add help to cli options

alan-turing-institute · Jun 11, 2024 · 096aee5 · 096aee5
1 parent af4ef86
commit 096aee5
Showing 1 changed file with 142 additions and 69 deletions.
diff --git a/reginald/cli.py b/reginald/cli.py
@@ -8,14 +8,27 @@
 from reginald.run import EMOJI_DEFAULT, main
 
 API_URL_PROPMPT = "No API URL was provided and REGINALD_API_URL not set. Please provide an API URL for the Reginald app"
+HELP_TEXT = {
+    "model": "Select which type of model to use..",
+    "model_name": "Select which sub-model to use (within the main model selected).",
+    "data_dir": "Location of the data (ignored if not using llama-index). Default is the data directory in the root of the repo.",
+    "which_index": "Which index to use (ignored if not using llama-index).",
+    "mode": "Select which mode to use (ignored if not using llama-index).",
+    "force_new_index": "Whether to force the creation of a new index (ignored if not using llama-index).",
+    "max_input_size": "Maximum input size for the model (ignored if not using llama-index).",
+    "k": "'similarity_top_k' to use in chat or query engine (ignored if not using llama-index).",
+    "chunk_size": "Chunk size for the model (ignored if not using llama-index).",
+    "chunk_overlap_ratio": "Chunk overlap ratio for the model (ignored if not using llama-index).",
+    "num_output": "Number of outputs to generate (ignored if not using llama-index).",
+    "is_path": "Whether the data is a path (ignored if not using llama-index-llama-cpp).",
+    "n_gpu_layers": "Number of GPU layers to use (ignored if not using llama-index).",
+    "device": "Device to use (ignored if not using llama-index).",
+    "api_url": "API URL for the Reginald app.",
+    "emoji": "Emoji to use for the bot.",
+}
 
 cli = typer.Typer()
 
-# TODO: before, using get_env_var would print out whenever it's being called and an env var is attempted to be loaded
-# (only when necessary, i.e. not provided), is it possible to tell the user that an env var is going to be used?
-# TODO: create config class
-# TODO: add help
-
 
 def set_up_logging_config(level: int = 20) -> None:
     logging.basicConfig(
@@ -31,46 +44,63 @@ def run_all(
         str,
         typer.Option(
             envvar="REGINALD_MODEL",
-            help="Select which type of model to use. Default is 'hello'.",
+            help=HELP_TEXT["model"],
         ),
     ] = DEFAULT_ARGS["model"],
     model_name: Annotated[
-        Optional[str], typer.Option(envvar="REGINALD_MODEL_NAME")
+        Optional[str],
+        typer.Option(envvar="REGINALD_MODEL_NAME", help=HELP_TEXT["model_name"]),
     ] = None,
-    mode: Annotated[str, typer.Option(envvar="LLAMA_INDEX_MODE")] = DEFAULT_ARGS[
-        "mode"
-    ],
+    mode: Annotated[
+        str, typer.Option(envvar="LLAMA_INDEX_MODE", help=HELP_TEXT["mode"])
+    ] = DEFAULT_ARGS["mode"],
     data_dir: Annotated[
-        pathlib.Path, typer.Option(envvar="LLAMA_INDEX_DATA_DIR")
+        pathlib.Path,
+        typer.Option(envvar="LLAMA_INDEX_DATA_DIR", help=HELP_TEXT["data_dir"]),
     ] = DEFAULT_ARGS["data_dir"],
     which_index: Annotated[
-        str, typer.Option(envvar="LLAMA_INDEX_WHICH_INDEX")
+        str,
+        typer.Option(envvar="LLAMA_INDEX_WHICH_INDEX", help=HELP_TEXT["which_index"]),
     ] = DEFAULT_ARGS["which_index"],
     force_new_index: Annotated[
-        bool, typer.Option(envvar="LLAMA_INDEX_FORCE_NEW_INDEX")
+        bool,
+        typer.Option(
+            envvar="LLAMA_INDEX_FORCE_NEW_INDEX", help=HELP_TEXT["force_new_index"]
+        ),
     ] = DEFAULT_ARGS["force_new_index"],
     max_input_size: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_MAX_INPUT_SIZE")
+        int,
+        typer.Option(
+            envvar="LLAMA_INDEX_MAX_INPUT_SIZE", help=HELP_TEXT["max_input_size"]
+        ),
     ] = DEFAULT_ARGS["max_input_size"],
-    k: Annotated[int, typer.Option(envvar="LLAMA_INDEX_K")] = DEFAULT_ARGS["k"],
+    k: Annotated[
+        int, typer.Option(envvar="LLAMA_INDEX_K", help=HELP_TEXT["k"])
+    ] = DEFAULT_ARGS["k"],
     chunk_size: Annotated[
-        Optional[int], typer.Option(envvar="LLAMA_INDEX_CHUNK_SIZE")
+        Optional[int],
+        typer.Option(envvar="LLAMA_INDEX_CHUNK_SIZE", help=HELP_TEXT["chunk_size"]),
     ] = DEFAULT_ARGS.get("chunk_size"),
     chunk_overlap_ratio: Annotated[
-        float, typer.Option(envvar="LLAMA_INDEX_CHUNK_OVERLAP_RATIO")
+        float,
+        typer.Option(
+            envvar="LLAMA_INDEX_CHUNK_OVERLAP_RATIO",
+            help=HELP_TEXT["chunk_overlap_ratio"],
+        ),
     ] = DEFAULT_ARGS["chunk_overlap_ratio"],
     num_output: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_NUM_OUTPUT")
+        int, typer.Option(envvar="LLAMA_INDEX_NUM_OUTPUT", help=HELP_TEXT["num_output"])
     ] = DEFAULT_ARGS["num_output"],
-    is_path: Annotated[bool, typer.Option(envvar="LLAMA_INDEX_IS_PATH")] = DEFAULT_ARGS[
-        "is_path"
-    ],
+    is_path: Annotated[
+        bool, typer.Option(envvar="LLAMA_INDEX_IS_PATH", help=HELP_TEXT["is_path"])
+    ] = DEFAULT_ARGS["is_path"],
     n_gpu_layers: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_N_GPU_LAYERS")
+        int,
+        typer.Option(envvar="LLAMA_INDEX_N_GPU_LAYERS", help=HELP_TEXT["n_gpu_layers"]),
     ] = DEFAULT_ARGS["n_gpu_layers"],
-    device: Annotated[str, typer.Option(envvar="LLAMA_INDEX_DEVICE")] = DEFAULT_ARGS[
-        "device"
-    ],
+    device: Annotated[
+        str, typer.Option(envvar="LLAMA_INDEX_DEVICE", help=HELP_TEXT["device"])
+    ] = DEFAULT_ARGS["device"],
 ) -> None:
     set_up_logging_config(level=20)
     main(
@@ -95,9 +125,14 @@ def run_all(
 @cli.command()
 def bot(
     api_url: Annotated[
-        str, typer.Option(prompt=API_URL_PROPMPT, envvar="REGINALD_API_URL")
+        str,
+        typer.Option(
+            prompt=API_URL_PROPMPT, envvar="REGINALD_API_URL", help=HELP_TEXT["api_url"]
+        ),
     ],
-    emoji: Annotated[str, typer.Option(envvar="REGINALD_EMOJI")] = EMOJI_DEFAULT,
+    emoji: Annotated[
+        str, typer.Option(envvar="REGINALD_EMOJI", help=HELP_TEXT["emoji"])
+    ] = EMOJI_DEFAULT,
 ) -> None:
     """
     Main function to run the Slack bot which sets up the bot
@@ -115,46 +150,67 @@ def bot(
 
 @cli.command()
 def app(
-    model: Annotated[str, typer.Option(envvar="REGINALD_MODEL")] = DEFAULT_ARGS[
-        "model"
-    ],
+    model: Annotated[
+        str,
+        typer.Option(
+            envvar="REGINALD_MODEL",
+            help=HELP_TEXT["model"],
+        ),
+    ] = DEFAULT_ARGS["model"],
     model_name: Annotated[
-        Optional[str], typer.Option(envvar="REGINALD_MODEL_NAME")
+        Optional[str],
+        typer.Option(envvar="REGINALD_MODEL_NAME", help=HELP_TEXT["model_name"]),
     ] = None,
-    mode: Annotated[str, typer.Option(envvar="LLAMA_INDEX_MODE")] = DEFAULT_ARGS[
-        "mode"
-    ],
+    mode: Annotated[
+        str, typer.Option(envvar="LLAMA_INDEX_MODE", help=HELP_TEXT["mode"])
+    ] = DEFAULT_ARGS["mode"],
     data_dir: Annotated[
-        pathlib.Path, typer.Option(envvar="LLAMA_INDEX_DATA_DIR")
+        pathlib.Path,
+        typer.Option(envvar="LLAMA_INDEX_DATA_DIR", help=HELP_TEXT["data_dir"]),
     ] = DEFAULT_ARGS["data_dir"],
     which_index: Annotated[
-        str, typer.Option(envvar="LLAMA_INDEX_WHICH_INDEX")
+        str,
+        typer.Option(envvar="LLAMA_INDEX_WHICH_INDEX", help=HELP_TEXT["which_index"]),
     ] = DEFAULT_ARGS["which_index"],
     force_new_index: Annotated[
-        bool, typer.Option(envvar="LLAMA_INDEX_FORCE_NEW_INDEX")
+        bool,
+        typer.Option(
+            envvar="LLAMA_INDEX_FORCE_NEW_INDEX", help=HELP_TEXT["force_new_index"]
+        ),
     ] = DEFAULT_ARGS["force_new_index"],
     max_input_size: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_MAX_INPUT_SIZE")
+        int,
+        typer.Option(
+            envvar="LLAMA_INDEX_MAX_INPUT_SIZE", help=HELP_TEXT["max_input_size"]
+        ),
     ] = DEFAULT_ARGS["max_input_size"],
-    k: Annotated[int, typer.Option(envvar="LLAMA_INDEX_K")] = DEFAULT_ARGS["k"],
+    k: Annotated[
+        int, typer.Option(envvar="LLAMA_INDEX_K", help=HELP_TEXT["k"])
+    ] = DEFAULT_ARGS["k"],
     chunk_size: Annotated[
-        Optional[int], typer.Option(envvar="LLAMA_INDEX_CHUNK_SIZE")
+        Optional[int],
+        typer.Option(envvar="LLAMA_INDEX_CHUNK_SIZE", help=HELP_TEXT["chunk_size"]),
     ] = DEFAULT_ARGS.get("chunk_size"),
     chunk_overlap_ratio: Annotated[
-        float, typer.Option(envvar="LLAMA_INDEX_CHUNK_OVERLAP_RATIO")
+        float,
+        typer.Option(
+            envvar="LLAMA_INDEX_CHUNK_OVERLAP_RATIO",
+            help=HELP_TEXT["chunk_overlap_ratio"],
+        ),
     ] = DEFAULT_ARGS["chunk_overlap_ratio"],
     num_output: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_NUM_OUTPUT")
+        int, typer.Option(envvar="LLAMA_INDEX_NUM_OUTPUT", help=HELP_TEXT["num_output"])
     ] = DEFAULT_ARGS["num_output"],
-    is_path: Annotated[bool, typer.Option(envvar="LLAMA_INDEX_IS_PATH")] = DEFAULT_ARGS[
-        "is_path"
-    ],
+    is_path: Annotated[
+        bool, typer.Option(envvar="LLAMA_INDEX_IS_PATH", help=HELP_TEXT["is_path"])
+    ] = DEFAULT_ARGS["is_path"],
     n_gpu_layers: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_N_GPU_LAYERS")
+        int,
+        typer.Option(envvar="LLAMA_INDEX_N_GPU_LAYERS", help=HELP_TEXT["n_gpu_layers"]),
     ] = DEFAULT_ARGS["n_gpu_layers"],
-    device: Annotated[str, typer.Option(envvar="LLAMA_INDEX_DEVICE")] = DEFAULT_ARGS[
-        "device"
-    ],
+    device: Annotated[
+        str, typer.Option(envvar="LLAMA_INDEX_DEVICE", help=HELP_TEXT["device"])
+    ] = DEFAULT_ARGS["device"],
 ) -> None:
     """
     Main function to run the app which sets up the response model
@@ -225,46 +281,63 @@ def chat(
         str,
         typer.Option(
             envvar="REGINALD_MODEL",
-            help="Select which type of model to use. Default is 'hello'.",
+            help=HELP_TEXT["model"],
         ),
     ] = DEFAULT_ARGS["model"],
     model_name: Annotated[
-        Optional[str], typer.Option(envvar="REGINALD_MODEL_NAME")
+        Optional[str],
+        typer.Option(envvar="REGINALD_MODEL_NAME", help=HELP_TEXT["model_name"]),
     ] = None,
-    mode: Annotated[str, typer.Option(envvar="LLAMA_INDEX_MODE")] = DEFAULT_ARGS[
-        "mode"
-    ],
+    mode: Annotated[
+        str, typer.Option(envvar="LLAMA_INDEX_MODE", help=HELP_TEXT["mode"])
+    ] = DEFAULT_ARGS["mode"],
     data_dir: Annotated[
-        pathlib.Path, typer.Option(envvar="LLAMA_INDEX_DATA_DIR")
+        pathlib.Path,
+        typer.Option(envvar="LLAMA_INDEX_DATA_DIR", help=HELP_TEXT["data_dir"]),
     ] = DEFAULT_ARGS["data_dir"],
     which_index: Annotated[
-        str, typer.Option(envvar="LLAMA_INDEX_WHICH_INDEX")
+        str,
+        typer.Option(envvar="LLAMA_INDEX_WHICH_INDEX", help=HELP_TEXT["which_index"]),
     ] = DEFAULT_ARGS["which_index"],
     force_new_index: Annotated[
-        bool, typer.Option(envvar="LLAMA_INDEX_FORCE_NEW_INDEX")
+        bool,
+        typer.Option(
+            envvar="LLAMA_INDEX_FORCE_NEW_INDEX", help=HELP_TEXT["force_new_index"]
+        ),
     ] = DEFAULT_ARGS["force_new_index"],
     max_input_size: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_MAX_INPUT_SIZE")
+        int,
+        typer.Option(
+            envvar="LLAMA_INDEX_MAX_INPUT_SIZE", help=HELP_TEXT["max_input_size"]
+        ),
     ] = DEFAULT_ARGS["max_input_size"],
-    k: Annotated[int, typer.Option(envvar="LLAMA_INDEX_K")] = DEFAULT_ARGS["k"],
+    k: Annotated[
+        int, typer.Option(envvar="LLAMA_INDEX_K", help=HELP_TEXT["k"])
+    ] = DEFAULT_ARGS["k"],
     chunk_size: Annotated[
-        Optional[int], typer.Option(envvar="LLAMA_INDEX_CHUNK_SIZE")
+        Optional[int],
+        typer.Option(envvar="LLAMA_INDEX_CHUNK_SIZE", help=HELP_TEXT["chunk_size"]),
     ] = DEFAULT_ARGS.get("chunk_size"),
     chunk_overlap_ratio: Annotated[
-        float, typer.Option(envvar="LLAMA_INDEX_CHUNK_OVERLAP_RATIO")
+        float,
+        typer.Option(
+            envvar="LLAMA_INDEX_CHUNK_OVERLAP_RATIO",
+            help=HELP_TEXT["chunk_overlap_ratio"],
+        ),
     ] = DEFAULT_ARGS["chunk_overlap_ratio"],
     num_output: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_NUM_OUTPUT")
+        int, typer.Option(envvar="LLAMA_INDEX_NUM_OUTPUT", help=HELP_TEXT["num_output"])
     ] = DEFAULT_ARGS["num_output"],
-    is_path: Annotated[bool, typer.Option(envvar="LLAMA_INDEX_IS_PATH")] = DEFAULT_ARGS[
-        "is_path"
-    ],
+    is_path: Annotated[
+        bool, typer.Option(envvar="LLAMA_INDEX_IS_PATH", help=HELP_TEXT["is_path"])
+    ] = DEFAULT_ARGS["is_path"],
     n_gpu_layers: Annotated[
-        int, typer.Option(envvar="LLAMA_INDEX_N_GPU_LAYERS")
+        int,
+        typer.Option(envvar="LLAMA_INDEX_N_GPU_LAYERS", help=HELP_TEXT["n_gpu_layers"]),
     ] = DEFAULT_ARGS["n_gpu_layers"],
-    device: Annotated[str, typer.Option(envvar="LLAMA_INDEX_DEVICE")] = DEFAULT_ARGS[
-        "device"
-    ],
+    device: Annotated[
+        str, typer.Option(envvar="LLAMA_INDEX_DEVICE", help=HELP_TEXT["device"])
+    ] = DEFAULT_ARGS["device"],
 ) -> None:
     set_up_logging_config(level=40)
     main(