Changes for 0.12.0 release #70
Conversation
|
I’ve also added the setting to resolve #68 but it didn’t seem to be working in the unit test - unsure if I’m misunderstanding what it’s supposed to do (or DynamoDB-local doesn’t support it) |
| @@ -104,6 +100,14 @@ type ``Conditional Expression Tests`` (fixture : TableFixture) = | |||
| fun () -> table.PutItem(item, precondition = itemDoesNotExist) | |||
| |> shouldFailwith<_, ConditionalCheckFailedException> | |||
|
|
|||
| let [<Fact(Skip = "Not sure if this should be working")>] ``Item not exists failure should add conflicting item data to the exception`` () = | |||
There was a problem hiding this comment.
this looks reasonable to me - don't have a prod env to hand so I guess that points the finget of blame at the simulators...
RELEASE_NOTES.md
Outdated
| @@ -1,3 +1,10 @@ | |||
| ### 0.11.3-beta | |||
There was a problem hiding this comment.
A .12 might not be unreasonable as there are new supported expression features (and a removal of an obsoletion)
chore: Switch to PackageLicenseExpression
…WS.DynamoDB into multi-key-attributes
| let item = mkItem() | ||
| let _key = table.PutItem(item, precondition = itemDoesNotExist) | ||
| try | ||
| table.PutItem(item, precondition = itemDoesNotExist) |> ignore |
There was a problem hiding this comment.
@samritchie Hm, probably should assert that it does not just succeed
by replacing the try-with with a single raisesWith expression
| @@ -1,27 +1,23 @@ | |||
| <Project Sdk="Microsoft.NET.Sdk"> | |||
| <PropertyGroup> | |||
| <TargetFramework>net6.0</TargetFramework> | |||
| <TargetFramework>net8.0</TargetFramework> | |||
| <IsPackable>false</IsPackable> | |||
There was a problem hiding this comment.
I should have removed this in my PR - has not been required since SDK v 6 (MS.Test reference sets this flag by default)
|
|
||
| type ``Inverse GSI Table Operation Tests`` (fixture : TableFixture) = | ||
|
|
||
| let rand = let r = Random() in fun () -> int64 <| r.Next() |
There was a problem hiding this comment.
.NET 7 has a Shard.Random which is threadsafe (and also means the seed can't be identical if lots of instances of the TestClass get constructed in the same timeslice)
| let [<Fact>] ``Query by Table Key and GSI`` () = | ||
| let values = set [ for _ in 1L .. 1000L -> mkItem() ] | ||
| for batch in values |> Set.toSeq |> Seq.chunkBySize 25 do | ||
| table.BatchPutItems batch |> ignore |
There was a problem hiding this comment.
|> ignores give me kittens in general. I guess the assertions are the point here, but I'd be tempted to assign to a temp and then do a skipped =! Array.empry
|
|
||
| let [<Fact>] ``Shared GSI Range Keys should be permitted`` () = | ||
| let template = RecordTemplate.Define<SameRangeKeyGSI>() | ||
| test <@ template.GlobalSecondaryIndices[0].RangeKey.IsSome @> |
There was a problem hiding this comment.
For future reference, IsSome is a big clue that you should match....
test <@ match template.GlobalSecondaryIndices[0].RangeKey, template.GlobalSecondaryIndices[1].RangeKey with
| Some r1, Some r2 -> r1 = r2
| _ -> false @>
OR
let rk1, rk2 = trap <@ template.GlobalSecondaryIndices[0].RangeKey |> Option.get
template.GlobalSecondaryIndices[1].RangeKey |> Option.get @>
r1 =! r2
BinaryFormatterAttribute