pitrou commented on code in PR #36344:
URL: https://github.com/apache/arrow/pull/36344#discussion_r1247480486
##########
cpp/src/arrow/ipc/reader.cc:
##########
@@ -951,60 +977,85 @@ class RecordBatchStreamReaderImpl : public
RecordBatchStreamReader {
return Status::OK();
}
- Status ReadInitialDictionaries() {
- // We must receive all dictionaries before reconstructing the
- // first record batch. Subsequent dictionary deltas modify the memo
- std::unique_ptr<Message> message;
-
- // TODO(wesm): In future, we may want to reconcile the ids in the stream
with
- // those found in the schema
- const auto num_dicts = dictionary_memo_.fields().num_dicts();
- for (int i = 0; i < num_dicts; ++i) {
- ARROW_ASSIGN_OR_RAISE(message, ReadNextMessage());
- if (!message) {
- if (i == 0) {
- /// ARROW-6006: If we fail to find any dictionaries in the stream,
then
- /// it may be that the stream has a schema but no actual data. In
such
- /// case we communicate that we were unable to find the dictionaries
- /// (but there was no failure otherwise), so the caller can decide
what
- /// to do
- empty_stream_ = true;
- break;
- } else {
- // ARROW-6126, the stream terminated before receiving the expected
- // number of dictionaries
- return Status::Invalid("IPC stream ended without reading the
expected number (",
- num_dicts, ") of dictionaries");
- }
- }
+ std::shared_ptr<Listener> listener_;
+ const IpcReadOptions options_;
+ State state_;
+ std::vector<bool> field_inclusion_mask_;
+ int num_required_initial_dictionaries_;
+ int num_read_initial_dictionaries_;
+ DictionaryMemo dictionary_memo_;
+ std::shared_ptr<Schema> schema_;
Review Comment:
That's a good idea, thanks for suggesting it.
--
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]
