add generate_uuid() in the functions

Signed-off-by: Xun Zhang <xunzh@amazon.com>

Author: Zhangxunmt

_data-prepper/pipelines/configuration/processors/add-entries.md

@@ -21,7 +21,7 @@ You can configure the `add_entries` processor with the following options.
 | `metadata_key` | No | The key for the new metadata attribute. The argument must be a literal string key and not a JSON Pointer. Either one string key or `metadata_key` is required. |
 | `value` | No | The value of the new entry to be added, which can be used with any of the following data types: strings, Booleans, numbers, null, nested objects, and arrays. |
 | `format` | No | A format string to use as the value of the new entry, for example, `${key1}-${key2}`, where `key1` and `key2` are existing keys in the event. Required if neither `value` nor `value_expression` is specified. |
-| `value_expression` | No | An expression string to use as the value of the new entry. For example, `/key` is an existing key in the event with a type of either a number, a string, or a Boolean. Expressions can also contain functions returning number/string/integer. For example, `length(/key)` will return the length of the key in the event when the key is a string. For more information about keys, see [Expression syntax]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/expression-syntax/).  |
+| `value_expression` | No | An expression string to use as the value of the new entry. For example, `/key` is an existing key in the event with a type of either a number, a string, or a Boolean. Expressions can also contain functions returning number/string/integer. For example, `length(/key)` will return the length of the key in the event when the key is a string. For more information about keys, see [Expression syntax]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/expression-syntax/). For more information about functions, see [Functions]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/functions/). |
 | `add_when` | No | A [conditional expression]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/expression-syntax/), such as `/some-key == "test"'`, that will be evaluated to determine whether the processor will be run on the event. |
 | `overwrite_if_key_exists` | No | When set to `true`, the existing value is overwritten if `key` already exists in the event. The default value is `false`. |
 | `append_if_key_exists` | No | When set to `true`, the existing value will be appended if a `key` already exists in the event. An array will be created if the existing value is not an array. Default is `false`. |

_data-prepper/pipelines/functions.md

@@ -8,10 +8,11 @@ has_children: true
 
 # Functions
 
-OpenSearch Data Prepper offers a range of built-in functions that can be used within expressions to perform common data preprocessing tasks, such as calculating lengths, checking for tags, retrieving metadata, searching for substrings, checking IP address ranges, and joining list elements. These functions include the following:
+OpenSearch Data Prepper offers a range of built-in functions that can be used within expressions to perform common data preprocessing tasks, such as calculating lengths, checking for tags, retrieving metadata, searching for substrings, checking IP address ranges, joining list elements, and generating unique identifiers. These functions include the following:
 
 - [`cidrContains()`]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/cidrcontains/)
 - [`contains()`]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/contains/)
+- [`generate_uuid()`]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/generate-uuid/)
 - [`getMetadata()`]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/get-metadata/)
 - [`getEventType()`]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/get-eventtype/)
 - [`hasTags()`]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/has-tags/)

_data-prepper/pipelines/generate-uuid.md

@@ -0,0 +1,72 @@
+---
+layout: default
+title: generate_uuid()
+parent: Functions
+grand_parent: Pipelines
+nav_order: 11
+---
+
+# generate_uuid()
+
+The `generate_uuid()` function takes no arguments and returns a randomly generated [UUID version 4](https://www.rfc-editor.org/rfc/rfc4122) string, for example, `"550e8400-e29b-41d4-a716-446655440000"`. Each call produces a unique value using a cryptographically strong random number generator, so collision probability is negligible in practice.
+
+This function is useful when source records do not contain a natural unique identifier---for example, when running asynchronous batch inference jobs that require a stable key to match inference results back to the original records.
+
+## Usage
+
+Use `generate_uuid()` as a `value_expression` in the [`add_entries`]({{site.url}}{{site.baseurl}}/data-prepper/pipelines/configuration/processors/add-entries/) processor:
+
+```yaml
+processor:
+  - add_entries:
+      entries:
+        - key: recordId
+          value_expression: 'generate_uuid()'
+```
+{% include copy.html %}
+
+This adds a `recordId` field containing a unique UUID to every event passing through the processor.
+
+## Example
+
+The following pipeline assigns a unique `recordId` to each incoming log record before forwarding it to OpenSearch:
+
+```yaml
+uuid-demo-pipeline:
+  source:
+    http:
+      ssl: false
+
+  processor:
+    - add_entries:
+        entries:
+          - key: recordId
+            value_expression: 'generate_uuid()'
+
+  sink:
+    - opensearch:
+        hosts: ["https://opensearch:9200"]
+        insecure: true
+        username: admin
+        password: admin_password
+        index_type: custom
+        index: demo-index-%{yyyy.MM.dd}
+```
+{% include copy.html %}
+
+Given the following input event:
+
+```json
+{ "message": "user login", "user": "alice" }
+```
+{% include copy.html %}
+
+The document stored in OpenSearch contains the following information:
+
+```json
+{
+  "message": "user login",
+  "user": "alice",
+  "recordId": "550e8400-e29b-41d4-a716-446655440000"
+}
+```