ARROW-5893: [C++][Python][GLib][Ruby][MATLAB][R] Remove arrow::Column…

… class This completely removes `arrow::Column` from the C++ library and adapts the Python bindings to follow, to help assist with mailing list discussion. There are other places that this would touch: - [x] GLib - [x] Ruby - [x] MATLAB - [x] R You can see the evident reduction of boilerplate and simplification in many places -- it is particularly pronounced in parquet/arrow/arrow-reader-writer-test.cc. This refactor also exposed a bug where a Column's data type did not match its Field type. Closes #4841 from wesm/remove-column-class and squashes the following commits: b389d42 <Wes McKinney> Fix up Python/C++ docs to remove references to arrow::Column cc34548 <Wes McKinney> Fix non-deterministic Table.from_pydict behavior. Add unicode/utf8 test for pyarrow.table. Fix usages of ChunkedArray.data and add unit test 4be8d5a <Wes McKinney> UTF-8 py2/3 compatibility issues 9b5a61e <Wes McKinney> Try to fix unicode issue a5ee7b6 <Wes McKinney> Fix MATLAB code (I hope) 2c7c51e <Wes McKinney> code review comments 020f52a <Wes McKinney> Re-render R README fd4473e <Wes McKinney> Fix up R C++ code and unit tests 244d2a7 <Wes McKinney> Begin refactoring R library. Change Feather to return ChunkedArray from GetColumn d30a365 <Sutou Kouhei> Remove unused variable d89366a <Sutou Kouhei> Follow API change 983d378 <Sutou Kouhei> Follow API change 3099328 <Sutou Kouhei> Follow API change f092f1b <Sutou Kouhei> Add missing available annotations 41dd1fe <Sutou Kouhei> Revert needless style change f120a7d <Sutou Kouhei> Suppress a warning with MSVC 62f3649 <Sutou Kouhei> Remove unused lambda captures 0090051 <Sutou Kouhei> Add API index for 1.0.0 bd6002a <Sutou Kouhei> Remove entries 335d7b6 <Sutou Kouhei> Implement Arrow::Column in Ruby fb5ad1c <Sutou Kouhei> Add garrow_chunked_array_get_n_rows() for consistency 60dfb0c <Sutou Kouhei> Add garrow_schema_get_field_index() c4ff972 <Sutou Kouhei> Remove backward compatible column API from GArrowRecordBatch 179aa67 <Sutou Kouhei> Use "column_data" instead of "column" 795b6c1 <Sutou Kouhei> Follow arrow::Column remove ac6f372 <Kevin Gurney> Remove use of arrow::Column from feather_reader.cc. Remove use of deprecated StatusCode enum class values in handle_status.cc fcaaf8c <Wes McKinney> Add pyarrow.ChunkedArray.flatten method. Remove Column from glib, but haven't fixed unit tests yet f0d48cc <Wes McKinney> Adapt Python bindings c161a9a <Wes McKinney> Fix up Parquet, too 93e3cad <Wes McKinney> arrow-tests all passing again a7344a8 <Wes McKinney> Stage 1 of cutting away Column Lead-authored-by: Wes McKinney <[email protected]> Co-authored-by: Sutou Kouhei <[email protected]> Co-authored-by: Kevin Gurney <[email protected]> Signed-off-by: Wes McKinney <[email protected]>
apache · Jul 17, 2019 · c350bba · c350bba
1 parent cbaa066
commit c350bba
Show file tree

Hide file tree

Showing 128 changed files with 1,630 additions and 3,061 deletions.
diff --git a/c_glib/arrow-glib/Makefile.am b/c_glib/arrow-glib/Makefile.am
@@ -55,7 +55,6 @@ libarrow_glib_la_headers =			\
 	buffer.h				\
 	chunked-array.h				\
 	codec.h					\
-	column.h				\
 	composite-array.h			\
 	composite-data-type.h			\
 	data-type.h				\
@@ -107,7 +106,6 @@ libarrow_glib_la_sources =			\
 	buffer.cpp				\
 	chunked-array.cpp			\
 	codec.cpp				\
-	column.cpp				\
 	composite-array.cpp			\
 	composite-data-type.cpp			\
 	decimal128.cpp				\
@@ -153,7 +151,6 @@ libarrow_glib_la_cpp_headers =			\
 	buffer.hpp				\
 	chunked-array.hpp			\
 	codec.hpp				\
-	column.hpp				\
 	data-type.hpp				\
 	decimal128.hpp				\
 	error.hpp				\
@@ -187,9 +184,13 @@ libarrow_glib_la_cpp_headers +=			\
 	orc-file-reader.hpp
 endif
 
+libarrow_glib_la_cpp_internal_headers =		\
+	internal-index.hpp
+
 libarrow_glib_la_SOURCES =			\
 	$(libarrow_glib_la_sources)		\
-	$(libarrow_glib_la_cpp_headers)
+	$(libarrow_glib_la_cpp_headers)		\
+	$(libarrow_glib_la_cpp_internal_headers)
 
 BUILT_SOURCES =					\
 	$(libarrow_glib_la_genearted_headers)	\

diff --git a/c_glib/arrow-glib/arrow-glib.h b/c_glib/arrow-glib/arrow-glib.h
@@ -26,7 +26,6 @@
 #include <arrow-glib/array-builder.h>
 #include <arrow-glib/chunked-array.h>
 #include <arrow-glib/codec.h>
-#include <arrow-glib/column.h>
 #include <arrow-glib/compute.h>
 #include <arrow-glib/data-type.h>
 #include <arrow-glib/enums.h>

diff --git a/c_glib/arrow-glib/arrow-glib.hpp b/c_glib/arrow-glib/arrow-glib.hpp
@@ -26,7 +26,6 @@
 #include <arrow-glib/buffer.hpp>
 #include <arrow-glib/chunked-array.hpp>
 #include <arrow-glib/codec.hpp>
-#include <arrow-glib/column.hpp>
 #include <arrow-glib/data-type.hpp>
 #include <arrow-glib/error.hpp>
 #include <arrow-glib/field.hpp>

diff --git a/c_glib/arrow-glib/chunked-array.cpp b/c_glib/arrow-glib/chunked-array.cpp
@@ -206,9 +206,25 @@ garrow_chunked_array_get_value_type(GArrowChunkedArray *chunked_array)
  * @chunked_array: A #GArrowChunkedArray.
  *
  * Returns: The total number of rows in the chunked array.
+ *
+ * Deprecated: 1.0.0: Use garrow_chunked_array_get_n_rows() instead.
  */
 guint64
 garrow_chunked_array_get_length(GArrowChunkedArray *chunked_array)
+{
+  return garrow_chunked_array_get_n_rows(chunked_array);
+}
+
+/**
+ * garrow_chunked_array_get_n_rows:
+ * @chunked_array: A #GArrowChunkedArray.
+ *
+ * Returns: The total number of rows in the chunked array.
+ *
+ * Since: 1.0.0
+ */
+guint64
+garrow_chunked_array_get_n_rows(GArrowChunkedArray *chunked_array)
 {
   const auto arrow_chunked_array = garrow_chunked_array_get_raw(chunked_array);
   return arrow_chunked_array->length();

diff --git a/c_glib/arrow-glib/chunked-array.h b/c_glib/arrow-glib/chunked-array.h
@@ -44,7 +44,10 @@ garrow_chunked_array_get_value_data_type(GArrowChunkedArray *chunked_array);
 GArrowType
 garrow_chunked_array_get_value_type(GArrowChunkedArray *chunked_array);
 
+GARROW_DEPRECATED_IN_1_0_FOR(garrow_chunked_array_get_n_rows)
 guint64 garrow_chunked_array_get_length (GArrowChunkedArray *chunked_array);
+GARROW_AVAILABLE_IN_1_0
+guint64 garrow_chunked_array_get_n_rows (GArrowChunkedArray *chunked_array);
 guint64 garrow_chunked_array_get_n_nulls(GArrowChunkedArray *chunked_array);
 guint   garrow_chunked_array_get_n_chunks (GArrowChunkedArray *chunked_array);