feat(cli)!: Change some defaults (#28)

* feat(cli): default to use a reranker for better out-of-the-box experience.

* test(cli): update test to reflect new default

* chore(nvim): remove deprecated API

* feat(cli): set a non-negative `chunk_size` so that large files are chunked by default

* fix(cli): use `NaiveReranker` when `"reranker"` is falsy.

* docs(cli): fix documentation

* fix(cli): fix broken test due to changed default

Author: Davidyz

docs/cli.md

@@ -144,7 +144,7 @@ Please try the following and see if any of these fix your issue:
 ## Advanced Usage
 ### Initialising a Project
 For each project, VectorCode creates a collection (similar to tables in
-traditional databases) and puts the code embeddings in the corresponsing
+traditional databases) and puts the code embeddings in the corresponding
 collection. In the root directory of a project, you may run `vectorcode init`.
 This will initialise the repository with a subdirectory
 `project_root/.vectorcode/`. This will mark this directory a _project root_, a
@@ -189,8 +189,7 @@ The JSON configuration file may hold the following values:
 - `chunk_size`: integer, the maximum number of characters per chunk. A larger
   value reduces the number of items in the database, and hence accelerates the
   search, but at the cost of potentially truncated data and lost information.
-  Default: `-1` (no chunking), but it's **highly** recommended to set it to a
-  positive integer that works for your model when working with large documents;
+  Default: `2500`. To disable chunking, set it to a negative number;
 - `overlap_ratio`: float between 0 and 1, the ratio of overlapping/shared content 
   between 2 adjacent chunks. A larger ratio improves the coherences of chunks,
   but at the cost of increasing number of entries in the database and hence
@@ -203,16 +202,18 @@ The JSON configuration file may hold the following values:
   works. If VectorCode fails to find an appropriate parser, it'll fallback to
   the legacy naive parser, in which case `overlap_ratio` works exactly in the
   same way as before;_
-- `query_multplier`: integer, when you use the `query` command to retrieve `n` documents,
-  VectorCode will check `n * query_multplier` chunks and return at most `n` 
-  documents. A larger value of `query_multplier`
+- `query_multiplier`: integer, when you use the `query` command to retrieve `n` documents,
+  VectorCode will check `n * query_multiplier` chunks and return at most `n` 
+  documents. A larger value of `query_multiplier`
   guarantees the return of `n` documents, but with the risk of including too
   many less-relevant chunks that may affect the document selection. Default: 
   `-1` (any negative value means selecting documents based on all indexed chunks);
 - `reranker`: string, a reranking model supported by 
   [`CrossEncoder`](https://sbert.net/docs/package_reference/cross_encoder/index.html). 
-  A list of available models is available on their documentation. The default is
-  not to use a reranker as it increases the time needed for each query;
+  A list of available models is available on their documentation. The default
+  model is `"cross-encoder/ms-marco-MiniLM-L-6-v2"`. You can disable the use of
+  `CrossEncoder` by setting this option to a falsy value that is not `null`,
+  such as `false` or `""` (empty string);
 - `reranker_params`: dictionary, similar to `embedding_params`. The options
   passed to `CrossEncoder` class constructor;
 - `db_settings`: dictionary, works in a similar way to `embedding_params`, but 

lua/vectorcode/cacher/init.lua

@@ -1,9 +1,4 @@
-vim.deprecate(
-  'require("vectorcode.cacher")',
-  'require("vectorcode.config").get_cacher_backend()',
-  "0.5.0",
-  "VectorCode",
-  true
-)
-
-return require("vectorcode.cacher.default")
+return {
+  lsp = require("vectorcode.cacher.lsp"),
+  default = require("vectorcode.cacher.default"),
+}

src/vectorcode/cli_utils.py

@@ -64,11 +64,11 @@ class Config:
     force: bool = False
     db_path: Optional[str] = "~/.local/share/vectorcode/chromadb/"
     db_settings: Optional[dict] = None
-    chunk_size: int = -1
+    chunk_size: int = 2500
     overlap_ratio: float = 0.2
     query_multiplier: int = -1
     query_exclude: list[PathLike] = field(default_factory=list)
-    reranker: Optional[str] = None
+    reranker: Optional[str] = "cross-encoder/ms-marco-MiniLM-L-6-v2"
     reranker_params: dict[str, Any] = field(default_factory=dict)
     check_item: Optional[str] = None
     use_absolute_path: bool = False
@@ -100,10 +100,12 @@ async def import_from(cls, config_dict: dict[str, Any]) -> "Config":
                 "host": host,
                 "port": port,
                 "db_path": db_path,
-                "chunk_size": config_dict.get("chunk_size", -1),
+                "chunk_size": config_dict.get("chunk_size", 2500),
                 "overlap_ratio": config_dict.get("overlap_ratio", 0.2),
                 "query_multiplier": config_dict.get("query_multiplier", -1),
-                "reranker": config_dict.get("reranker", None),
+                "reranker": config_dict.get(
+                    "reranker", "cross-encoder/ms-marco-MiniLM-L-6-v2"
+                ),
                 "reranker_params": config_dict.get("reranker_params", {}),
                 "db_settings": config_dict.get("db_settings", None),
                 "hnsw": config_dict.get("hnsw", {}),

src/vectorcode/subcommands/query/__init__.py

@@ -57,7 +57,7 @@ async def get_query_result_files(
         # no results found
         return []
 
-    if configs.reranker is None:
+    if not configs.reranker:
         from .reranker import NaiveReranker
 
         aggregated_results = NaiveReranker(configs).rerank(results)

tests/test_cli_utils.py

@@ -106,10 +106,10 @@ async def test_config_import_from_missing_keys():
     assert config.host == "localhost"
     assert config.port == 8000
     assert config.db_path == os.path.expanduser("~/.local/share/vectorcode/chromadb/")
-    assert config.chunk_size == -1
+    assert config.chunk_size == 2500
     assert config.overlap_ratio == 0.2
     assert config.query_multiplier == -1
-    assert config.reranker is None
+    assert config.reranker == "cross-encoder/ms-marco-MiniLM-L-6-v2"
     assert config.reranker_params == {}
     assert config.db_settings is None
 

tests/test_mcp.py

@@ -84,7 +84,7 @@ async def test_query_tool_success():
         patch("os.path.relpath", return_value="rel/path.py"),
         patch("vectorcode.cli_utils.load_config_file") as mock_load_config_file,
     ):
-        mock_config = Config(chunk_size=100, overlap_ratio=0.1)
+        mock_config = Config(chunk_size=100, overlap_ratio=0.1, reranker=None)
         mock_load_config_file.return_value = mock_config
         mock_get_project_config.return_value = mock_config
         mock_client = AsyncMock()