From ca359ba7b478f369f715cc620de610ed2cc13e6c Mon Sep 17 00:00:00 2001
From: Antoine Pitrou <antoine@python.org>
Date: Mon, 19 Feb 2024 16:37:43 +0100
Subject: [PATCH] MINOR: [C++][Parquet][Tools] Print FIXED_LEN_BYTE_ARRAY
 length

In `ParquetFilePrinter`, when printing the type of the column, also print its byte width if the type is FIXED_LEN_BYTE_ARRAY.

Before:
```
Column 0: float16_plain (FIXED_LEN_BYTE_ARRAY / Float16)
Column 1: float16_byte_stream_split (FIXED_LEN_BYTE_ARRAY / Float16)
Column 2: float_plain (FLOAT)
Column 3: float_byte_stream_split (FLOAT)
Column 4: double_plain (DOUBLE)
Column 5: double_byte_stream_split (DOUBLE)
Column 6: int32_plain (INT32)
Column 7: int32_byte_stream_split (INT32)
Column 8: int64_plain (INT64)
Column 9: int64_byte_stream_split (INT64)
Column 10: flba5_plain (FIXED_LEN_BYTE_ARRAY)
Column 11: flba5_byte_stream_split (FIXED_LEN_BYTE_ARRAY)
Column 12: decimal_plain (FIXED_LEN_BYTE_ARRAY / Decimal(precision=7, scale=3) / DECIMAL(7,3))
Column 13: decimal_byte_stream_split (FIXED_LEN_BYTE_ARRAY / Decimal(precision=7, scale=3) / DECIMAL(7,3))
```

After:
```
Column 0: float16_plain (FIXED_LEN_BYTE_ARRAY(2) / Float16)
Column 1: float16_byte_stream_split (FIXED_LEN_BYTE_ARRAY(2) / Float16)
Column 2: float_plain (FLOAT)
Column 3: float_byte_stream_split (FLOAT)
Column 4: double_plain (DOUBLE)
Column 5: double_byte_stream_split (DOUBLE)
Column 6: int32_plain (INT32)
Column 7: int32_byte_stream_split (INT32)
Column 8: int64_plain (INT64)
Column 9: int64_byte_stream_split (INT64)
Column 10: flba5_plain (FIXED_LEN_BYTE_ARRAY(5))
Column 11: flba5_byte_stream_split (FIXED_LEN_BYTE_ARRAY(5))
Column 12: decimal_plain (FIXED_LEN_BYTE_ARRAY(4) / Decimal(precision=7, scale=3) / DECIMAL(7,3))
Column 13: decimal_byte_stream_split (FIXED_LEN_BYTE_ARRAY(4) / Decimal(precision=7, scale=3) / DECIMAL(7,3))
```
---
 cpp/src/parquet/printer.cc |  2 +-
 cpp/src/parquet/types.cc   | 10 ++++++++++
 cpp/src/parquet/types.h    |  2 ++
 3 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/cpp/src/parquet/printer.cc b/cpp/src/parquet/printer.cc
index f11397ab96ed8..ce194f897e44d 100644
--- a/cpp/src/parquet/printer.cc
+++ b/cpp/src/parquet/printer.cc
@@ -105,7 +105,7 @@ void ParquetFilePrinter::DebugPrint(std::ostream& stream, std::list<int> selecte
   for (auto i : selected_columns) {
     const ColumnDescriptor* descr = file_metadata->schema()->Column(i);
     stream << "Column " << i << ": " << descr->path()->ToDotString() << " ("
-           << TypeToString(descr->physical_type());
+           << TypeToString(descr->physical_type(), descr->type_length());
     const auto& logical_type = descr->logical_type();
     if (!logical_type->is_none()) {
       stream << " / " << logical_type->ToString();
diff --git a/cpp/src/parquet/types.cc b/cpp/src/parquet/types.cc
index 33fed01ba324f..7b50ed48d06b0 100644
--- a/cpp/src/parquet/types.cc
+++ b/cpp/src/parquet/types.cc
@@ -177,6 +177,16 @@ std::string TypeToString(Type::type t) {
   }
 }
 
+std::string TypeToString(Type::type t, int type_length) {
+  auto s = TypeToString(t);
+  if (t == Type::FIXED_LEN_BYTE_ARRAY) {
+    s += '(';
+    s += std::to_string(type_length);
+    s += ')';
+  }
+  return s;
+}
+
 std::string ConvertedTypeToString(ConvertedType::type t) {
   switch (t) {
     case ConvertedType::NONE:
diff --git a/cpp/src/parquet/types.h b/cpp/src/parquet/types.h
index 76dd0efc7cb4a..38529bceae85f 100644
--- a/cpp/src/parquet/types.h
+++ b/cpp/src/parquet/types.h
@@ -796,6 +796,8 @@ PARQUET_EXPORT std::string ConvertedTypeToString(ConvertedType::type t);
 
 PARQUET_EXPORT std::string TypeToString(Type::type t);
 
+PARQUET_EXPORT std::string TypeToString(Type::type t, int type_length);
+
 PARQUET_EXPORT std::string FormatStatValue(Type::type parquet_type,
                                            ::std::string_view val);