Merge pull request #75 from diksha-sf/main

Update dc_function to function

Author: diksha-sf

README.md

@@ -40,6 +40,8 @@ pip install salesforce-data-customcode
 datacustomcode init my_package
 ```
 
+To create a package of type function, pass the parameter `--code-type=function` with the init command.
+
 This will yield all necessary files to get started:
 ```
 .
@@ -244,6 +246,8 @@ Initialize a new development environment with a code package template.
 
 Argument:
 - `DIRECTORY`: Directory to create project in (default: ".")
+Options:
+- `--code-type TEXT`: This can be either `function` or `script`. The default value is `script` if the argument is missing.
 
 
 #### `datacustomcode scan`
@@ -293,6 +297,7 @@ Options:
 - `--description TEXT`: Description of the transformation job (default: "")
 - `--network TEXT`: docker network (default: "default")
 - `--cpu-size TEXT`: CPU size for the deployment (default: `CPU_2XL`). Available options: CPU_L(Large), CPU_XL(Extra Large), CPU_2XL(2X Large), CPU_4XL(4X Large)
+- `--function-invoke-opt TEXT`: Currently we support only `UnstructuredChunking` for functions.
 
 
 ## Docker usage

src/datacustomcode/templates/function/Dockerfile

@@ -1,4 +1,4 @@
-FROM public.ecr.aws/emr-on-eks/spark/emr-7.3.0:latest
+FROM python:3.11-slim
 
 USER root
 

src/datacustomcode/templates/function/Dockerfile.dependencies

@@ -33,8 +33,8 @@ def chunk_text(text: str, chunk_size: int = 1000) -> List[str]:
     return chunks
 
 
-def dc_function(request: dict) -> dict:
-    logger.info("Inside DC Function")
+def function(request: dict) -> dict:
+    logger.info("Inside Function")
     logger.info(request)
 
     items = request["input"]
@@ -107,7 +107,7 @@ def dc_function(request: dict) -> dict:
     }
 
     # Run the function
-    result = dc_function(test_request)
+    result = function(test_request)
 
     # Print the results in a more readable format
     print("\nChunking Results:")

src/datacustomcode/templates/function/account.ipynb

@@ -1,10 +1,3 @@
 # Required packages for the project - Do not modify
-salesforce-cdp-connector>=1.0.16
-pyspark==3.5.1
-pandas
-numpy
 pydantic
-jupyterlab
-ipywidgets
-tqdm
 salesforce-data-customcode