zeroshade commented on code in PR #356:
URL: https://github.com/apache/iceberg-go/pull/356#discussion_r2019216734
##########
table/table_test.go:
##########
@@ -732,6 +732,106 @@ func (t *TableWritingTestSuite)
TestAddFilesReferencedCurrentSnapshotIgnoreDupli
t.Equal([]int32{0, 0, 0}, deleted)
}
+type mockedCatalog struct{}
+
+func (m *mockedCatalog) LoadTable(ctx context.Context, ident table.Identifier,
props iceberg.Properties) (*table.Table, error) {
+ return nil, nil
+}
+
+func (m *mockedCatalog) CommitTable(ctx context.Context, tbl *table.Table,
reqs []table.Requirement, updates []table.Update) (table.Metadata, string,
error) {
+ bldr, err := table.MetadataBuilderFromBase(tbl.Metadata())
+ if err != nil {
+ return nil, "", err
+ }
+
+ for _, u := range updates {
+ if err := u.Apply(bldr); err != nil {
+ return nil, "", err
+ }
+ }
+
+ meta, err := bldr.Build()
+ if err != nil {
+ return nil, "", err
+ }
+
+ return meta, "", nil
+}
+
+func (t *TableWritingTestSuite) TestReplaceDataFiles() {
+ fs := iceio.LocalFS{}
+
+ files := make([]string, 0)
+ for i := range 5 {
+ filePath :=
fmt.Sprintf("%s/replace_data_files_v%d/data-%d.parquet", t.location,
t.formatVersion, i)
+ t.writeParquet(fs, filePath, t.arrTablePromotedTypes)
+ files = append(files, filePath)
+ }
+
+ ident := table.Identifier{"default", "replace_data_files_v" +
strconv.Itoa(t.formatVersion)}
+ meta, err := table.NewMetadata(t.tableSchemaPromotedTypes,
iceberg.UnpartitionedSpec,
+ table.UnsortedSortOrder, t.location,
iceberg.Properties{"format-version": strconv.Itoa(t.formatVersion)})
+ t.Require().NoError(err)
+
+ ctx := context.Background()
+
+ tbl := table.New(ident, meta, t.getMetadataLoc(), fs, &mockedCatalog{})
+ for i := range 5 {
+ tx := tbl.NewTransaction()
+ t.Require().NoError(tx.AddFiles(files[i:i+1], nil, false))
+ tbl, err = tx.Commit(ctx)
+ t.Require().NoError(err)
+ }
+
+ mflist, err := tbl.CurrentSnapshot().Manifests(tbl.FS())
+ t.Require().NoError(err)
+ t.Len(mflist, 5)
+
+ // create a parquet file that is essentially as if we merged two of
+ // the data files together
+ cols := make([]arrow.Column, 0, t.arrTablePromotedTypes.NumCols())
+ for i := range int(t.arrTablePromotedTypes.NumCols()) {
+ chkd := t.arrTablePromotedTypes.Column(i).Data()
+ duplicated := arrow.NewChunked(chkd.DataType(),
append(chkd.Chunks(), chkd.Chunks()...))
+ defer duplicated.Release()
+
+ col := arrow.NewColumn(t.arrSchemaPromotedTypes.Fields()[i],
duplicated)
+ defer col.Release()
+
+ cols = append(cols, *col)
+ }
+
+ combined := array.NewTable(t.arrSchemaPromotedTypes, cols, -1)
+ defer combined.Release()
+
+ combinedFilePath :=
fmt.Sprintf("%s/replace_data_files_v%d/combined.parquet", t.location,
t.formatVersion)
+ t.writeParquet(fs, combinedFilePath, combined)
+
+ tx := tbl.NewTransaction()
+ t.Require().NoError(tx.ReplaceDataFiles(files[:2],
[]string{combinedFilePath}, nil))
Review Comment:
Yup that's exactly what's happening and what we see on the snapshot summary
(which is being confirmed in the test just below where you commented). Are you
saying I should add a comment with the explanation?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For additional commands, e-mail: issues-h...@iceberg.apache.org
