docs: update tutorial (#1080)

Signed-off-by: Louis Mandel <[email protected]>

Author: mandel

docs/README.md

@@ -201,7 +201,7 @@ text:
 ```
 
 
-A PDL program computes 2 data structures. The first is a JSON corresponding to the result of the overall program, obtained by aggregating the results of each block. This is what is printed by default when we run the interpreter. The second is a conversational background context, which is a list of role/content pairs, where we implicitly keep track of roles and content for the purpose of communicating with models that support chat APIs. The contents in the latter correspond to the results of each block. The conversational background context is the list of messages used to make calls to LLMs via LiteLLM.
+A PDL program computes two data structures. The first is a JSON corresponding to the result of the overall program, obtained by aggregating the results of each block. This is what is printed by default when we run the interpreter. The second is a conversational background context, which is a list of role/content pairs, where we implicitly keep track of roles and content for the purpose of communicating with models that support chat APIs. The contents in the latter correspond to the results of each block. The conversational background context is the list of messages used to make calls to LLMs via LiteLLM.
 
 The PDL interpreter can also stream the background conversation instead of the result:
 

docs/tutorial.md

@@ -1,6 +1,6 @@
-description: Hello world calling a model
+description: Calling a model on the implicit background context
 text:
 - "Hello\n"
 - model: ollama_chat/granite3.2:2b
   parameters:
-    stop: ['!']
+    stop: ['!']

examples/sdk/hello_dict.py

@@ -1,4 +1,4 @@
-description: Hello world calling a model
+description: Calling a model with an input text
 text:
 - "Hello\n"
 - model: ollama_chat/granite3.2:2b

examples/sdk/hello_prog.py

@@ -1,4 +1,4 @@
-description: Hello world calling a model
+description: Calling a model with an explicit list of messages
 text:
 - "Hello\n"
 - model: ollama_chat/granite3.2:2b

examples/tutorial/calling_llm.pdl

@@ -3,8 +3,10 @@ defs:
   prompt:
     array:
     - role: system
-      content: You are a helpful software engineer. You write clear, concise, well-commented code.
+      content: You are a helpful assistant that is fluent in French.
     - role: user
-      content: Write a Python function that implement merge sort.
-model: ollama_chat/granite3.2:2b
-input: ${ prompt }
+      content: Translate the word 'Hello' to French
+text:
+- "Hello\n"
+- model: ollama_chat/granite3.2:2b
+  input: ${ prompt }

examples/tutorial/calling_llm_with_input.pdl

@@ -1,22 +1,16 @@
-description: Function def and call
-defs:
-  translate:
-    function:
-      sentence: string
-      language: string
-    return:
+text:
+- "Hello\n"
+- defs:
+    fr:
       lastOf:
-      - "\nTranslate the sentence '${ sentence }' to ${ language }.\n"
+      - "\nTranslate to French\n"
       - model: ollama_chat/granite3.2:2b
-        parameters:
-          stop: ["\n"]
-text:
-- call: ${ translate }
-  args:
-    sentence: I love Paris!
-    language: French
-- "\n"
-- call: ${ translate }
-  args:
-    sentence: I love Madrid!
-    language: Spanish
+    es:
+      lastOf:
+      - "\nTranslate to Spanish\n"
+      - model: ollama_chat/granite3.2:2b
+  text: |
+  
+    In Fench: ${ fr }
+
+    In Spanish: ${ es }