From b9c22c1d3436a0b7d411756496da5b02555654af Mon Sep 17 00:00:00 2001
From: Gang Wu <ustcwg@gmail.com>
Date: Wed, 13 Nov 2024 10:57:10 +0800
Subject: [PATCH] add per column config

---
 parquet-hadoop/README.md                         |  9 +++++++++
 .../parquet/hadoop/ParquetOutputFormat.java      | 16 ++++++++++++++++
 2 files changed, 25 insertions(+)

diff --git a/parquet-hadoop/README.md b/parquet-hadoop/README.md
index c00fea2439..5817069f4b 100644
--- a/parquet-hadoop/README.md
+++ b/parquet-hadoop/README.md
@@ -515,4 +515,13 @@ if not found then the library will use the classic non-vectored reads: it is saf
 **Description:** Whether to enable column statistics collection.
 If `true`, statistics will be collected for all columns unless explicitly disabled for specific columns.
 If `false`, statistics will be disabled for all columns regardless of column-specific settings.
+It is possible to enable or disable statistics for specific columns by appending `#` followed by the column path.
 **Default value:** `true`
+**Example:**
+```java
+// Enable statistics for all columns
+conf.set("parquet.column.statistics.enabled", true);
+// Disable statistics for 'column.path'
+conf.set("parquet.column.statistics.enabled#column.path", false);
+// The final configuration will be: Enable statistics for all columns except 'column.path'
+```
diff --git a/parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetOutputFormat.java b/parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetOutputFormat.java
index b16d743a98..c13781a685 100644
--- a/parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetOutputFormat.java
+++ b/parquet-hadoop/src/main/java/org/apache/parquet/hadoop/ParquetOutputFormat.java
@@ -397,6 +397,18 @@ public static boolean getStatisticsEnabled(Configuration conf) {
     return conf.getBoolean(STATISTICS_ENABLED, ParquetProperties.DEFAULT_STATISTICS_ENABLED);
   }
 
+  public static void setStatisticsEnabled(JobContext jobContext, String columnPath, boolean enabled) {
+    getConfiguration(jobContext).set(STATISTICS_ENABLED + "#" + columnPath, String.valueOf(enabled));
+  }
+
+  public static boolean getStatisticsEnabled(Configuration conf, String columnPath) {
+    String columnSpecific = conf.get(STATISTICS_ENABLED + "#" + columnPath);
+    if (columnSpecific != null) {
+      return Boolean.parseBoolean(columnSpecific);
+    }
+    return conf.getBoolean(STATISTICS_ENABLED, ParquetProperties.DEFAULT_STATISTICS_ENABLED);
+  }
+
   private WriteSupport<T> writeSupport;
   private ParquetOutputCommitter committer;
 
@@ -489,6 +501,10 @@ public RecordWriter<Void, T> getRecordWriter(Configuration conf, Path file, Comp
             BLOOM_FILTER_CANDIDATES_NUMBER,
             key -> conf.getInt(key, ParquetProperties.DEFAULT_BLOOM_FILTER_CANDIDATES_NUMBER),
             propsBuilder::withBloomFilterCandidatesNumber)
+        .withColumnConfig(
+            STATISTICS_ENABLED,
+            key -> conf.getBoolean(key, ParquetProperties.DEFAULT_STATISTICS_ENABLED),
+            propsBuilder::withStatisticsEnabled)
         .parseConfig(conf);
 
     ParquetProperties props = propsBuilder.build();