wgtmac commented on code in PR #106:
URL: https://github.com/apache/iceberg-cpp/pull/106#discussion_r2100379480
##########
src/iceberg/avro/avro_schema_util.cc:
##########
@@ -263,4 +265,116 @@ Status ToAvroNodeVisitor::Visit(const SchemaField& field,
::avro::NodePtr* node)
return {};
}
+namespace {
+
+bool HasId(const ::avro::NodePtr& parent_node, size_t field_idx,
+ const std::string& attr_name) {
+ if (field_idx >= parent_node->customAttributes()) {
+ return false;
+ }
+ return
parent_node->customAttributesAt(field_idx).getAttribute(attr_name).has_value();
+}
+
+} // namespace
+
+Status HasIdVisitor::Visit(const ::avro::NodePtr& node) {
+ if (!node) [[unlikely]] {
+ return InvalidSchema("Avro node is null");
+ }
+
+ switch (node->type()) {
+ case ::avro::AVRO_RECORD:
+ return VisitRecord(node);
+ case ::avro::AVRO_ARRAY:
+ return VisitArray(node);
+ case ::avro::AVRO_MAP:
+ return VisitMap(node);
+ case ::avro::AVRO_UNION:
+ return VisitUnion(node);
+ case ::avro::AVRO_BOOL:
+ case ::avro::AVRO_INT:
+ case ::avro::AVRO_LONG:
+ case ::avro::AVRO_FLOAT:
+ case ::avro::AVRO_DOUBLE:
+ case ::avro::AVRO_STRING:
+ case ::avro::AVRO_BYTES:
+ case ::avro::AVRO_FIXED:
+ return {};
+ case ::avro::AVRO_NULL:
+ case ::avro::AVRO_ENUM:
+ default:
+ return InvalidSchema("Unsupported Avro type: {}",
static_cast<int>(node->type()));
+ }
+}
+
+Status HasIdVisitor::VisitRecord(const ::avro::NodePtr& node) {
+ static const std::string kFieldIdKey{kFieldIdProp};
+ total_fields_ += node->leaves();
+ for (size_t i = 0; i < node->leaves(); ++i) {
+ if (HasId(node, i, kFieldIdKey)) {
+ fields_with_id_++;
Review Comment:
This visitor mixes the feature of `HasIds` and `MissingIds` from the Java
implementation so I can check them in a single pass. I think an Avro schema
should be considered valid when all or none of fields have field-ids, but not
in between.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
