v1.2.1 correlated subquery support

New for v1.2.1:

SqlHydra.Query

• Support for correlated subqueries (#51). Thanks to @ntwilson for this feature!
  readme

SqlHydra CLI

• Fixes issue #50 for the SqlHydra CLI generators.

v1.2.0 Table Declarations

SqlHydra v1.2.0 now generates a table declaration for each table, along with the generated table record.

For example, if you have a table, Sales.Person, it will now generate a table declaration, Sales.Person.
Behind the scenes in the generated code, this look like this:

module Sales =

    [<CLIMutable>]
    type Person = { FName: string; LName: string }
    
    let Person = SqlHydra.Query.Table.table<Person>

You can then use this in your queries:

  select {
      for p in Sales.Person do
      select p
  }

How to enable feature

Starting a new project

If running the sqlhydra CLI tool for the first time, you will be prompted to "Select a use case" with three options:

• "SqlHydra.Query integration (default)"
• "Other data library"
• "Standalone"

This option will only be enabled if you choose "SqlHydra.Query integration (default)".

Upgrading from a previous version

If upgrading a previous version, you will need to manually edit your .toml file in your project folder to add table_declarations = true within the sqlhydra_query_integration section.

Example:

[general]
connection = "Data Source=TestData/AdventureWorksLT.db"
output = "Sqlite/AdventureWorksNet6.fs"
namespace = "Sqlite.AdventureWorksNet6"
cli_mutable = true
[sqlhydra_query_integration]
provider_db_type_attributes = true
table_declarations = true
[readers]
reader_type = "System.Data.Common.DbDataReader"
[filters]
include = [ "*" ]
exclude = [ "hr/*" ]

See .toml configuration docs for more details:

SqlHydra.Query v1.1.2

Adds support in select statement where clauses for reversing order of column/value. Previously, the table column/property always had to be on left side of the comparison:

  select {
      for a in addressTable do
      where (a.AddressID > 5)
  }

Now it can be on either side:

  select {
      for a in addressTable do
      where (5 < a.AddressID && a.AddressID <= 20)
  }

Thanks to @ntwilson for this improvement! 🎉

v1.1.1 - Simplified CLI wizard with "use cases"

v1.1.1 updates all the SqlHydra code generator tools to simplify the TOML configuration wizard that runs when a TOML file is not detected.
This update establishes three main "use cases" supported by SqlHydra.
Please see the updated TOML wiki page for more details:

https://github.com/JordanMarr/SqlHydra/wiki/TOML-Configuration#generating-toml-file

v1.1.0 / .NET 7 / Same Table Joins

SqlHydra.Query and SqlHydra.* (code gen)

• All packages have been updated to v1.1.0
• All packages support .NET 5, .NET 6 and now .NET 7

SqlHydra.Query

• Same table joins are now supported.
  • Behind the scenes, all select queries now use table aliases instead of fully qualified table names (required to support same table joins).
• inSchema function is no longer needed (now deprecated/obsolete because schema can be inferred from generated types parent module).
  • This makes it easier to specify your tables within your query instead of doing it ahead of time:

    select {
        for order in table<dbo.Orders> do
        select token
    }

• Updated to SqlKata v2.4.0.

Thanks to @jwosty for his huge contributions toward the same table join feature!

Support for Implicit Conversions

SqlHydra.Query v1.0.5 adds support for F# implicit conversions in where expressions.

Example 1: Comparing a decimal value (p.ListPrice) with an int constant now works as expected:

select {
    for p in productTable do
    where (p.ListPrice > 5)
}

Example 2: Comparing a decimal option (p.Weight) with an int constant now works as expected:

select {
    for p in productTable do
    where (p.Weight = Some 5)
}

Thanks to aciq for his awesome pull request!

Npgsql Support for (more) Arrays

SqlHydra.Npgsql v1.0.5 adds adds support for arrays for the remaining PostgreSQL primitive data types, including:

• boolean
• smallint
• integer
• bigint
• real
• double precision
• numeric
• money
• text
• bit(1)
• date
• date without timezone
• timestamp with time zone
• time with time zone

Npgsql Support for Arrays

SqlHydra.Npgsql v1.0.4 adds support for text[] and integer[] columns.

There are other types that can support arrays, but I wanted to add the requested types first.
If you would like to see more array types added, please post an issue that we can use to come up with a more comprehensive list.

SQL Server getId uniqueidentifier fix

Fixes getId for SQL Server inserts when using GUID/UNIQUEIDENTIFIER column.
#38

SQL Server DATETIME2 Fix

This release affects SqlHydra.SqlServer and SqlHydra.Query so that the SQL Server DATETIME2 parameter type is properly detected and set based on the generated types. (Previously, DATETIME2 columns were being set as DATETIME parameter types which caused a loss of precision.)

SqlHydra.SqlServer v1.0.3

• Generated record properties now specify the ProviderDbType attribute for DateTime fields with either DATETIME or DATETIME2.

SqlHydra.Query v1.0.3

• The ProviderDbType has now been fixed to check for generated ProviderDbType attributes in both where and having parameters as well as in updates and inserts.

Thanks to @ntwilson for contributing this fix!