You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@iceberg.apache.org by at...@apache.org on 2026/08/07 15:56:31 UTC

(iceberg-go) branch main updated: fix(codec): reject nil data files in scan tasks (#1611)

This is an automated email from the ASF dual-hosted git repository.

laskoviymishka pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/iceberg-go.git


The following commit(s) were added to refs/heads/main by this push:
     new 687852e7 fix(codec): reject nil data files in scan tasks (#1611)
687852e7 is described below

commit 687852e7a831e2d5fc5862f2fa88473d4006ae60
Author: Minh Vu <vu...@gmail.com>
AuthorDate: Fri Aug 7 17:56:25 2026 +0200

    fix(codec): reject nil data files in scan tasks (#1611)
    
    * fix(codec): reject nil data files in scan tasks
    
    Signed-off-by: Minh Vu <vu...@gmail.com>
    
    * fix(codec): reject typed-nil data files
    
    * chore(ci): retrigger pull request checks
    
    ---------
    
    Signed-off-by: Minh Vu <vu...@gmail.com>
---
 codec/file_scan_task.go      | 18 ++++++++++++++++
 codec/file_scan_task_test.go | 49 ++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 67 insertions(+)

diff --git a/codec/file_scan_task.go b/codec/file_scan_task.go
index 0c681a20..43001ca5 100644
--- a/codec/file_scan_task.go
+++ b/codec/file_scan_task.go
@@ -23,6 +23,7 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"reflect"
 
 	"github.com/apache/iceberg-go"
 	"github.com/apache/iceberg-go/table"
@@ -255,6 +256,9 @@ func decodeFileScanTaskResidual(data []byte, schema *iceberg.Schema) (iceberg.Bo
 // carrying a different spec would silently mis-map or drop partition values and
 // still succeed. Every file in a FileScanTask must therefore share spec.ID().
 func checkDataFileSpecID(f iceberg.DataFile, spec iceberg.PartitionSpec) error {
+	if isNilDataFile(f) {
+		return fmt.Errorf("%w: data file is nil", iceberg.ErrInvalidArgument)
+	}
 	if int(f.SpecID()) != spec.ID() {
 		return fmt.Errorf("data file spec id %d does not match codec spec id %d (partition evolution requires per-spec grouping)", f.SpecID(), spec.ID())
 	}
@@ -262,6 +266,20 @@ func checkDataFileSpecID(f iceberg.DataFile, spec iceberg.PartitionSpec) error {
 	return nil
 }
 
+func isNilDataFile(f iceberg.DataFile) bool {
+	if f == nil {
+		return true
+	}
+
+	v := reflect.ValueOf(f)
+	switch v.Kind() {
+	case reflect.Chan, reflect.Func, reflect.Interface, reflect.Map, reflect.Pointer, reflect.Slice, reflect.UnsafePointer:
+		return v.IsNil()
+	default:
+		return false
+	}
+}
+
 func encodeDataFileSlice(files []iceberg.DataFile, spec iceberg.PartitionSpec, schema *iceberg.Schema, version int) ([][]byte, error) {
 	if len(files) == 0 {
 		return nil, nil
diff --git a/codec/file_scan_task_test.go b/codec/file_scan_task_test.go
index ac7058fb..576d1762 100644
--- a/codec/file_scan_task_test.go
+++ b/codec/file_scan_task_test.go
@@ -137,6 +137,55 @@ func TestEncodeFileScanTaskRejectsMismatchedPrimaryDataFileSpec(t *testing.T) {
 		"error must include the codec's spec id")
 }
 
+func TestEncodeFileScanTaskRejectsNilDataFiles(t *testing.T) {
+	spec, schema, task := fullyPopulatedFileScanTask(t, 2)
+	tests := []struct {
+		name   string
+		update func(*table.FileScanTask)
+		part   string
+	}{
+		{"primary file", func(task *table.FileScanTask) { task.File = nil }, "data file is nil"},
+		{"delete file", func(task *table.FileScanTask) { task.DeleteFiles = []iceberg.DataFile{nil} }, "delete files: entry 0"},
+		{"equality delete file", func(task *table.FileScanTask) { task.EqualityDeleteFiles = []iceberg.DataFile{nil} }, "equality delete files: entry 0"},
+		{"deletion vector file", func(task *table.FileScanTask) { task.DeletionVectorFiles = []iceberg.DataFile{nil} }, "deletion vector files: entry 0"},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			task := task
+			tt.update(&task)
+			_, err := codec.EncodeFileScanTask(task, spec, schema, 2)
+			require.ErrorIs(t, err, iceberg.ErrInvalidArgument)
+			require.ErrorContains(t, err, tt.part)
+		})
+	}
+}
+
+func TestEncodeFileScanTaskRejectsTypedNilDataFiles(t *testing.T) {
+	spec, schema, task := fullyPopulatedFileScanTask(t, 3)
+	var typedNil *stubDataFile
+	tests := []struct {
+		name   string
+		update func(*table.FileScanTask)
+		part   string
+	}{
+		{"primary file", func(task *table.FileScanTask) { task.File = typedNil }, "data file is nil"},
+		{"delete file", func(task *table.FileScanTask) { task.DeleteFiles = []iceberg.DataFile{typedNil} }, "delete files: entry 0"},
+		{"equality delete file", func(task *table.FileScanTask) { task.EqualityDeleteFiles = []iceberg.DataFile{typedNil} }, "equality delete files: entry 0"},
+		{"deletion vector file", func(task *table.FileScanTask) { task.DeletionVectorFiles = []iceberg.DataFile{typedNil} }, "deletion vector files: entry 0"},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			task := task
+			tt.update(&task)
+			_, err := codec.EncodeFileScanTask(task, spec, schema, 3)
+			require.ErrorIs(t, err, iceberg.ErrInvalidArgument)
+			require.ErrorContains(t, err, tt.part)
+		})
+	}
+}
+
 func TestDecodeFileScanTaskErrorCarriesMarker(t *testing.T) {
 	schema := iceberg.NewSchema(123,
 		iceberg.NestedField{ID: 1, Name: "id", Type: iceberg.Int64Type{}, Required: true},