Error implementations (#232)

Solves #216 by adding a check on table open for existence of the
`.lance` file. Does not check for it for remote connections.

.gitignore

@@ -3,6 +3,7 @@
 *.egg-info
 **/__pycache__
 .DS_Store
+venv
 
 rust/target
 rust/Cargo.lock

docs/src/examples/modal_langchain.py

@@ -79,10 +79,7 @@ def qanda_langchain(query):
     download_docs()
     docs = store_docs()
 
-    text_splitter = RecursiveCharacterTextSplitter(
-        chunk_size=1000,
-        chunk_overlap=200,
-    )
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200,)
     documents = text_splitter.split_documents(docs)
     embeddings = OpenAIEmbeddings()
 

node/README.md

@@ -16,7 +16,7 @@ npm install vectordb
 const lancedb = require('vectordb');
 const db = lancedb.connect('<PATH_TO_LANCEDB_DATASET>');
 const table = await db.openTable('my_table');
-const query = await table.search([0.1, 0.3]).limit(20).execute();
+const results = await table.search([0.1, 0.3]).limit(20).execute();
 console.log(results);
 ```
 

python/README.md

@@ -0,0 +1,85 @@
+# LanceDB
+
+A Python library for [LanceDB](https://github.com/lancedb/lancedb).
+
+## Installation
+
+```bash
+pip install lancedb
+```
+
+## Usage
+
+### Basic Example
+
+```python
+import lancedb
+db = lancedb.connect('<PATH_TO_LANCEDB_DATASET>')
+table = db.open_table('my_table')
+results = table.search([0.1, 0.3]).limit(20).to_df()
+print(results)
+```
+
+
+## Development
+
+Create a virtual environment and activate it:
+
+```bash
+python -m venv venv
+. ./venv/bin/activate
+```
+
+Install the necessary packages:
+
+```bash
+python -m pip install .
+```
+
+To run the unit tests:
+
+```bash
+pytest
+```
+
+To run linter and automatically fix all errors:
+
+```bash
+black .
+isort .
+```
+
+If any packages are missing, install them with:
+
+```bash
+pip install <PACKAGE_NAME>
+```
+
+
+___
+For **Windows** users, there may be errors when installing packages, so these commands may be helpful:
+
+Activate the virtual environment:
+```bash
+. .\venv\Scripts\activate
+```
+
+You may need to run the installs separately:
+```bash
+pip install -e .[tests]
+pip install -e .[dev]
+```
+
+
+`tantivy` requires `rust` to be installed, so install it with `conda`, as it doesn't support windows installation:
+```bash
+pip install wheel
+pip install cargo
+conda install rust
+pip install tantivy
+```
+
+To run the unit tests:
+```bash
+pytest
+```

python/lancedb/db.py

@@ -252,7 +252,7 @@ class LanceDBConnection:
         if data is not None:
             tbl = LanceTable.create(self, name, data, schema, mode=mode)
         else:
-            tbl = LanceTable(self, name)
+            tbl = LanceTable.open(self, name)
         return tbl
 
     def open_table(self, name: str) -> LanceTable:
@@ -267,7 +267,7 @@ class LanceDBConnection:
         -------
         A LanceTable object representing the table.
         """
-        return LanceTable(self, name)
+        return LanceTable.open(self, name)
 
     def drop_table(self, name: str):
         """Drop a table from the database.

python/lancedb/table.py

@@ -308,6 +308,19 @@ class LanceTable:
         lance.write_dataset(data, tbl._dataset_uri, mode=mode)
         return tbl
 
+    @classmethod
+    def open(cls, db, name):
+        tbl = cls(db, name)
+        if tbl._conn.is_managed_remote:
+            # Not completely sure how to check for remote table existence yet.
+            return tbl
+        if not os.path.exists(tbl._dataset_uri):
+            raise FileNotFoundError(
+                f"Table {name} does not exist. Please first call db.create_table({name}, data)"
+            )
+
+        return tbl
+
     def delete(self, where: str):
         """Delete rows from the table.
 

python/tests/test_db.py

@@ -123,6 +123,15 @@ def test_delete_table(tmp_path):
     assert db.table_names() == ["test"]
 
 
+def test_empty_or_nonexistent_table(tmp_path):
+    db = lancedb.connect(tmp_path)
+    with pytest.raises(Exception):
+        db.create_table("test_with_no_data")
+
+    with pytest.raises(Exception):
+        db.open_table("does_not_exist")
+
+
 def test_replace_index(tmp_path):
     db = lancedb.connect(uri=tmp_path)
     table = db.create_table(