From e3233fa91ae29554ab14f432aec927aa5c543433 Mon Sep 17 00:00:00 2001
From: yixiutt <yixiu@selectdb.com>
Date: Thu, 12 May 2022 11:14:54 +0800
Subject: [PATCH] [bugfix](load) fix coredump in ordinal index flush

commit #9123 introduce the bug. bitshuffle page return error when
page is full, so scalar column write cannot switch to next page, which make
ordinal index is null when flush.

All page builder should return ok when page full, and column writer procedure
shoud be append_data, check is_page_full, switch to next page
---
 be/src/olap/rowset/segment_v2/binary_dict_page.cpp | 6 +++---
 be/src/olap/rowset/segment_v2/bitshuffle_page.h    | 2 +-
 be/src/olap/rowset/segment_v2/page_builder.h       | 4 +++-
 3 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/be/src/olap/rowset/segment_v2/binary_dict_page.cpp b/be/src/olap/rowset/segment_v2/binary_dict_page.cpp
index 28859732266068..0b027659a70ba7 100644
--- a/be/src/olap/rowset/segment_v2/binary_dict_page.cpp
+++ b/be/src/olap/rowset/segment_v2/binary_dict_page.cpp
@@ -75,13 +75,13 @@ Status BinaryDictPageBuilder::add(const uint8_t* vals, size_t* count) {
         }
 
         for (int i = 0; i < *count; ++i, ++src) {
+            if (is_page_full()) {
+                break;
+            }
             auto iter = _dictionary.find(*src);
             if (iter != _dictionary.end()) {
                 value_code = iter->second;
             } else {
-                if (_dict_builder->is_page_full()) {
-                    break;
-                }
                 Slice dict_item(src->data, src->size);
                 if (src->size > 0) {
                     char* item_mem = (char*)_pool.allocate(src->size);
diff --git a/be/src/olap/rowset/segment_v2/bitshuffle_page.h b/be/src/olap/rowset/segment_v2/bitshuffle_page.h
index 50edb30c2a32a3..c7fca81be1975e 100644
--- a/be/src/olap/rowset/segment_v2/bitshuffle_page.h
+++ b/be/src/olap/rowset/segment_v2/bitshuffle_page.h
@@ -105,7 +105,7 @@ class BitshufflePageBuilder : public PageBuilder {
         DCHECK(!_finished);
         if (_remain_element_capacity <= 0) {
             *count = 0;
-            return Status::RuntimeError("page is full.");
+            return Status::OK();
         }
         int to_add = std::min<int>(_remain_element_capacity, *count);
         int to_add_size = to_add * SIZE_OF_TYPE;
diff --git a/be/src/olap/rowset/segment_v2/page_builder.h b/be/src/olap/rowset/segment_v2/page_builder.h
index 26eec4b43045ee..ab74ad7fcaed9d 100644
--- a/be/src/olap/rowset/segment_v2/page_builder.h
+++ b/be/src/olap/rowset/segment_v2/page_builder.h
@@ -49,7 +49,9 @@ class PageBuilder {
     // Add a sequence of values to the page.
     // The number of values actually added will be returned through count, which may be less
     // than requested if the page is full.
-    //
+
+    // check page if full before truly add, return ok when page is full so that column write
+    // will switch to next page
     // vals size should be decided according to the page build type
     // TODO make sure vals is naturally-aligned to its type so that impls can use aligned load
     // instead of memcpy to copy values.