Skip to content

fix(csharp): normalize TypeName once in TryGetSchemaFromManifest #337

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
eric-wang-1990 merged 4 commits into from
Mar 12, 2026
Merged

fix(csharp): normalize TypeName once in TryGetSchemaFromManifest #337

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
47ba26e
fix(csharp): normalize TypeName once in TryGetSchemaFromManifest; add…
eric-wang-1990 Mar 12, 2026
cc71688
test(csharp): remove redundant Spark:DataType:SqlName assertion from …
eric-wang-1990 Mar 12, 2026
4254d7c
test(csharp): remove ComplexAndDecimalTypesAreMappedCorrectly test
eric-wang-1990 Mar 12, 2026
3860e05
test(csharp): move null TypeName coverage to ColumnMetadataHelperTests
eric-wang-1990 Mar 12, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions csharp/src/StatementExecution/StatementExecutionStatement.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -482,7 +482,8 @@ private Schema GetSchemaFromManifest(ResultManifest manifest)
var fields = new List<Field>();
foreach (var column in manifest.Schema.Columns)
{
var arrowType = MapDatabricksTypeToArrowType(column.TypeName);
var typeName = column.TypeName ?? string.Empty;
var arrowType = MapDatabricksTypeToArrowType(typeName);
// Embed the SQL type name as Arrow field metadata so that consumers
// (e.g. the PowerBI connector's AdjustNativeTypes) can read it via
// the "Spark:DataType:SqlName" key — the same metadata the Databricks
Expand All @@ -494,7 +495,7 @@ private Schema GetSchemaFromManifest(ResultManifest manifest)
// it here for now. Add it if a consumer requires it (PECO-2950).
var metadata = new Dictionary<string, string>
{
["Spark:DataType:SqlName"] = ColumnMetadataHelper.GetSparkSqlName(column.TypeName ?? string.Empty)
["Spark:DataType:SqlName"] = ColumnMetadataHelper.GetSparkSqlName(typeName)
};
fields.Add(new Field(column.Name, arrowType, true, metadata));
}
Expand Down
81 changes: 81 additions & 0 deletions csharp/test/Unit/StatementExecution/StatementExecutionEmptyResultSchemaTests.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -249,6 +249,87 @@ public async Task ExecuteQuery_EmptyTable_SqlNameAliasesNormalized()
Assert.Equal("SMALLINT", fields[2].Metadata["Spark:DataType:SqlName"]);
}

[Fact]
public async Task ExecuteQuery_EmptyTable_ComplexAndDecimalTypesAreMappedCorrectly()
Copy link
Copy Markdown
Collaborator Author

@eric-wang-1990 eric-wang-1990 Mar 12, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why need add these, we already have csharp/test/ColumnMetadataHelperTests.cs which test all type mapping?

Copy link
Copy Markdown
Collaborator Author

@eric-wang-1990 eric-wang-1990 Mar 12, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good point — removed the redundant Spark:DataType:SqlName assertion for DECIMAL (already covered by ColumnMetadataHelperTests).

Kept the two tests because they cover things ColumnMetadataHelperTests doesn't:

  • ComplexAndDecimalTypesAreMappedCorrectly: exercises MapDatabricksTypeToArrowType end-to-end (e.g. DECIMAL(10,2) → Decimal128Type, ARRAY → StringType) — the existing ArrowTypesAreMappedCorrectly test doesn't include these types
  • NullTypeName_TreatedAsUnknownType: verifies null TypeName doesn't throw — not covered anywhere else

Copy link
Copy Markdown
Collaborator Author

@eric-wang-1990 eric-wang-1990 Mar 12, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You're right, removed it. The production change is just extracting a typeName local — the complex/decimal type mapping is out of scope.

Copy link
Copy Markdown
Collaborator Author

@eric-wang-1990 eric-wang-1990 Mar 12, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Moved it — added [InlineData("", "")] to the existing GetBaseTypeName and GetSparkSqlName theories in ColumnMetadataHelperTests and removed the statement-level test entirely.

{
// Arrange: DECIMAL(10,2), ARRAY, MAP, STRUCT — all must survive the schema path
var manifest = BuildManifest(
("price", "DECIMAL(10,2)"),
("tags", "ARRAY<STRING>"),
("props", "MAP<STRING,INT>"),
("addr", "STRUCT<city:STRING,zip:INT>"));

var mockClient = new Mock<IStatementExecutionClient>();
mockClient
.Setup(c => c.ExecuteStatementAsync(
It.IsAny<ExecuteStatementRequest>(),
It.IsAny<CancellationToken>()))
.ReturnsAsync(new ExecuteStatementResponse
{
StatementId = StatementId,
Status = new StatementStatus { State = "SUCCEEDED" },
Manifest = manifest,
Result = new ResultData { Attachment = null },
});

using var stmt = CreateStatement(mockClient.Object);
stmt.SqlQuery = "SELECT price, tags, props, addr FROM complex_table WHERE 0=1";

var queryResult = await stmt.ExecuteQueryAsync(CancellationToken.None);
var fields = queryResult.Stream!.Schema.FieldsList;

Assert.Equal(4, fields.Count);
Assert.IsType<Decimal128Type>(fields[0].DataType);
Assert.Equal("DECIMAL", fields[0].Metadata["Spark:DataType:SqlName"]);
// ARRAY, MAP, STRUCT are represented as STRING in Arrow for now
Assert.IsType<StringType>(fields[1].DataType);
Assert.IsType<StringType>(fields[2].DataType);
Assert.IsType<StringType>(fields[3].DataType);
}

[Fact]
public async Task ExecuteQuery_NullTypeName_TreatedAsUnknownType()
{
// Arrange: a column with a null TypeName must not throw — it falls back to StringType
var manifest = new ResultManifest
{
Format = "ARROW_STREAM",
Schema = new ResultSchema
{
Columns = new List<ColumnInfo>
{
new ColumnInfo { Name = "mystery", TypeName = null },
}
},
TotalRowCount = 0,
Chunks = new List<ResultChunk>(),
};

var mockClient = new Mock<IStatementExecutionClient>();
mockClient
.Setup(c => c.ExecuteStatementAsync(
It.IsAny<ExecuteStatementRequest>(),
It.IsAny<CancellationToken>()))
.ReturnsAsync(new ExecuteStatementResponse
{
StatementId = StatementId,
Status = new StatementStatus { State = "SUCCEEDED" },
Manifest = manifest,
Result = new ResultData { Attachment = null },
});

using var stmt = CreateStatement(mockClient.Object);
stmt.SqlQuery = "SELECT mystery FROM weird_table WHERE 0=1";

var queryResult = await stmt.ExecuteQueryAsync(CancellationToken.None);
var fields = queryResult.Stream!.Schema.FieldsList;

Assert.Single(fields);
Assert.Equal("mystery", fields[0].Name);
Assert.IsType<StringType>(fields[0].DataType);
Assert.Equal(string.Empty, fields[0].Metadata["Spark:DataType:SqlName"]);
}

[Fact]
public async Task ExecuteQuery_NullManifest_ReturnsEmptySchema()
{
Expand Down
Loading