Add docs for yaml dependencies. (apache#34345)

Author: robertwb

CHANGES.md

@@ -111,6 +111,8 @@
 * [Java] Support for `--add-modules` JVM option is added through a new pipeline option `JdkAddRootModules`. This allows extending the module graph with optional modules such as SDK incubator modules. Sample usage: `<pipeline invocation> --jdkAddRootModules=jdk.incubator.vector` ([#30281](https://github.com/apache/beam/issues/30281)).
 * X feature added (Java/Python) ([#X](https://github.com/apache/beam/issues/X)).
 * Managed API for [Java](https://beam.apache.org/releases/javadoc/current/org/apache/beam/sdk/managed/Managed.html) and [Python](https://beam.apache.org/releases/pydoc/current/apache_beam.transforms.managed.html#module-apache_beam.transforms.managed) supports [key I/O connectors](https://beam.apache.org/documentation/io/connectors/) Iceberg, Kafka, and BigQuery.
+* [YAML] Beam YAML UDFs (such as those used in MapToFields) can now have declared dependencies
+  (e.g. pypi packages for Python, or extra jars for Java).
 * Prism now supports event time triggers for most common cases. ([#31438](https://github.com/apache/beam/issues/31438))
   * Prism does not yet support triggered side inputs, or triggers on merging windows (such as session windows).
 

sdks/python/apache_beam/yaml/readme_test.py

@@ -299,6 +299,8 @@ def extract_name(input_spec):
         if code_lines:
           if code_lines[0].startswith('- type:'):
             specs = yaml.load('\n'.join(code_lines), Loader=SafeLoader)
+            if 'dependencies:' in specs:
+              test_type = 'PARSE'
             is_chain = not any('input' in spec for spec in specs)
             if is_chain:
               undefined_inputs = set(['input'])

website/www/site/content/en/documentation/sdks/yaml-udf.md

@@ -508,3 +508,33 @@ a `{type: 'basic_type_name'}` nesting.
 
 This can be especially useful to resolve errors involving the inability to
 handle the `beam:logical:pythonsdk_any:v1` type.
+
+
+## Dependencies
+
+Often user defined functions need to rely on external dependencies.
+These can be provided with a `dependencies` attribute in the transform
+config. For example
+
+```
+- type: MapToFields
+  config:
+    language: python
+    dependencies:
+      - 'scipy>=1.15'
+    fields:
+      new_col:
+        callable: |
+          import scipy.special
+
+          def func(t):
+            return scipy.special.zeta(complex(1/2, t))
+```
+
+The dependencies are interpreted according to the language, e.g.
+for Java one provides a list of maven identifiers and/or jars,
+and for Python one provides a list of pypi package specifiers and/or sdist tarballs.
+See also the full examples using
+[java dependencies](https://github.com/apache/beam/blob/master/sdks/python/apache_beam/yaml/examples/transforms/elementwise/map_to_fields_with_java_deps.yaml)
+and
+[python dependencies](https://github.com/apache/beam/blob/master/sdks/python/apache_beam/yaml/examples/transforms/elementwise/map_to_fields_with_deps.yaml).