You are viewing a plain text version of this content. The canonical link for it is here.
Posted to jira@arrow.apache.org by "Chao Sun (Jira)" <ji...@apache.org> on 2021/09/08 22:46:00 UTC

[jira] [Created] (ARROW-13941) [C++] Parquet writes incorrect file_offset

Chao Sun created ARROW-13941:
--------------------------------

             Summary: [C++] Parquet writes incorrect file_offset 
                 Key: ARROW-13941
                 URL: https://issues.apache.org/jira/browse/ARROW-13941
             Project: Apache Arrow
          Issue Type: Bug
          Components: Parquet
            Reporter: Chao Sun


Currently the Parquet writer set {{file_offset}} in the following way:
{code:cpp}
    if (dictionary_page_offset > 0) {
      column_chunk_->meta_data.__set_dictionary_page_offset(dictionary_page_offset);
      column_chunk_->__set_file_offset(dictionary_page_offset + compressed_size);
    } else {
      column_chunk_->__set_file_offset(data_page_offset + compressed_size);
    }{code}
This doesn't look correct, and seems it should not take {{compressed_size}} into consideration.

The {{file_offset}} is used when filtering row groups, and the issue could cause correctness issue. See SPARK-36696.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)