nandorKollar commented on code in PR #431:
URL: https://github.com/apache/iceberg-go/pull/431#discussion_r2132934564
##########
table/update_schema.go:
##########
@@ -0,0 +1,532 @@
+package table
+
+import (
+ "github.com/apache/iceberg-go"
+)
+
+type UpdateSchema struct {
+ base *Metadata
+ schema *iceberg.Schema
+ idToParent map[int]int
+ deletes []int
+ updates map[int]*iceberg.NestedField
+ parentToAddedIDs map[int][]int
+ addedNameToID map[string]int
+ lastColumnID int
+ allowIncompatibleChanges bool
+ identifierFields map[string]struct{}
+ caseSensitive bool
+}
+
+func NewUpdateSchema(base *Metadata, s *iceberg.Schema, lastColumnID int)
*UpdateSchema {
+ identifierFields := make(map[string]struct{})
+
+ return &UpdateSchema{
+ base: base,
+ schema: s,
+ idToParent: make(map[int]int),
+ deletes: make([]int, 0),
+ updates: make(map[int]*iceberg.NestedField),
+ parentToAddedIDs: make(map[int][]int),
+ addedNameToID: make(map[string]int),
+ lastColumnID: lastColumnID,
+ allowIncompatibleChanges: false,
+ identifierFields: identifierFields,
+ caseSensitive: true,
+ }
+}
+
+// AllowIncompatibleChanges permits incompatible schema changes.
+func (us *UpdateSchema) AllowIncompatibleChanges() *UpdateSchema {
+ us.allowIncompatibleChanges = true
+
+ return us
+}
+
+func (us *UpdateSchema) AddColumn(parent, name string, new_id int, required
bool, dataType iceberg.Type, doc string, initialDefaultValue any) *UpdateSchema
{
+ parentID := -1
+ fullName := ""
+
+ if parent != "" {
+ parentField := us.findField(parent)
+ if parentField == nil {
+ panic("Cannot find parent struct: " + parent)
+ }
+
+ // Get the parent struct's ID
+ parentID = parentField.ID
+
+ // Store the parent-child relationship
+ us.idToParent[new_id] = parentID // Add this line
+ if _, ok := us.parentToAddedIDs[parentID]; !ok {
+ us.parentToAddedIDs[parentID] = make([]int, 0)
+ }
+ us.parentToAddedIDs[parentID] =
append(us.parentToAddedIDs[parentID], new_id)
+
+ // Use full qualified name
+ fullName = parent + "." + name
+ us.addedNameToID[fullName] = new_id
+
+ parentType := parentField.Type
+
+ // ToDo: handle Nested columns
+ if nestedType, ok := parentType.(iceberg.NestedType); ok {
+ if mapType, ok := nestedType.(*iceberg.MapType); ok {
+ // fields are added to the map value type
+ vf := mapType.ValueField()
+ parentField = &vf
+ } else if listType, ok :=
nestedType.(*iceberg.ListType); ok {
+ // fields are added to the element type
+ ef := listType.ElementField()
+ parentField = &ef
+ } else if structType, ok :=
nestedType.(*iceberg.StructType); ok {
+ // fields are added to the struct
+ fields := structType.Fields()
+ parentField = &fields[len(fields)-1]
+ }
+ } else {
+ panic("Cannot add to non-nested DataType: " + parent +
": " + parentField.Type.String())
+ }
+
+ parentID = parentField.ID
+ currentField := us.findField(parent + "." + name)
+
+ for _, id := range us.deletes {
+ if id == parentID {
+ panic("Cannot add to a column that will be
deleted: " + parent)
+ }
+ }
+
+ if currentField != nil {
+ foundInDeletes := false
Review Comment:
I think with the help of `slices` package, you can rewrite this to a more
concise form:
`foundInDeletes := slices.Contains(us.deletes, currentField.ID)`
--
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
