Skip to content

CHANGES.rst

Latest commit

 

History

History
309 lines (239 loc) · 13.4 KB

CHANGES.rst

File metadata and controls

309 lines (239 loc) · 13.4 KB

Version history

UNRELEASED

  • Added autoincrement to primary key columns to prevent missing field errors. (#473; PR by @jtmonroe)

4.0.3

  • Improved rendering of Identity server defaults by explicitly rendering non-default parameters; Decimal values (as returned by some databases) are now cast to int (PR by @NotCarlosSerrano)

4.0.2

  • Fixed rendering of inherited keyword arguments for dialect-specific types that use **kwargs in their initializers (such as MySQL CHAR with collation) while preserving existing *args rendering behavior (PR by @hyoj0942)
  • Fixed missing metadata argument when rendering plain tables with the SQLModel
  • Added support for self-referential tables in the SQLModel generator (PR by @sheinbergon)
  • Fixed empty dialect kwargs (e.g. postgresql_include=[]) being included in rendered indexes, tables, and columns (PR by @sheinbergon)

4.0.1

  • Fix enum column definitions to explicitly include schema and name if reflected via SQLAlchemy's Metadata (pr by @sheinbergon)

4.0.0

  • BACKWARD INCOMPATIBLE API changes (for those who customize code generation by subclassing the existing generators):
    • Added new optional keyword argument, explicit_foreign_keys to DeclarativeGenerator, to force foreign keys to be rendered as ClassName.attribute_name string references
    • Removed the render_relationship_args() method from the SQLModel generator
    • Added two new methods for customizing relationship rendering in DeclarativeGenerator:
      • render_relationship_annotation(): returns the appropriate type annotation (without the Mapped wrapper) for the relationship
      • render_relationship_arguments(): returns a dictionary of keyword arguments to sqlalchemy.orm.relationship()

4.0.0rc3

  • BACKWARD INCOMPATIBLE Relationship names changed when multiple FKs or junction tables connect to the same target table. Regenerating models will break existing code.
  • Added support for generating Python enum classes for ARRAY(Enum(...)) columns (e.g., PostgreSQL ARRAY(ENUM)). Supports named/unnamed enums, shared enums across columns, and multi-dimensional arrays. Respects --options nonativeenums. (PR by @sheinbergon)
  • Improved relationship naming: one-to-many uses FK column names (e.g., simple_items_parent_container), many-to-many uses junction table names (e.g., students_enrollments). Use --options nofknames to revert to old behavior. (PR by @sheinbergon)
  • Fixed Index kwargs (e.g. mysql_length) being ignored during code generation (PR by @luliangce)
  • Fixed the SQLModel generator not adding the foreign_keys parameters when generating multiple relationships between the same two tables

4.0.0rc2

  • Add values_callable lambda to generated native enums column definitions. This allows for proper enum value insertion when working with ORM models (PR by @sheinbergon)

4.0.0rc1

  • BACKWARD INCOMPATIBLE TablesGenerator.render_column_type() was changed to receive the Column object instead of the column type object as its sole argument
  • Added Python enum generation for native database ENUM types (e.g., PostgreSQL / MySQL ENUM). Retained synthetic Python enum generation from CHECK constraints with IN clauses (e.g., column IN ('val1', 'val2', ...)). Use --options nonativeenums to disable enum generation for native database enums. Use --options nosyntheticenums to disable enum generation for synthetic database enums (VARCHAR columns with check constraints). (PR by @sheinbergon)

3.2.0

  • Dropped support for Python 3.9
  • Fix Postgres DOMAIN adaptation regression introduced in SQLAlchemy 2.0.42 (PR by @sheinbergon)
  • Support disabling special naming logic for single column many-to-one and one-to-one relationships (PR by @Henkhogan, revised by @sheinbergon)
  • Add include_dialect_options option to render Table and Column dialect-specific kwargs and info in generated code. (PR by @jaogoy)
  • Add keep_dialect_types option to preserve dialect-specific column types instead of adapting to generic SQLAlchemy types. (PR by @jaogoy)

3.1.1

  • Fallback NotImplemented errors encountered when accessing python_type for non-native types to typing.Any (PR by @sheinbergon, based on work by @danplischke)

3.1.0

  • Type annotations for ARRAY column attributes now include the Python type of the array elements
  • Added support for specifying engine arguments via --engine-arg (PR by @LajosCseppento)
  • Fixed incorrect package name used in importlib.metadata.version for sqlalchemy-citext, resolving PackageNotFoundError (PR by @oaimtiaz)
  • Prevent double pluralization (PR by @dkratzert)
  • Fixes DOMAIN extending JSON/JSONB data types (PR by @sheinbergon)
  • Temporarily restrict SQLAlchemy version to 2.0.41 (PR by @sheinbergon)
  • Fixes add_import behavior when adding imports from sqlalchemy and overall better alignment of import behavior(s) across generators
  • Fixes nullable column behavior for non-null columns for both sqlmodels and declarative generators (PR by @sheinbergon)

3.0.0

  • Dropped support for Python 3.8
  • Changed nullable relationships to include Optional in their type annotations
  • Fixed SQLModel code generation
  • Fixed two rendering issues in ENUM columns when a non-default schema is used: an unwarranted positional argument and missing the schema argument
  • Fixed AttributeError when metadata contains user defined column types
  • Fixed AssertionError when metadata contains a column type that is a type decorator with an all-uppercase name
  • Fixed MySQL DOUBLE column types being rendered with the wrong arguments

3.0.0rc5

  • Fixed pgvector support not working

3.0.0rc4

  • Dropped support for Python 3.7
  • Dropped support for SQLAlchemy 1.x
  • Added support for the pgvector extension (with help from KellyRousselHoomano)

3.0.0rc3

  • Added support for SQLAlchemy 2 (PR by rbuffat with help from mhauru)
  • Renamed --option to --options and made its values delimited by commas
  • Restored CIText and GeoAlchemy2 support (PR by stavvy-rotte)

3.0.0rc2

  • Added support for generating SQLModel classes (PR by Andrii Khirilov)
  • Fixed code generation when a single-column index is unique or does not match the dialect's naming convention (PR by Leonardus Chen)
  • Fixed another problem where sequence schemas were not properly separated from the sequence name
  • Fixed invalid generated primary/secondaryjoin expressions in self-referential many-to-many relationships by using lambdas instead of strings
  • Fixed AttributeError when the declarative generator encounters a table name already in singular form when --option use_inflect is enabled
  • Increased minimum SQLAlchemy version to 1.4.36 to address issues with ForeignKey and indexes, and to eliminate the PostgreSQL UUID column type annotation hack

3.0.0rc1

  • Migrated all packaging/testing configuration to pyproject.toml
  • Fixed unwarranted ForeignKey declarations appearing in column attributes when there are named, single column foreign key constraints (PR by Leonardus Chen)

. Fixed KeyError when rendering an index without any columns - Fixed improper handling of schema prefixes in sequence names in server defaults - Fixed identically named tables from different schemas resulting in invalid generated

code
  • Fixed imports caused by server_default conflicting with class attribute names
  • Worked around PostgreSQL UUID columns getting Any as the type annotation

3.0.0b3

  • Dropped support for Python < 3.7
  • Dropped support for SQLAlchemy 1.3
  • Added a __main__ module which can be used as an alternate entry point to the CLI
  • Added detection for sequence use in column defaults on PostgreSQL
  • Fixed sqlalchemy.exc.InvalidRequestError when encountering a column named "metadata" (regression from 2.0)
  • Fixed missing MetaData import with DeclarativeGenerator when only plain tables are generated
  • Fixed invalid data classes being generated due to some relationships having been rendered without a default value
  • Improved translation of column names into column attributes where the column name has whitespace at the beginning or end
  • Modified constraint and index rendering to add them explicitly instead of using shortcuts like unique=True, index=True or primary=True when the constraint or index has a name that does not match the default naming convention

3.0.0b2

  • Fixed IDENTITY columns not rendering properly when they are part of the primary key

3.0.0b1

NOTE: Both the API and the command line interface have been refactored in a backwards incompatible fashion. Notably several command line options have been moved to specific generators and are no longer visible from sqlacodegen --help. Their replacement are documented in the README.

  • Dropped support for Python < 3.6
  • Added support for Python 3.10
  • Added support for SQLAlchemy 1.4
  • Added support for bidirectional relationships (use --option nobidi) to disable
  • Added support for multiple schemas via --schemas
  • Added support for IDENTITY columns
  • Disabled inflection during table/relationship name generation by default (use --option use_inflect to re-enable)
  • Refactored the old CodeGenerator class into separate generator classes, selectable via --generator
  • Refactored several command line options into generator specific options:
    • --noindexes → --option noindexes
    • --noconstraints → --option noconstraints
    • --nocomments → --option nocomments
    • --nojoined → --option nojoined (declarative and dataclass generators only)
    • --noinflect → (now the default; use --option use_inflect instead) (declarative and dataclass generators only)
  • Fixed missing import for JSONB astext_type argument
  • Fixed generated column or relationship names colliding with imports or each other
  • Fixed CompileError when encountering server defaults that contain colons (:)

2.3.0

  • Added support for rendering computed columns
  • Fixed --nocomments not taking effect (fix proposed by AzuresYang)
  • Fixed handling of MySQL SET column types (and possibly others as well)

2.2.0

  • Added support for rendering table comments (PR by David Hirschfeld)
  • Fixed bad identifier names being generated for plain tables (PR by softwarepk)

2.1.0

  • Dropped support for Python 3.4
  • Dropped support for SQLAlchemy 0.8
  • Added support for Python 3.7 and 3.8
  • Added support for SQLAlchemy 1.3
  • Added support for column comments (requires SQLAlchemy 1.2+; based on PR by koalas8)
  • Fixed crash on unknown column types (NullType)

2.0.1

  • Don't adapt dialect specific column types if they need special constructor arguments (thanks Nicholas Martin for the PR)

2.0.0

  • Refactored code for better reuse
  • Dropped support for Python 2.6, 3.2 and 3.3
  • Dropped support for SQLAlchemy < 0.8
  • Worked around a bug regarding Enum on SQLAlchemy 1.2+ (name was missing)
  • Added support for Geoalchemy2
  • Fixed invalid class names being generated (fixes #60; PR by Dan O'Huiginn)
  • Fixed array item types not being adapted or imported (fixes #46; thanks to Martin Glauer and Shawn Koschik for help)
  • Fixed attribute name of columns named metadata in mapped classes (fixes #62)
  • Fixed rendered column types being changed from the original (fixes #11)
  • Fixed server defaults which contain double quotes (fixes #7, #17, #28, #33, #36)
  • Fixed secondary= not taking into account the association table's schema name (fixes #30)
  • Sort models by foreign key dependencies instead of schema and name (fixes #15, #16)

1.1.6

  • Fixed compatibility with SQLAlchemy 1.0
  • Added an option to only generate tables

1.1.5

  • Fixed potential assignment of columns or relationships into invalid attribute names (fixes #10)
  • Fixed unique=True missing from unique Index declarations
  • Fixed several issues with server defaults
  • Fixed potential assignment of columns or relationships into invalid attribute names
  • Allowed pascal case for tables already using it
  • Switched from Mercurial to Git

1.1.4

  • Fixed compatibility with SQLAlchemy 0.9.0

1.1.3

  • Fixed compatibility with SQLAlchemy 0.8.3+
  • Migrated tests from nose to pytest

1.1.2

  • Fixed non-default schema name not being present in __table_args__ (fixes #2)
  • Fixed self referential foreign key causing column type to not be rendered
  • Fixed missing "deferrable" and "initially" keyword arguments in ForeignKey constructs
  • Fixed foreign key and check constraint handling with alternate schemas (fixes #3)

1.1.1

  • Fixed TypeError when inflect could not determine the singular name of a table for a many-to-1 relationship
  • Fixed _IntegerType, _StringType etc. being rendered instead of proper types on MySQL

1.1.0

  • Added automatic detection of joined-table inheritance
  • Fixed missing class name prefix in primary/secondary joins in relationships
  • Instead of wildcard imports, generate explicit imports dynamically (fixes #1)
  • Use the inflect library to produce better guesses for table to class name conversion
  • Automatically detect Boolean columns based on CheckConstraints
  • Skip redundant CheckConstraints for Enum and Boolean columns

1.0.0

  • Initial release