This is an automated email from the ASF dual-hosted git repository.
morningman pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/doris-website.git
The following commit(s) were added to refs/heads/master by this push:
new cb12a36243c1 [docs] Clarify Sequence mapping and full-row INSERT
requirements (#4624)
cb12a36243c1 is described below
commit cb12a36243c1fba429f1998f8bf5cbbe8e73a83a
Author: Mingyu Chen (Rayner) <[email protected]>
AuthorDate: Fri Nov 13 08:21:08 2026 +0800
[docs] Clarify Sequence mapping and full-row INSERT requirements (#4624)
Correct the 3.x sequence_type/sequence_col mix-up and distinguish
ordinary-column values from load-time hidden-column mappings. Clarify
full-row INSERT requirements, omitted target-column lists,
CURRENT_TIMESTAMP defaults and partial-update exceptions across active
versions.
Fixes #4066
Versions and languages: dev, 4.x and 3.x; English and Chinese.
Source verification:
- branch-3.0/3.1/4.0 and master BindSink distinguish mapped ordinary
columns from the unmapped hidden Sequence column; omitted INSERT column
lists include the mapped column
- branch-3.1 DataDescription.analyzeSequenceCol returns immediately when
a table already has a Sequence map column
- Current/4.x mapping sections already distinguish sequence_type; their
broad caution had the same ambiguity
Validation:
- Checked both corrected 3.x mapping paragraphs against load source
- Reviewed sequence checks and partial-update exceptions in BindSink
across active branches
- Confirmed six locale/version copies express equivalent requirements
- git diff --check
Review:
- Goal: correct the reported documentation issue; the changed content
directly addresses it.
- Scope: targeted documentation edits.
- Information architecture: active version and locale counterparts
inspected; archived snapshots unchanged.
- Navigation: no paths, slugs, anchors or sidebar entries changed.
- Frontend/build: no framework, styling or build-script changes.
- Validation limits: no live Doris execution or full website build
performed.
- Other concerns: No additional correctness, usability, accessibility,
SEO or maintainability issues found in the changed content.
Related PR: #4617 addresses the independent hidden-Sequence setup issue
in these same six pages. Both PRs correct the 3.x introductory sentence
from sequence_col to sequence_type. Merge either PR first and rebase the
other, preserving #4617’s standalone hidden-table examples and this PR’s
clarified usage notes.
Co-authored-by: morningman <[email protected]>
---
docs/data-operate/update/unique-update-concurrent-control.md | 2 +-
.../current/data-operate/update/unique-update-concurrent-control.md | 2 +-
.../data-operate/update/unique-update-concurrent-control.md | 4 ++--
.../data-operate/update/unique-update-concurrent-control.md | 2 +-
.../data-operate/update/unique-update-concurrent-control.md | 4 ++--
.../data-operate/update/unique-update-concurrent-control.md | 2 +-
6 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/docs/data-operate/update/unique-update-concurrent-control.md
b/docs/data-operate/update/unique-update-concurrent-control.md
index b7c6ed895b61..66d405072f69 100644
--- a/docs/data-operate/update/unique-update-concurrent-control.md
+++ b/docs/data-operate/update/unique-update-concurrent-control.md
@@ -302,7 +302,7 @@ Because the Sequence value `2020-03-23` of the newly
imported data is greater th
<!-- Knowledge type: Caveats -->
-1. **The Sequence column must be specified explicitly**: in import jobs such
as Stream Load and Broker Load, as well as in row-update `INSERT` statements,
you must explicitly specify the Sequence column (unless its default value is
`CURRENT_TIMESTAMP`). Otherwise, you will receive the following error:
+1. **Sequence requirements depend on the table definition**: with
`function_column.sequence_col`, load the mapped ordinary column using the
normal column rules; no extra load-time Sequence mapping parameter is needed.
With `function_column.sequence_type`, configure the source-column mapping for
Stream Load or Broker Load as shown above. For a full-row `INSERT` with an
explicit target-column list, include the mapped ordinary column (unless its
default is `CURRENT_TIMESTAMP`), or the hidde [...]
```Plain
Table test_tbl has sequence column, need to specify the sequence column
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/data-operate/update/unique-update-concurrent-control.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/data-operate/update/unique-update-concurrent-control.md
index 9857923e8573..d1acca1c0f35 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/current/data-operate/update/unique-update-concurrent-control.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/current/data-operate/update/unique-update-concurrent-control.md
@@ -302,7 +302,7 @@ MySQL [test]> select * from test_table;
<!-- 知识类型: 注意事项 -->
-1. **必须显式指定 Sequence 列**：在 Stream Load / Broker Load 等导入任务以及行更新 `INSERT`
语句中，用户必须显式指定 Sequence 列（除非 Sequence 列的默认值为 `CURRENT_TIMESTAMP`），否则会收到如下报错：
+1. **Sequence 列的要求取决于建表方式**：使用 `function_column.sequence_col`
时，按普通列规则导入映射列即可，无需额外指定导入时的 Sequence 映射参数。使用 `function_column.sequence_type`
时，Stream Load / Broker Load 需要按上文配置源数据列映射。对于显式指定目标列列表的全行
`INSERT`，需要包含映射的普通列（除非其默认值为 `CURRENT_TIMESTAMP`）；未映射普通列时，需要包含隐藏列
`__DORIS_SEQUENCE_COL__`。不指定目标列列表的 `INSERT` 已包含映射的普通列。部分列更新的例外见下文第 3 点。缺少必要的
Sequence 输入时，会收到如下报错：
```Plain
Table test_tbl has sequence column, need to specify the sequence column
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.x/data-operate/update/unique-update-concurrent-control.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.x/data-operate/update/unique-update-concurrent-control.md
index 2b9d757715c3..095b064a205f 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.x/data-operate/update/unique-update-concurrent-control.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-3.x/data-operate/update/unique-update-concurrent-control.md
@@ -254,12 +254,12 @@ MySQL [test]> select * from test_table;
### 注意
-1. 为防止误用，在 StreamLoad/BrokerLoad 等导入任务以及行更新 insert 语句中，用户必须显示指定 sequence 列 (除非
sequence 列的默认值为 CURRENT_TIMESTAMP)，不然会收到以下报错信息：
+1. **Sequence 列的要求取决于建表方式**：使用 `function_column.sequence_col`
时，按普通列规则导入映射列即可，无需额外指定导入时的 Sequence 映射参数。使用 `function_column.sequence_type`
时，Stream Load / Broker Load 需要按上文配置源数据列映射。对于显式指定目标列列表的全行
`INSERT`，需要包含映射的普通列（除非其默认值为 `CURRENT_TIMESTAMP`）；未映射普通列时，需要包含隐藏列
`__DORIS_SEQUENCE_COL__`。不指定目标列列表的 `INSERT` 已包含映射的普通列。部分列更新的例外见下文第 3 点。缺少必要的
Sequence 输入时，会收到如下报错：
```Plain
Table test_tbl has sequence column, need to specify the sequence column
```
-2. 在使用 Insert 语句插入数据时，由于用户必须显示指定 sequence
列，否则会报如上异常。为了方便用户在一些场景下（表复制，内部数据迁移等场景）使用，Doris 可以通过 session 参数控制，来关闭 sequence
列的检查约束：
+2. 在表复制、内部数据迁移等场景下，可以通过 session 参数关闭全行 `INSERT` 的 Sequence 列检查约束：
```sql
set require_sequence_in_insert = false;
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/data-operate/update/unique-update-concurrent-control.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/data-operate/update/unique-update-concurrent-control.md
index 05a9eb3008eb..1ff439d76daf 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/data-operate/update/unique-update-concurrent-control.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/data-operate/update/unique-update-concurrent-control.md
@@ -302,7 +302,7 @@ MySQL [test]> select * from test_table;
<!-- 知识类型: 注意事项 -->
-1. **必须显式指定 Sequence 列**：在 Stream Load / Broker Load 等导入任务以及行更新 `INSERT`
语句中，用户必须显式指定 Sequence 列（除非 Sequence 列的默认值为 `CURRENT_TIMESTAMP`），否则会收到如下报错：
+1. **Sequence 列的要求取决于建表方式**：使用 `function_column.sequence_col`
时，按普通列规则导入映射列即可，无需额外指定导入时的 Sequence 映射参数。使用 `function_column.sequence_type`
时，Stream Load / Broker Load 需要按上文配置源数据列映射。对于显式指定目标列列表的全行
`INSERT`，需要包含映射的普通列（除非其默认值为 `CURRENT_TIMESTAMP`）；未映射普通列时，需要包含隐藏列
`__DORIS_SEQUENCE_COL__`。不指定目标列列表的 `INSERT` 已包含映射的普通列。部分列更新的例外见下文第 3 点。缺少必要的
Sequence 输入时，会收到如下报错：
```Plain
Table test_tbl has sequence column, need to specify the sequence column
diff --git
a/versioned_docs/version-3.x/data-operate/update/unique-update-concurrent-control.md
b/versioned_docs/version-3.x/data-operate/update/unique-update-concurrent-control.md
index 95d15459f747..74cc5c266c3b 100644
---
a/versioned_docs/version-3.x/data-operate/update/unique-update-concurrent-control.md
+++
b/versioned_docs/version-3.x/data-operate/update/unique-update-concurrent-control.md
@@ -254,13 +254,13 @@ This time, the data in the table is replaced. In summary,
during the load proces
### Note
-1. To prevent misuse, in Stream Load/Broker Load load tasks and row update
insert statements, users must explicitly specify the sequence column (unless
the default value of the sequence column is CURRENT_TIMESTAMP), otherwise, the
following error message will be received:
+1. **Sequence requirements depend on the table definition**: with
`function_column.sequence_col`, load the mapped ordinary column using the
normal column rules; no extra load-time Sequence mapping parameter is needed.
With `function_column.sequence_type`, configure the source-column mapping for
Stream Load or Broker Load as shown above. For a full-row `INSERT` with an
explicit target-column list, include the mapped ordinary column (unless its
default is `CURRENT_TIMESTAMP`), or the hidde [...]
```Plain
Table test_tbl has sequence column, need to specify the sequence column
```
-2. When you use the Insert statement to insert data, you must display the
specified sequence column; otherwise, the preceding exception is reported. In
order to facilitate the use of Doris in some scenarios (table replication,
internal data migration, etc.), Doris can be controlled by the session
parameter to close the check constraint on the sequence column:
+2. In scenarios such as table replication or internal data migration, you can
disable the Sequence check for full-row `INSERT` statements through a session
variable:
```sql
set require_sequence_in_insert = false;
diff --git
a/versioned_docs/version-4.x/data-operate/update/unique-update-concurrent-control.md
b/versioned_docs/version-4.x/data-operate/update/unique-update-concurrent-control.md
index 09ca762426ab..f3450204025a 100644
---
a/versioned_docs/version-4.x/data-operate/update/unique-update-concurrent-control.md
+++
b/versioned_docs/version-4.x/data-operate/update/unique-update-concurrent-control.md
@@ -302,7 +302,7 @@ Because the Sequence value `2020-03-23` of the newly
imported data is greater th
<!-- Knowledge type: Caveats -->
-1. **The Sequence column must be specified explicitly**: in import jobs such
as Stream Load and Broker Load, as well as in row-update `INSERT` statements,
you must explicitly specify the Sequence column (unless its default value is
`CURRENT_TIMESTAMP`). Otherwise, you will receive the following error:
+1. **Sequence requirements depend on the table definition**: with
`function_column.sequence_col`, load the mapped ordinary column using the
normal column rules; no extra load-time Sequence mapping parameter is needed.
With `function_column.sequence_type`, configure the source-column mapping for
Stream Load or Broker Load as shown above. For a full-row `INSERT` with an
explicit target-column list, include the mapped ordinary column (unless its
default is `CURRENT_TIMESTAMP`), or the hidde [...]
```Plain
Table test_tbl has sequence column, need to specify the sequence column
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
