Merge 95ed437 into 682dadf

Author: mukunku

.github/workflows/build-test-publish.yaml

@@ -2,6 +2,11 @@ name: Test and Publish
 
 on:
  pull_request:
+    types:
+      - opened
+      - reopened
+      - synchronize
+      - ready_for_review
 
 permissions:
   checks: write
@@ -46,7 +51,7 @@ jobs:
   checkPublish:
     runs-on: windows-latest
     needs: test
-    if: github.repository == 'mukunku/ParquetViewer' 
+    if: github.repository == 'mukunku/ParquetViewer' && github.event.pull_request.draft == false
     outputs:
       release_version: ${{ steps.release-version.outputs.release_version }}
       should_publish: ${{ steps.should-publish.outputs.should_publish }}
@@ -141,26 +146,9 @@ jobs:
           wait-for-completion: true
           output-artifact-directory: '/signed-package'
 
-      - name: Generate _checksums.txt
-        run: |
-          $fileHash = (Get-FileHash "signed-package/ParquetViewer.exe" -Algorithm SHA256)
-          $fileHashSelfContained = (Get-FileHash "signed-package/ParquetViewer_SelfContained.exe" -Algorithm SHA256)
-          $fileSize = (Get-Item -Path "signed-package/ParquetViewer.exe").Length
-          $fileSizeSelfContained = (Get-Item -Path "signed-package/ParquetViewer_SelfContained.exe").Length
-          "v${{ env.VERSION_NUMBER }}" >> "_checksums.txt"
-          "" >> "_checksums.txt"
-          "Name: ParquetViewer.exe" >> "_checksums.txt"
-          "Size: $fileSize bytes ($([math]::floor($fileSize/1KB)) KiB)" >> "_checksums.txt"
-          "SHA256: $($fileHash.Hash)" >> "_checksums.txt"
-          "" >> "_checksums.txt"
-          "Name: ParquetViewer_SelfContained.exe" >> "_checksums.txt"
-          "Size: $fileSizeSelfContained bytes ($([math]::floor($fileSizeSelfContained/1MB)) MiB)" >> "_checksums.txt"
-          "SHA256: $($fileHashSelfContained.Hash)" >> "_checksums.txt"
-          "" >> "_checksums.txt"
-
       - uses: ncipollo/release-action@v1
         with:
-          artifacts: "signed-package/ParquetViewer.exe,signed-package/ParquetViewer_SelfContained.exe,_checksums.txt"
+          artifacts: "signed-package/ParquetViewer.exe,signed-package/ParquetViewer_SelfContained.exe"
           body: "PR: #${{ env.PR_NUMBER }}"
           allowUpdates: ${{ env.BRANCH_NAME != 'main' }}
           omitBodyDuringUpdate: true

src/Directory.Packages.props

@@ -3,8 +3,8 @@
     <ManagePackageVersionsCentrally>true</ManagePackageVersionsCentrally>
   </PropertyGroup>
   <ItemGroup>
-    <PackageVersion Include="Apache.Arrow" Version="20.0.0" />
-    <PackageVersion Include="Parquet.Net" Version="5.1.1" />
+    <PackageVersion Include="Apache.Arrow" Version="21.0.0" />
+    <PackageVersion Include="Parquet.Net" Version="5.2.0" />
     <PackageVersion Include="Microsoft.CSharp" Version="4.7.0" />
     <PackageVersion Include="Microsoft.NET.Test.Sdk" Version="17.14.0" />
     <PackageVersion Include="RichardSzalay.MockHttp" Version="7.0.0" />

src/ParquetViewer.Engine/DataTableLite.cs

@@ -1,4 +1,5 @@
-using System.Data;
+using ParquetViewer.Engine.Exceptions;
+using System.Data;
 
 namespace ParquetViewer.Engine
 {
@@ -83,8 +84,31 @@ public DataTable ToDataTable(CancellationToken token, IProgress<int>? progress =
             {
                 token.ThrowIfCancellationRequested();
 
-                //supposedly this is the fastest way to load data into a datatable https://stackoverflow.com/a/17123914/1458738
-                dataTable.LoadDataRow(_rows[i]!, false);
+                try
+                {
+                    //supposedly this is the fastest way to load data into a datatable https://stackoverflow.com/a/17123914/1458738
+                    dataTable.LoadDataRow(_rows[i]!, false);
+                }
+                catch (Exception ex)
+                {
+                    if (ex.Message.Contains("Type of value has a mismatch with column type"))
+                    {
+                        //Try figure out where the mismatch is
+                        var columnIndex = 0;
+                        foreach (var column in this._columns.Values)
+                        {
+                            if (_rows[i][columnIndex] != DBNull.Value && column.Type != _rows[i][columnIndex].GetType())
+                            {
+                                throw new TypeMismatchException($"Value type '{_rows[i][columnIndex]?.GetType()}' doesn't match column type {column.Type} for field `{column.Name}`");
+                            }
+                            columnIndex++;
+                        }
+
+                        throw new TypeMismatchException(null, ex);
+                    }
+
+                    throw;
+                }
 
                 progress?.Report(_columns.Count);
             }

src/ParquetViewer.Engine/Exceptions/TypeMismatchException.cs

@@ -0,0 +1,10 @@
+namespace ParquetViewer.Engine.Exceptions
+{
+    public class TypeMismatchException : Exception
+    {
+        public TypeMismatchException(string? message, Exception? ex = null) : base(message, ex)
+        {
+
+        }
+}
+}