Prevent setting chunk paths if not specified

rapidsai · Dec 9, 2021 · 200d1b0 · 200d1b0
1 parent 8c2927d
commit 200d1b0
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 9 deletions.
diff --git a/cpp/include/cudf/io/parquet.hpp b/cpp/include/cudf/io/parquet.hpp
@@ -485,7 +485,7 @@ class parquet_writer_options {
    * @param partitions Partitions of input table in {start_row, num_rows} pairs. If specified, must
    * be same size as number of sinks in sink_info
    */
-  void set_partitions(std::vector<partition_info> const& partitions)
+  void set_partitions(std::vector<partition_info> partitions)
   {
     CUDF_EXPECTS(partitions.size() == _sink.num_sinks,
                  "Mismatch between number of sinks and number of partitions");
@@ -602,7 +602,7 @@ class parquet_writer_options_builder {
   {
     CUDF_EXPECTS(partitions.size() == options._sink.num_sinks,
                  "Mismatch between number of sinks and number of partitions");
-    options._partitions = std::move(partitions);
+    options.set_partitions(std::move(partitions));
     return *this;
   }
 
@@ -664,12 +664,11 @@ class parquet_writer_options_builder {
    * data sinks
    * @return this for chaining.
    */
-  parquet_writer_options_builder& column_chunks_file_paths(
-    std::vector<std::string> const& file_paths)
+  parquet_writer_options_builder& column_chunks_file_paths(std::vector<std::string> file_paths)
   {
     CUDF_EXPECTS(file_paths.size() == options._sink.num_sinks,
                  "Mismatch between number of sinks and number of chunk paths to set");
-    options._column_chunks_file_paths = file_paths;
+    options.set_column_chunks_file_paths(std::move(file_paths));
     return *this;
   }
 
@@ -956,7 +955,7 @@ class chunked_parquet_writer_options_builder {
   {
     CUDF_EXPECTS(metadata.size() == options._sink.num_sinks,
                  "Mismatch between number of sinks and number of metadata maps");
-    options._user_data = std::move(metadata);
+    options.set_key_value_metadata(std::move(metadata));
     return *this;
   }
 

diff --git a/python/cudf/cudf/_lib/parquet.pyx b/python/cudf/cudf/_lib/parquet.pyx
@@ -337,8 +337,6 @@ cpdef write_parquet(
     cdef unique_ptr[vector[uint8_t]] out_metadata_c
     cdef vector[string] c_column_chunks_file_paths
     cdef bool _int96_timestamps = int96_timestamps
-    if metadata_file_path is not None:
-        c_column_chunks_file_paths.push_back(str.encode(metadata_file_path))
 
     # Perform write
     cdef parquet_writer_options args = move(
@@ -347,10 +345,12 @@ cpdef write_parquet(
         .key_value_metadata(move(user_data))
         .compression(comp_type)
         .stats_level(stat_freq)
-        .column_chunks_file_paths(c_column_chunks_file_paths)
         .int96_timestamps(_int96_timestamps)
         .build()
     )
+    if metadata_file_path is not None:
+        c_column_chunks_file_paths.push_back(str.encode(metadata_file_path))
+        args.set_column_chunks_file_paths(move(c_column_chunks_file_paths))
     if row_group_size_bytes is not None:
         args.set_row_group_size_bytes(row_group_size_bytes)
     if row_group_size_rows is not None: