Skip to content

Latest commit

 

History

History
123 lines (71 loc) · 3.05 KB

CHANGELOG.md

File metadata and controls

123 lines (71 loc) · 3.05 KB

[1.0.8] - 2022.03.20

misc: updating packages ci: switching to github actions

[1.0.7] - 2022.11.27

feat: add --no-bigint #17 (wirekang) feat: add --no-optional #18 (wirekang) fix: --no-write-header to --no-header (wirekang) misc: updating packages

[1.0.6] - 2022.08.29

chore: upgrading dependencies

[1.0.5] - 2022.06.26

feat: infer bigint type chore: upgrading dependencies

[1.0.4] - 2022.02.22

fix(mysql): for some versions of mysql key casing results in empty hash lookups feat(postgres): Adding types 'mol', 'bfp' and 'bit'

[1.0.3] - 2022.02.22

chore: adding ci

[1.0.2] - 2022.02.19

chore: Upgrading dependencies

[1.0.1] - 2022.02.03

fix(postgres): (bchrobot) adding missing cli command throwOnMissingType

fix(postgres): (bchrobot) typo in write-header option

[0.0.12] - 2021.11.01

fix: typo in CLI

[0.0.11] - 2021.10.31

feat: adding mysql compatability

This allows you to do the same thing just with mysql using /bin/schemats mysql $connection_string -s $schema_name

[0.0.10] - 2021.09.02

chore: updating all dependencies

feat: add -C --camelCaseTypes option

This option adds the ability to camel case just the type names - which
gives a good mix between using JS Standard Camel Case and still
following the actual definitions of database.

The issue with using camel case for both the types and the keys
is that we would have to provide a layer within the programs using the
types to convert back to the original form if the attributes are
different in JS than in the schema.

There are definately
issues with this, especially with a database schema with an inconsistent
naming convention - we would have to provide some sort of mapping file
to acheive correct conversion.

The types on the other hand, only exist in JS and therefore can be named
whatever we want when generating the types.

fix(schema): add 'tsvector' to string types

Text Search Vectors are a complex type inside of postgres, but can
generally be expressed as strings within TS.

fix(generator): quote string enum keys

This helps prevent issues in the generated file due to special
characters like `:` present in the postgres enum keys.

[0.0.9] - 2021.08.27

doc: adding example documentation fix: Don't export custom types if empty

[0.0.8] - 2021.08.22

Feat: Exporting tables and Custom types for typed-postgres

[0.0.7] - 2021.08.05

Fix: array regression due to bad merge

[0.0.6] - 2021.08.05

Feat: using the -f flag to reference a file with non DB types and adding comments to columns in postgres using COMMENT ON COLUMN schema.table.column is '@type {TYPE}'; now allows us to type jsonb columns directly

[0.0.5] - 2021.07.26

Fix: isArray overrides real value with false

[0.0.4] - 2021.07.26

Fix: publish dist and src packages

[0.0.3] - 2021.07.26

Fix: nullable fields are also optional

[0.0.2] - 2021.07.26

Fix: Adding support for arrays

[0.0.1] - 2021.06.20

Include README file in published package

[0.0.0] - 2021.06.20

First release