commit 991e4a5f362714ca15facbd6b074ea44d756e0d0
author Jacek Pliszka <Jacek.Pliszka@gmail.com> Fri Aug 17 17:52:29 2018 -0400
committer Wes McKinney <wesm+git@apache.org> Fri Aug 17 17:52:29 2018 -0400
tree f4b9850288cdd8f192b955f164882e72962b18c9
parent 79a1bc5d5368bd780a2c0e745d28f562e248a59f

PARQUET-1256: Add --print-key-value-metadata option to parquet_reader tool

This is a minor change useful for debugging.

Now parquet_reader tool has --print-key-value-metadata which when present, dump of key values of file metadata is done.

Created https://issues.apache.org/jira/browse/PARQUET-1256

Author: Jacek Pliszka <Jacek.Pliszka@gmail.com>

Closes #450 from JacekPliszka/master and squashes the following commits:

0d9a108 [Jacek Pliszka] Added --print-key-value-metadata option to parquet_reader tool

src/parquet/printer.cc

diff --git a/src/parquet/printer.cc b/src/parquet/printer.cc
index 3f18a5c..9f26a41 100644
--- a/src/parquet/printer.cc
+++ b/src/parquet/printer.cc
@@ -33,13 +33,25 @@
 #define COL_WIDTH "30"
 
 void ParquetFilePrinter::DebugPrint(std::ostream& stream, std::list<int> selected_columns,
-                                    bool print_values, const char* filename) {
+                                    bool print_values, bool print_key_value_metadata,
+                                    const char* filename) {
   const FileMetaData* file_metadata = fileReader->metadata().get();
 
   stream << "File Name: " << filename << "\n";
   stream << "Version: " << file_metadata->version() << "\n";
   stream << "Created By: " << file_metadata->created_by() << "\n";
   stream << "Total rows: " << file_metadata->num_rows() << "\n";
+
+  if (print_key_value_metadata) {
+    auto key_value_metadata = file_metadata->key_value_metadata();
+    int64_t size_of_key_value_metadata = key_value_metadata->size();
+    stream << "Key Value File Metadata: " << size_of_key_value_metadata << " entries\n";
+    for (int64_t i = 0; i < size_of_key_value_metadata; i++) {
+      stream << " Key nr " << i << " " << key_value_metadata->key(i) << ": "
+             << key_value_metadata->value(i) << "\n";
+    }
+  }
+
   stream << "Number of RowGroups: " << file_metadata->num_row_groups() << "\n";
   stream << "Number of Real Columns: "
          << file_metadata->schema()->group_node()->field_count() << "\n";

src/parquet/printer.h

diff --git a/src/parquet/printer.h b/src/parquet/printer.h
index 3b82882..1113c3f 100644
--- a/src/parquet/printer.h
+++ b/src/parquet/printer.h
@@ -38,7 +38,8 @@
   ~ParquetFilePrinter() {}
 
   void DebugPrint(std::ostream& stream, std::list<int> selected_columns,
-                  bool print_values = true, const char* fileame = "No Name");
+                  bool print_values = true, bool print_key_value_metadata = false,
+                  const char* filename = "No Name");
 
   void JSONPrint(std::ostream& stream, std::list<int> selected_columns,
                  const char* filename = "No Name");

tools/parquet_reader.cc

diff --git a/tools/parquet_reader.cc b/tools/parquet_reader.cc
index 7ef59dc..34bdfc1 100644
--- a/tools/parquet_reader.cc
+++ b/tools/parquet_reader.cc
@@ -24,13 +24,14 @@
 int main(int argc, char** argv) {
   if (argc > 5 || argc < 2) {
     std::cerr << "Usage: parquet_reader [--only-metadata] [--no-memory-map] [--json]"
-                 "[--columns=...] <file>"
+                 "[--print-key-value-metadata] [--columns=...] <file>"
               << std::endl;
     return -1;
   }
 
   std::string filename;
   bool print_values = true;
+  bool print_key_value_metadata = false;
   bool memory_map = true;
   bool format_json = false;
 
@@ -42,6 +43,8 @@
   for (int i = 1; i < argc; i++) {
     if ((param = std::strstr(argv[i], "--only-metadata"))) {
       print_values = false;
+    } else if ((param = std::strstr(argv[i], "--print-key-value-metadata"))) {
+      print_key_value_metadata = true;
     } else if ((param = std::strstr(argv[i], "--no-memory-map"))) {
       memory_map = false;
     } else if ((param = std::strstr(argv[i], "--json"))) {
@@ -64,7 +67,8 @@
     if (format_json) {
       printer.JSONPrint(std::cout, columns, filename.c_str());
     } else {
-      printer.DebugPrint(std::cout, columns, print_values, filename.c_str());
+      printer.DebugPrint(std::cout, columns, print_values,
+        print_key_value_metadata, filename.c_str());
     }
   } catch (const std::exception& e) {
     std::cerr << "Parquet error: " << e.what() << std::endl;