Support lf.Message to be used to create lf.Template.

- This allows metadata passing seamlessly from `lf.Message` to `lf.Template`, which enables multi-modal object chaining
- Pull up `lf.Template.from_value` to `lf.LangFunc`.
- Eval framework to create prompt bound with input example, allowing multi-modal object passing.

PiperOrigin-RevId: 631431707

langfun/core/eval/base.py

@@ -1179,7 +1179,7 @@ def call_postprocess(self, lm_response: str) -> str:
 
   def process(self, example: Any, **kwargs) -> lf.Message:
     """Process an example and returns its output."""
-    prompt = self.prompt.render(example=example).text
+    prompt = lf.Template.from_value(self.prompt, example=example)
     if self.method == 'call':
       return lf_structured.call(
           prompt,
@@ -1207,7 +1207,9 @@ def process(self, example: Any, **kwargs) -> lf.Message:
     else:
       assert self.method == 'complete', self.method
       assert isinstance(self.schema.spec, pg.typing.Object), self.schema
-      input_value = self.schema.spec.cls.partial(prompt)
+      # TODO(daiyip): Currently multi-modal inputs within the prompt for
+      # completion is not supported.
+      input_value = self.schema.spec.cls.partial(prompt.render().text)
       return lf_structured.complete(
           input_value,
           lm=self.lm,

langfun/core/langfunc.py

@@ -14,7 +14,7 @@
 """LangFunc: Language-based functions."""
 
 import dataclasses
-from typing import Annotated, Type, Union
+from typing import Annotated, Type
 
 from langfun.core import component
 from langfun.core import language_model
@@ -328,22 +328,6 @@ def transform_output(
     """Transforms the output message before returning from __call__."""
     return lm_output
 
-  @classmethod
-  def from_value(
-      cls, value: Union[str, template_lib.Template], **kwargs
-  ) -> 'LangFunc':
-    """Create a LangFunc object from a string or template."""
-    if isinstance(value, LangFunc):
-      return value
-    if isinstance(value, template_lib.Template):
-      lfun = LangFunc(value.template_str, **kwargs)
-      # So lfun could acccess all attributes from value.
-      lfun.sym_setparent(value)
-      return lfun
-    if isinstance(value, str):
-      return LangFunc(template_str=value, **kwargs)
-    return LangFunc('{{input}}', input=value, **kwargs)
-
 
 # Register converter from str to LangFunc, therefore we can always
 # pass strs to attributes that accept LangFunc.

langfun/core/langfunc_test.py

@@ -57,6 +57,10 @@ def test_from_value(self):
     l2 = LangFunc.from_value(l1)
     self.assertIs(l2, l1)
 
+    l3 = LangFunc.from_value(l1, x=1)
+    self.assertIsNot(l3, l1)
+    self.assertTrue(pg.eq(l3, LangFunc('Hello', x=1)))
+
     c = template_lib.Template(
         '{{x}} + {{l}}',
         x=1,

langfun/core/structured/prompting.py

@@ -214,13 +214,8 @@ class Flight(pg.Object):
   # prompt rendering.
   prompt_kwargs.pop('template_str', None)
 
-  if isinstance(prompt, str):
-    prompt = lf.Template(prompt, **prompt_kwargs)
-  elif isinstance(prompt, lf.Template):
-    prompt = prompt.rebind(**prompt_kwargs, raise_on_no_change=False)
-
-  if isinstance(prompt, lf.Template):
-    prompt = prompt.render(lm=lm)
+  if isinstance(prompt, (str, lf.Message, lf.Template)):
+    prompt = lf.Template.from_value(prompt, **prompt_kwargs).render(lm=lm)
   else:
     prompt = schema_lib.mark_missing(prompt)
 

langfun/core/template.py

@@ -17,7 +17,7 @@
 import dataclasses
 import functools
 import inspect
-from typing import Annotated, Any, Callable, Iterator, Set, Tuple, Type
+from typing import Annotated, Any, Callable, Iterator, Set, Tuple, Type, Union
 
 import jinja2
 from jinja2 import meta as jinja2_meta
@@ -495,6 +495,27 @@ class Bar(lf.Template):
     t.sym_setparent(self)
     return t
 
+  @classmethod
+  def from_value(
+      cls,
+      value: Union[str, message_lib.Message, 'Template'],
+      **kwargs
+  ) -> 'Template':
+    """Create a template object from a string or template."""
+    if isinstance(value, cls):
+      return value.clone(override=kwargs) if kwargs else value  # pylint: disable=no-value-for-parameter
+    if isinstance(value, str):
+      return cls(template_str=value, **kwargs)
+    if isinstance(value, message_lib.Message):
+      kwargs.update(value.metadata)
+      return cls(template_str=value.text, **kwargs)
+    if isinstance(value, Template):
+      lfun = cls(template_str=value.template_str, **kwargs)
+      # So lfun could acccess all attributes from value.
+      lfun.sym_setparent(value)
+      return lfun
+    return cls(template_str='{{input}}', input=value, **kwargs)
+
 
 # Register converter from str to LangFunc, therefore we can always
 # pass strs to attributes that accept LangFunc.