/release_notes/5.24.0.txt +56 -0
- data/doc/sharding.rdoc +2 -0
- data/doc/testing.rdoc +1 -0
- data/doc/transactions.rdoc +38 -0
- data/lib/sequel/adapters/ado.rb +27 -19
- data/lib/sequel/adapters/jdbc.rb +7 -1
- data/lib/sequel/adapters/jdbc/mysql.rb +2 -2
- data/lib/sequel/adapters/jdbc/postgresql.rb +1 -13
- data/lib/sequel/adapters/jdbc/sqlite.rb +29 -0
- data/lib/sequel/adapters/mysql2.rb +2 -3
- data/lib/sequel/adapters/shared/mssql.rb +7 -7
- data/lib/sequel/adapters/shared/postgres.rb +37 -19
- data/lib/sequel/adapters/shared/sqlite.rb +27 -3
- data/lib/sequel/adapters/sqlite.rb +1 -1
- data/lib/sequel/adapters/tinytds.rb +12 -0
- data/lib/sequel/adapters/utils/mysql_mysql2.rb +2 -0
- data/lib/sequel/database/logging.rb +7 -1
- data/lib/sequel/database/query.rb +1 -1
- data/lib/sequel/database/schema_generator.rb +12 -3
- data/lib/sequel/database/schema_methods.rb +2 -0
- data/lib/sequel/database/transactions.rb +57 -5
- data/lib/sequel/dataset.rb +4 -2
- data/lib/sequel/dataset/actions.rb +3 -2
- data/lib/sequel/dataset/placeholder_literalizer.rb +4 -1
- data/lib/sequel/dataset/query.rb +5 -1
- data/lib/sequel/dataset/sql.rb +11 -7
- data/lib/sequel/extensions/named_timezones.rb +52 -8
- data/lib/sequel/extensions/pg_array.rb +4 -0
- data/lib/sequel/extensions/pg_json.rb +387 -123
- data/lib/sequel/extensions/pg_range.rb +3 -2
- data/lib/sequel/extensions/pg_row.rb +3 -1
- data/lib/sequel/extensions/schema_dumper.rb +1 -1
- data/lib/sequel/extensions/server_block.rb +15 -4
- data/lib/sequel/model/associations.rb +35 -9
- data/lib/sequel/model/plugins.rb +104 -0
- data/lib/sequel/plugins/association_dependencies.rb +3 -3
- data/lib/sequel/plugins/association_pks.rb +14 -4
- data/lib/sequel/plugins/association_proxies.rb +3 -2
- data/lib/sequel/plugins/class_table_inheritance.rb +11 -0
- data/lib/sequel/plugins/composition.rb +13 -9
- data/lib/sequel/plugins/finder.rb +2 -2
- data/lib/sequel/plugins/hook_class_methods.rb +17 -5
- data/lib/sequel/plugins/insert_conflict.rb +72 -0
- data/lib/sequel/plugins/inverted_subsets.rb +2 -2
- data/lib/sequel/plugins/pg_auto_constraint_validations.rb +147 -59
- data/lib/sequel/plugins/rcte_tree.rb +6 -0
- data/lib/sequel/plugins/static_cache.rb +8 -3
- data/lib/sequel/plugins/static_cache_cache.rb +53 -0
- data/lib/sequel/plugins/subset_conditions.rb +2 -2
- data/lib/sequel/plugins/validation_class_methods.rb +5 -3
- data/lib/sequel/sql.rb +15 -3
- data/lib/sequel/timezones.rb +50 -11
- data/lib/sequel/version.rb +1 -1
- data/spec/adapters/mssql_spec.rb +24 -0
- data/spec/adapters/mysql_spec.rb +0 -5
- data/spec/adapters/postgres_spec.rb +319 -1
- data/spec/bin_spec.rb +1 -1
- data/spec/core/database_spec.rb +123 -2
- data/spec/core/dataset_spec.rb +33 -1
- data/spec/core/expression_filters_spec.rb +25 -1
- data/spec/core/schema_spec.rb +24 -0
- data/spec/extensions/class_table_inheritance_spec.rb +30 -8
- data/spec/extensions/core_refinements_spec.rb +1 -1
- data/spec/extensions/hook_class_methods_spec.rb +22 -0
- data/spec/extensions/insert_conflict_spec.rb +103 -0
- data/spec/extensions/migration_spec.rb +13 -0
- data/spec/extensions/named_timezones_spec.rb +109 -2
- data/spec/extensions/pg_auto_constraint_validations_spec.rb +45 -0
- data/spec/extensions/pg_json_spec.rb +218 -29
- data/spec/extensions/pg_range_spec.rb +76 -9
- data/spec/extensions/rcte_tree_spec.rb +6 -0
- data/spec/extensions/s_spec.rb +1 -1
- data/spec/extensions/schema_dumper_spec.rb +4 -2
- data/spec/extensions/server_block_spec.rb +38 -0
- data/spec/extensions/spec_helper.rb +8 -1
- data/spec/extensions/static_cache_cache_spec.rb +35 -0
- data/spec/integration/dataset_test.rb +25 -9
- data/spec/integration/plugin_test.rb +42 -0
- data/spec/integration/schema_test.rb +7 -2
- data/spec/integration/transaction_test.rb +50 -0
- data/spec/model/associations_spec.rb +84 -4
- data/spec/model/plugins_spec.rb +111 -0
- metadata +16 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 7ea0327d7fbbc76458fc9dc6a088069f04bd100b37f91288e9db9f37db84bbb7
|
4
|
+
data.tar.gz: 6195027a34de796b5109ee2b2ed57407ce9a58f75b1661a3be0ab9b2acbba885
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: aaa9a71ee0562ccedaeee6144c4764a8024699214dfd4fd7ce46e8a301d8df72f1a709d373f7f5d47c455ff70407b61e1e0bf05ee7f30e20c49a3be51a1bbea2
|
7
|
+
data.tar.gz: d8343280ef3c09e930df1db0e12eeac349ce23f396fccf1dfcc055a5e760d5dbb50fd5d0a4f8ce591ac4d12644833b9afea6a94fa1b52991c580baba28c26926
|
data/CHANGELOG
CHANGED
@@ -1,3 +1,105 @@
|
|
1
|
+
=== 5.24.0 (2019-09-01)
|
2
|
+
|
3
|
+
* Add Database#skip_logging? private method designed for extensions to force query timing even if no logger is present (adam12) (#1640)
|
4
|
+
|
5
|
+
* Allow a hostname specified in a defaults_file in the mysql2 adapter, by not explicitly setting :host (sapio-bdeamer) (#1638)
|
6
|
+
|
7
|
+
* Convert all database array types to Ruby arrays in the jdbc adapter (jeremyevans)
|
8
|
+
|
9
|
+
* Add static_cache_cache plugin for caching rows for static_cache models to a file to avoid database queries during model initialization (jeremyevans)
|
10
|
+
|
11
|
+
* Add :cache_file plugin option to pg_auto_constraint_validations plugin, for caching metadata to a file for faster initialization (jeremyevans)
|
12
|
+
|
13
|
+
* Support :unique_deferrable and :primary_key_deferrable column options (jeremyevans)
|
14
|
+
|
15
|
+
* Support :generated_always_as column option on PostgreSQL 12+ (jeremyevans)
|
16
|
+
|
17
|
+
=== 5.23.0 (2019-08-01)
|
18
|
+
|
19
|
+
* Work around a bug on jdbc-sqlite3 3.27.2.1 when parsing schema for tables with columns with default values (jeremyevans)
|
20
|
+
|
21
|
+
* Work around a bug in jdbc-sqlite3 3.27.2.1 when in Database#foreign_key_list in the jdbc/sqlite3 adapter (jeremyevans)
|
22
|
+
|
23
|
+
* Make Dataset#execute* private methods respect explicit servers option, fixing Dataset#paged_each in the postgres adapter when sharding (jeremyevans) (#1632)
|
24
|
+
|
25
|
+
* Handle instances of subclasses of core classes when wrapping objects in the pg_json extension (jeremyevans) (#1631)
|
26
|
+
|
27
|
+
* Support :ansi Database option in the tinytds adapter (kenaniah) (#1629)
|
28
|
+
|
29
|
+
* Support cross-database and linked servers when parsing schema on Microsoft SQL Server (kenaniah) (#1629)
|
30
|
+
|
31
|
+
* Add insert_conflict plugin for automatically handling unique constraint conflicts when saving new model instances on PostgreSQL 9.5+ and SQLite 3.24.0+ (jeremyevans)
|
32
|
+
|
33
|
+
* Avoid errors when parsing schema in the mock sqlite adapter (jeremyevans)
|
34
|
+
|
35
|
+
* Avoid possible thread-safety issue in the timezones support (jeremyevans)
|
36
|
+
|
37
|
+
* Handle offsets when typecasting an array or hash to datetime when Sequel.datetime_class = Time (jeremyevans)
|
38
|
+
|
39
|
+
* Support Sequel.datetime_class = Time when using the named_timezones extension (jeremyevans)
|
40
|
+
|
41
|
+
=== 5.22.0 (2019-07-01)
|
42
|
+
|
43
|
+
* Fix Dataset#multi_insert and #import with return: :primary_key on MSSQL when the dataset has a row_proc (jeremyevans) (#1627)
|
44
|
+
|
45
|
+
* Support Dataset#with :materialized option on PostgreSQL 12 for [NOT] MATERIALIZED (jeremyevans)
|
46
|
+
|
47
|
+
* Make Database#primary_key_sequence work on tables without serial sequences on PostgreSQL 12 (jeremyevans)
|
48
|
+
|
49
|
+
* Support ruby 2.7+ startless ranges in the pg_range extension (jeremyevans)
|
50
|
+
|
51
|
+
* Support ruby 2.7+ startless, endless ranges in filters, using an always true condition for them (jeremyevans)
|
52
|
+
|
53
|
+
* Support ruby 2.7+ startless ranges in filters, using just a <= or < operator for them (jeremyevans)
|
54
|
+
|
55
|
+
=== 5.21.0 (2019-06-01)
|
56
|
+
|
57
|
+
* Recognize additional DatabaseLockTimeout errors in mysql and mysql2 adapters (jeremyevans)
|
58
|
+
|
59
|
+
* Disallow eager_graph of ancestors and descendants associations when using the rcte_tree plugin (jeremyevans)
|
60
|
+
|
61
|
+
* Make jdbc/mysql adapter work when using JRuby with Java 11 (jeremyevans)
|
62
|
+
|
63
|
+
* Support window function options :window, :exclude, and :frame :type=>:groups, :start, and :end on SQLite 3.28.0+ (jeremyevans)
|
64
|
+
|
65
|
+
* Make the server_block extension respect the :servers_hash Database option (jeremyevans)
|
66
|
+
|
67
|
+
* Typecast string input for json/jsonb types as JSON strings instead of parsing as JSON in the pg_json extension when Database#typecast_json_strings is set to true (jeremyevans)
|
68
|
+
|
69
|
+
* Wrap JSON primitives (string, number, true, false, nil) in the pg_json extension when Database#wrap_json_primitives is set to true (jeremyevans)
|
70
|
+
|
71
|
+
* Convert the Database :timeout option to an integer in the sqlite adapter (jeremyevans) (#1620)
|
72
|
+
|
73
|
+
* Improve performance in ado adapter using more efficient inner loop (jeremyevans)
|
74
|
+
|
75
|
+
* Improve performance in ado adapter using faster callables for type conversion (jeremyevans)
|
76
|
+
|
77
|
+
* Fix handling of decimal values in the ado adapter when using locales where the decimal separator is , and not . (jeremyevans) (#1619)
|
78
|
+
|
79
|
+
=== 5.20.0 (2019-05-01)
|
80
|
+
|
81
|
+
* Fix reversing of alter_table add_foreign_key when :type option is used (jeremyevans) (#1615)
|
82
|
+
|
83
|
+
* Switch from using instance_exec to define_method for model associations and in some plugins (jeremyevans)
|
84
|
+
|
85
|
+
* Fix Database#server_version when using mysql2 adapter with mysql driver on MariaDB 10+ database (v-kolesnikov) (#1614)
|
86
|
+
|
87
|
+
* Make one_to_one setter method handle models that use joined datasets (jeremyevans) (#1612)
|
88
|
+
|
89
|
+
* Make auto_validations plugin work with the class_table_inheritance plugin (jeremyevans) (#1611)
|
90
|
+
|
91
|
+
* Avoid use of instance_exec for PlaceholderLiteralString#with_dataset (jeremyevans)
|
92
|
+
|
93
|
+
* Recognize float unsigned database types as float (keeguon, jeremyevans) (#1609)
|
94
|
+
|
95
|
+
* Support :savepoint options to Database#{after_commit,after_rollback} for making the hooks handle savepoints (jeremyevans)
|
96
|
+
|
97
|
+
* Avoid use of instance_exec in association_dependencies plugin (jeremyevans)
|
98
|
+
|
99
|
+
* Add pg_auto_constraint_validation_override to the pg_auto_constraint_validations plugin, for customizing columns and error message per constraint (jeremyevans)
|
100
|
+
|
101
|
+
* Make Database#check_constraints on PostgreSQL also include constraints where the related columns are not known (jeremyevans)
|
102
|
+
|
1
103
|
=== 5.19.0 (2019-04-02)
|
2
104
|
|
3
105
|
* Use more optimized approach to merging hashes in ruby 2.5+ (jeremyevans)
|
data/doc/dataset_filtering.rdoc
CHANGED
@@ -106,6 +106,21 @@ You can also use placeholders with :placeholder and a hash of placeholder values
|
|
106
106
|
items.where(Sequel.lit('category = :category', category: "ruby")).sql
|
107
107
|
# "SELECT * FROM items WHERE category = 'ruby'"
|
108
108
|
|
109
|
+
In order to combine AND and OR together, you have a few options:
|
110
|
+
|
111
|
+
items.where(category: nil).or(category: "ruby")
|
112
|
+
# SELECT * FROM items WHERE (category IS NULL) OR (category = 'ruby')
|
113
|
+
|
114
|
+
This won't work if you add other conditions:
|
115
|
+
|
116
|
+
items.where(name: "Programming in Ruby").where(category: nil).or(category: 'ruby')
|
117
|
+
# SELECT * FROM items WHERE ((name = 'Programming in Ruby') AND (category IS NULL)) OR (category = 'ruby')
|
118
|
+
|
119
|
+
The OR applies globally and not locally. To fix this, use & and |:
|
120
|
+
|
121
|
+
items.where(Sequel[name: "Programming in Ruby"] & (Sequel[category: nil] | Sequel[category: "ruby"]))
|
122
|
+
# SELECT * FROM items WHERE ((name = 'Programming in Ruby') AND ((category IS NULL) OR (category = 'ruby')))
|
123
|
+
|
109
124
|
=== Specifying SQL functions
|
110
125
|
|
111
126
|
Sequel also allows you to specify functions by using the Sequel.function method:
|
data/doc/opening_databases.rdoc
CHANGED
@@ -264,7 +264,8 @@ The following additional options are supported:
|
|
264
264
|
or an array of symbols or strings (e.g. <tt>:sql_mode=>[:no_zero_date, :pipes_as_concat]</tt>).
|
265
265
|
:timeout :: Sets the wait_timeout for the connection, defaults to 1 month.
|
266
266
|
:read_timeout :: Set the timeout in seconds for reading back results to a query.
|
267
|
-
:connect_timeout :: Set the timeout in seconds before a connection attempt is abandoned
|
267
|
+
:connect_timeout :: Set the timeout in seconds before a connection attempt is abandoned
|
268
|
+
(may not be supported when using MariaDB 10.2+ client libraries).
|
268
269
|
|
269
270
|
The :sslkey, :sslcert, :sslca, :sslcapath, and :sslca options (in that order) are passed to Mysql#ssl_set method
|
270
271
|
if either the :sslca or :sslkey option is given.
|
@@ -400,6 +401,9 @@ options that you may want to set are :login_timeout, :timeout, :tds_version, :az
|
|
400
401
|
|
401
402
|
Other Sequel specific options:
|
402
403
|
|
404
|
+
:ansi :: Set to true to enable the ANSI compatibility settings when connecting
|
405
|
+
(ANSI_NULLS, ANSI_PADDING, ANSI_WARNINGS, ANSI_NULL_DFLT_ON, QUOTED_IDENTIFIER,
|
406
|
+
CONCAT_NULL_YIELDS_NULL).
|
403
407
|
:server_version :: Override the server version to use (9000000 = SQL Server 2005).
|
404
408
|
This also works on any other adapter that connects to Microsoft
|
405
409
|
SQL Server.
|
@@ -0,0 +1,89 @@
|
|
1
|
+
= New Features
|
2
|
+
|
3
|
+
* Database#after_commit and #after_rollback transaction hook methods
|
4
|
+
now support a :savepoint option. Using the :savepoint option makes
|
5
|
+
the hooks savepoint-aware, so after_commit will only be called if
|
6
|
+
all enclosing savepoints and the transaction are committed, and
|
7
|
+
after_rollback will be called when any of the enclosing savepoints
|
8
|
+
are rolled back (which may be before transaction commit/rollback).
|
9
|
+
Examples:
|
10
|
+
|
11
|
+
x = nil
|
12
|
+
DB.transaction do # BEGIN
|
13
|
+
DB.transaction(savepoint: true) do # SAVEPOINT
|
14
|
+
DB.after_commit(savepoint: true){x = 1}
|
15
|
+
DB.after_rollback(savepoint: true){x = 2}
|
16
|
+
x # nil
|
17
|
+
end # RELEASE SAVEPOINT
|
18
|
+
x # nil
|
19
|
+
end # COMMIT
|
20
|
+
x # 1
|
21
|
+
|
22
|
+
x = nil
|
23
|
+
DB.transaction do # BEGIN
|
24
|
+
DB.transaction(savepoint: true) do # SAVEPOINT
|
25
|
+
DB.after_commit(savepoint: true){x = 1}
|
26
|
+
DB.after_rollback(savepoint: true){x = 2}
|
27
|
+
x # nil
|
28
|
+
raise Sequel::Rollback
|
29
|
+
end # ROLLBACK TO SAVEPOINT
|
30
|
+
x # 2
|
31
|
+
end # COMMIT
|
32
|
+
x # 2
|
33
|
+
|
34
|
+
x = nil
|
35
|
+
DB.transaction do # BEGIN
|
36
|
+
DB.transaction(savepoint: true) do # SAVEPOINT
|
37
|
+
DB.after_commit(savepoint: true){x = 1}
|
38
|
+
DB.after_rollback(savepoint: true){x = 2}
|
39
|
+
end # RELEASE SAVEPOINT
|
40
|
+
x # nil
|
41
|
+
raise Sequel::Rollback
|
42
|
+
end
|
43
|
+
x # 2
|
44
|
+
|
45
|
+
* The pg_auto_constraint_validations plugin now supports a
|
46
|
+
pg_auto_constraint_validation_override method for overriding
|
47
|
+
the columns and message for a specific constraint. This is
|
48
|
+
useful if the database cannot determine the columns (due
|
49
|
+
to the constraint containing a database function call), or
|
50
|
+
if you would like to customize the message per constraint.
|
51
|
+
|
52
|
+
= Other Improvements
|
53
|
+
|
54
|
+
* The one_to_one association setter now works with models that use
|
55
|
+
joined datasets, such as child models when using the
|
56
|
+
class_table_inheritance plugin.
|
57
|
+
|
58
|
+
* Database#check_constraints on PostgreSQL now also includes CHECK
|
59
|
+
constraints where the related columns are not known. The :columns
|
60
|
+
entry in the hash will be an empty array in such cases. The
|
61
|
+
exclusion of such constraints in previous versions was not
|
62
|
+
intentional, and the documentation implied that all CHECK
|
63
|
+
constraints were returned.
|
64
|
+
|
65
|
+
* Many cases where instance_exec was previously used on model
|
66
|
+
instances have been changed so that instance methods are defined
|
67
|
+
and called instead. This avoids the creation of singleton classes
|
68
|
+
for model instances, and can significantly improve performance in
|
69
|
+
some cases. This affects all associations as well as the
|
70
|
+
following plugins:
|
71
|
+
|
72
|
+
* composition
|
73
|
+
* hook_class_methods
|
74
|
+
* validation_class_methods
|
75
|
+
|
76
|
+
Other cases where instance_exec is now avoided and a different
|
77
|
+
approach is used:
|
78
|
+
|
79
|
+
* association_dependencies plugin
|
80
|
+
* PlaceholderLiteralString#with_dataset
|
81
|
+
|
82
|
+
* The auto_validations plugin now works with child models when using
|
83
|
+
the class_table_inheritance plugin.
|
84
|
+
|
85
|
+
* Database#server_version now works correctly in the mysql2 adapter
|
86
|
+
when using the MySQL driver with MariaDB 10+.
|
87
|
+
|
88
|
+
* The float unsigned type is now recognized and supported in the
|
89
|
+
schema parser and schema_dumper extension.
|
@@ -0,0 +1,87 @@
|
|
1
|
+
= New Features
|
2
|
+
|
3
|
+
* The pg_json extension now adds a Database#wrap_json_primitives
|
4
|
+
accessor. When set to true, JSON primitive values (string, number,
|
5
|
+
true, false, and null) will be wrapped by delegate Ruby objects
|
6
|
+
instead of using Ruby primitives. This allows the values to round
|
7
|
+
trip, so the following code will work even for primitive values in
|
8
|
+
json_column:
|
9
|
+
|
10
|
+
DB.extension :pg_json
|
11
|
+
DB.wrap_json_primitives = true
|
12
|
+
value = DB[:table].get(:json_column)
|
13
|
+
DB[:other_table].insert(json_column: value)
|
14
|
+
|
15
|
+
This should be enabled with care, especially in cases where false
|
16
|
+
and null JSON values are used, as the behavior will change if
|
17
|
+
the objects are used in a boolean context in Ruby, as only false
|
18
|
+
and nil in Ruby are treated as false:
|
19
|
+
|
20
|
+
# assume JSON false or null value
|
21
|
+
value = DB[:table].get(:json_column)
|
22
|
+
|
23
|
+
if value
|
24
|
+
# executed if wrap_json_primitives is true
|
25
|
+
else
|
26
|
+
# executed by default
|
27
|
+
end
|
28
|
+
|
29
|
+
When typecasting input in model objects to a JSON type, string
|
30
|
+
input will still be parsed as JSON. However, you can set the
|
31
|
+
Database#typecast_json_strings accessor to true, and then string
|
32
|
+
input will be considered as a JSON string instead of parsing the
|
33
|
+
string as JSON.
|
34
|
+
|
35
|
+
To prevent backwards compatibility issues, Sequel.pg_json/pg_jsonb
|
36
|
+
behavior has not changed. To support wrapping Ruby primitives in
|
37
|
+
the delegate objects, new Sequel.pg_json_wrap/pg_jsonb_wrap methods
|
38
|
+
have been added. These methods only handle the Ruby primitives,
|
39
|
+
they cannot be used if the existing object is already a delegate
|
40
|
+
object.
|
41
|
+
|
42
|
+
As model objects always consider a nil value as SQL NULL and do
|
43
|
+
not typecast it, if you want to explicitly set a JSON null value,
|
44
|
+
you need to wrap it explicitly:
|
45
|
+
|
46
|
+
model_object.json_column = Sequel.pg_json_wrap(nil)
|
47
|
+
|
48
|
+
= Other Improvements
|
49
|
+
|
50
|
+
* Sequel now supports window function options :window, :exclude, and
|
51
|
+
:frame :type=>:groups, :start, and :end on SQLite 3.28.0+.
|
52
|
+
|
53
|
+
* The server_block extension now respects the :servers_hash Database
|
54
|
+
option. This makes it more similar to Sequel's default behavior.
|
55
|
+
However, that means by default, the server_block extension will
|
56
|
+
default to handling unknown shards as the default shard, instead
|
57
|
+
of raising an error for them.
|
58
|
+
|
59
|
+
* The rcte_tree plugin now disallows eager graphing of the ancestors
|
60
|
+
and descendants associations. Previously, eager graphing of these
|
61
|
+
associations generated incorrect results. It is not possible to
|
62
|
+
eager graph these extensions, but normal eager loading does work.
|
63
|
+
|
64
|
+
* The ado adapter's performance has been improved by using faster
|
65
|
+
callables for type conversion and a more efficient inner loop.
|
66
|
+
|
67
|
+
* The sqlite adapter now converts a :timeout option given as a string
|
68
|
+
to an integer. This allows you to use the option inside of a
|
69
|
+
connection string.
|
70
|
+
|
71
|
+
* The mysql and mysql2 adapters now recognize an additional
|
72
|
+
DatabaseLockTimeout error.
|
73
|
+
|
74
|
+
* The jdbc/mysql adapter now works correctly when using JRuby with
|
75
|
+
Java 11.
|
76
|
+
|
77
|
+
* The ado adapter now handles numeric values when using locales that
|
78
|
+
use comma instead of period as the decimal separator.
|
79
|
+
|
80
|
+
= Backwards Compatibility
|
81
|
+
|
82
|
+
* In the pg_json extension, the following singleton methods of
|
83
|
+
Sequel::Postgres::JSONDatabaseMethods are now deprecated:
|
84
|
+
|
85
|
+
* parse_json
|
86
|
+
* db_parse_json
|
87
|
+
* db_parse_jsonb
|
@@ -0,0 +1,48 @@
|
|
1
|
+
= New Features
|
2
|
+
|
3
|
+
* Sequel now supports Ruby 2.7+ startless ranges in filters:
|
4
|
+
|
5
|
+
DB[:table].where(:column=>(..10))
|
6
|
+
# SELECT * FROM table WHERE (column <= 10)
|
7
|
+
|
8
|
+
DB[:table].where(:column=>(...10))
|
9
|
+
# SELECT * FROM table WHERE (column < 10)
|
10
|
+
|
11
|
+
It also supports startless, endless ranges in filters, using a
|
12
|
+
condition that is always true:
|
13
|
+
|
14
|
+
DB[:table].where(:column=>(nil..nil))
|
15
|
+
# SELECT * FROM table WHERE (1 = 1)
|
16
|
+
|
17
|
+
* Sequel now supports startless ranges in the pg_range extension:
|
18
|
+
|
19
|
+
DB.extension :pg_range
|
20
|
+
|
21
|
+
DB[:table].insert(:column=>(..10))
|
22
|
+
# INSERT INTO "table" ("column") VALUES ('[,10]') RETURNING "id"
|
23
|
+
|
24
|
+
DB[:table].insert(:column=>(...10))
|
25
|
+
# INSERT INTO "table" ("column") VALUES ('[,10)') RETURNING "id"
|
26
|
+
|
27
|
+
DB[:table].insert(:column=>(nil..nil))
|
28
|
+
# INSERT INTO "table" ("column") VALUES ('[,]') RETURNING "id"
|
29
|
+
|
30
|
+
* Sequel now supports a :materialized option in Dataset#with on
|
31
|
+
PostgreSQL 12+, to control the inlining of common table expressions:
|
32
|
+
|
33
|
+
DB[:t].with(:t, DB[:t2], :materialized=>false)
|
34
|
+
# WITH "t" AS NOT MATERIALIZED (SELECT * FROM "t2")
|
35
|
+
# SELECT * FROM "t"
|
36
|
+
|
37
|
+
DB[:t].with(:t, DB[:t2], :materialized=>true)
|
38
|
+
# WITH "t" AS MATERIALIZED (SELECT * FROM "t2")
|
39
|
+
# SELECT * FROM "t"
|
40
|
+
|
41
|
+
= Other Improvements
|
42
|
+
|
43
|
+
* Database#primary_key_sequence now works for tables without serial
|
44
|
+
sequences on PostgreSQL 12+.
|
45
|
+
|
46
|
+
* Dataset#multi_insert and #import with return: :primary_key option
|
47
|
+
on Microsoft SQL Server now work correctly if the dataset uses
|
48
|
+
a row_proc (e.g. for model datasets).
|
@@ -0,0 +1,56 @@
|
|
1
|
+
= New Features
|
2
|
+
|
3
|
+
* An insert_conflict plugin has been added for automatically handling
|
4
|
+
constraint conflicts when saving new model instances. It is
|
5
|
+
supported on PostgreSQL 9.5+ and SQLite 3.24.0+.
|
6
|
+
|
7
|
+
Album.new(name: 'Foo', copies_sold: 1000).
|
8
|
+
insert_conflict(
|
9
|
+
target: :name,
|
10
|
+
update: {copies_sold: Sequel[:excluded][:b]}
|
11
|
+
).
|
12
|
+
save
|
13
|
+
|
14
|
+
* On Microsoft SQL Server, the Database :ansi option has been added,
|
15
|
+
which sets the following ANSI related options:
|
16
|
+
|
17
|
+
* ANSI_NULLS
|
18
|
+
* ANSI_PADDING
|
19
|
+
* ANSI_WARNINGS
|
20
|
+
* ANSI_NULL_DFLT_ON
|
21
|
+
* QUOTED_IDENTIFIER
|
22
|
+
* CONCAT_NULL_YIELDS_NULL
|
23
|
+
|
24
|
+
= Other Improvements
|
25
|
+
|
26
|
+
* Sequel.datetime_class = Time is now supported when using the
|
27
|
+
named_timezones extension. For backwards compatibility, the
|
28
|
+
named_timezones extension still sets
|
29
|
+
Sequel.datetime_class = DateTime. When using Ruby 2.6+, the
|
30
|
+
Time instances have the timezone set on them using Ruby 2.6+'s
|
31
|
+
timezone support, but basic support works correctly in earlier
|
32
|
+
versions of Ruby.
|
33
|
+
|
34
|
+
* On Microsoft SQL Server, Sequel now handles parsing schema for
|
35
|
+
tables in another database on the same server or in a database
|
36
|
+
on a linked server.
|
37
|
+
|
38
|
+
* The pg_json extension now correctly handles subclasses of core
|
39
|
+
classes when wrapping objects. This stopped working in Sequel
|
40
|
+
5.21.0, when support for wrapping JSON primitives was added.
|
41
|
+
|
42
|
+
* Sequel now works around a couple bugs in jdbc-sqlite 3.27.2.1,
|
43
|
+
allowing schema parsing and foreign key parsing to work.
|
44
|
+
|
45
|
+
* Dataset#execute* private methods now respect an explicitly given
|
46
|
+
:server option, fixing Dataset#paged_each in the postgres adapter
|
47
|
+
when using sharding.
|
48
|
+
|
49
|
+
* Timezone offsets are now handled correctly when typecasting an array
|
50
|
+
or hash to datetime when Sequel.datetime_class = Time.
|
51
|
+
|
52
|
+
* Sequel now avoids errors when parsing schema when using the mock
|
53
|
+
SQLite adapter.
|
54
|
+
|
55
|
+
* A minor thread-safety issue has been fixed in the named_timezones
|
56
|
+
extension.
|
@@ -0,0 +1,56 @@
|
|
1
|
+
= New Features
|
2
|
+
|
3
|
+
* A :cache_file plugin option has been added to the
|
4
|
+
pg_auto_constraint_validations plugin. This option specifies
|
5
|
+
a file to use to cache the metadata the plugin uses, so the
|
6
|
+
plugin does not need to run 5 queries per model at startup to
|
7
|
+
load the metadata. This can dramatically improve startup time
|
8
|
+
when using the plugin with a large number of models.
|
9
|
+
|
10
|
+
To create the metadata file, load the plugin into Sequel::Model
|
11
|
+
(or whatever class you are using as the base class for your
|
12
|
+
model classes) with the :cache_file option, and after loading
|
13
|
+
all of the subclasses of that class, run:
|
14
|
+
|
15
|
+
Sequel::Model.dump_pg_auto_constraint_validations_cache
|
16
|
+
|
17
|
+
As when using the schema_caching and index_caching extensions,
|
18
|
+
it is up to the user to ensure that the cached metadata matches
|
19
|
+
the current database schema. Sequel does no checking of this,
|
20
|
+
as checking would take more time, and the point of this plugin
|
21
|
+
is to improve startup performance.
|
22
|
+
|
23
|
+
* A static_cache_cache plugin has been added. This plugin allows
|
24
|
+
for caching rows for models using the static_cache plugin. This
|
25
|
+
prevents the need to issue a query at model creation time to
|
26
|
+
get the rows. This plugin should be loaded into Sequel::Model
|
27
|
+
(or whatever class you are using as the base class for your
|
28
|
+
model classes) before loading the models using the static_cache
|
29
|
+
plugin. To create the metadata file, after all subclasses of
|
30
|
+
that class have been loaded, run:
|
31
|
+
|
32
|
+
Sequel::Model.dump_static_cache_cache
|
33
|
+
|
34
|
+
* :unique_deferrable and :primary_key_deferrable column
|
35
|
+
options are now supported on PostgreSQL 9+ and Oracle. This
|
36
|
+
allows you to created deferrable unique and primary key
|
37
|
+
column constraints. You could already create deferrable
|
38
|
+
table constraints using the :deferrable option to the primary_key
|
39
|
+
and unique methods.
|
40
|
+
|
41
|
+
* A :generated_always_as column option is now supported on
|
42
|
+
PostgreSQL 12+, for creating generated columns.
|
43
|
+
|
44
|
+
* A Database#skip_logging? private method has been added. This
|
45
|
+
is designed for use in extensions, to force log timing even
|
46
|
+
when no loggers are configured.
|
47
|
+
|
48
|
+
= Other Improvements
|
49
|
+
|
50
|
+
* Sequel no longer sets the :host option to localhost by default
|
51
|
+
in the mysql2 adapter. This prevents Sequel from overriding
|
52
|
+
a host specified in the defaults_file.
|
53
|
+
|
54
|
+
* All database array types are converted to Ruby arrays in the
|
55
|
+
jdbc adapter. Previously, this was only done in the
|
56
|
+
jdbc/postgresql subadapter.
|