bump version for v0.1.5-python

Co-authored-by: Will Jones <willjones127@gmail.com>

.github/workflows/python.yml

@@ -30,7 +30,8 @@ jobs:
         python-version: 3.${{ matrix.python-minor-version }}
     - name: Install lancedb
       run: |
-        pip install -e ".[fts]"
+        pip install -e .
+        pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
         pip install pytest
     - name: Run tests
       run: pytest -x -v --durations=30 tests
@@ -52,7 +53,8 @@ jobs:
         python-version: "3.11"
     - name: Install lancedb
       run: |
-        pip install -e ".[fts]"
+        pip install -e .
+        pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
         pip install pytest
     - name: Run tests
       run: pytest -x -v --durations=30 tests

Cargo.lock

@@ -1052,6 +1052,7 @@ dependencies = [
  "paste",
  "petgraph",
  "rand",
+ "regex",
  "uuid",
 ]
 
@@ -1645,9 +1646,9 @@ dependencies = [
 
 [[package]]
 name = "lance"
-version = "0.4.12"
+version = "0.4.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "fc96cf89139af6f439a0e28ccd04ddf81be795b79fda3105b7a8952fadeb778e"
+checksum = "86dda8185bd1ffae7b910c1f68035af23be9b717c52e9cc4de176cd30b47f772"
 dependencies = [
  "accelerate-src",
  "arrow",
@@ -1684,6 +1685,7 @@ dependencies = [
  "rand",
  "reqwest",
  "shellexpand",
+ "snafu",
  "sqlparser-lance",
  "tokio",
  "url",
@@ -3362,7 +3364,9 @@ dependencies = [
  "arrow-data",
  "arrow-schema",
  "lance",
+ "object_store",
  "rand",
+ "snafu",
  "tempfile",
  "tokio",
 ]

docs/src/fts.md

@@ -6,9 +6,10 @@ to make this available for JS as well.
 
 ## Installation
 
-To use full text search, you must install the fts optional dependencies:
+To use full text search, you must install optional dependency tantivy-py:
 
-`pip install lancedb[fts]`
+# tantivy 0.19.2
+pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985
 
 
 ## Quickstart

docs/src/index.md

@@ -38,7 +38,7 @@ result = table.search([100, 100]).limit(2).to_df()
 ## Complete Demos
 
 We will be adding completed demo apps built using LanceDB.
-- [YouTube Transcript Search](../notebooks/youtube_transcript_search.ipynb)
+- [YouTube Transcript Search](../../notebooks/youtube_transcript_search.ipynb)
 
 
 ## Documentation Quick Links

node/examples/js-openai/index.js

@@ -0,0 +1,41 @@
+// Copyright 2023 Lance Developers.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+'use strict'
+
+async function example () {
+  const lancedb = require('vectordb')
+  // You need to provide an OpenAI API key, here we read it from the OPENAI_API_KEY environment variable
+  const apiKey = process.env.OPENAI_API_KEY
+  // The embedding function will create embeddings for the 'text' column(text in this case)
+  const embedding = new lancedb.OpenAIEmbeddingFunction('text', apiKey)
+
+  const db = await lancedb.connect('data/sample-lancedb')
+
+  const data = [
+    { id: 1, text: 'Black T-Shirt', price: 10 },
+    { id: 2, text: 'Leather Jacket', price: 50 }
+  ]
+
+  const table = await db.createTable('vectors', data, embedding)
+  console.log(await db.tableNames())
+
+  const results = await table
+    .search('keeps me warm')
+    .limit(1)
+    .execute()
+  console.log(results[0].text)
+}
+
+example().then(_ => { console.log('All done!') })

node/examples/js-openai/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "vectordb-example-js-openai",
+  "version": "1.0.0",
+  "description": "",
+  "main": "index.js",
+  "scripts": {
+    "test": "echo \"Error: no test specified\" && exit 1"
+  },
+  "author": "Lance Devs",
+  "license": "Apache-2.0",
+  "dependencies": {
+    "vectordb": "file:../..",
+    "openai": "^3.2.1"
+  }
+}

node/examples/js/package.json

@@ -9,6 +9,6 @@
   "author": "Lance Devs",
   "license": "Apache-2.0",
   "dependencies": {
-    "vectordb": "^0.1.0"
+    "vectordb": "file:../.."
   }
 }

node/examples/ts/package.json

@@ -17,6 +17,6 @@
     "typescript": "*"
   },
   "dependencies": {
-    "vectordb": "^0.1.0"
+    "vectordb": "file:../.."
   }
 }

node/package-lock.json

@@ -1,12 +1,12 @@
 {
   "name": "vectordb",
-  "version": "0.1.1",
+  "version": "0.1.3",
   "lockfileVersion": 2,
   "requires": true,
   "packages": {
     "": {
       "name": "vectordb",
-      "version": "0.1.1",
+      "version": "0.1.3",
       "license": "Apache-2.0",
       "dependencies": {
         "@apache-arrow/ts": "^12.0.0",
@@ -16,6 +16,7 @@
         "@types/chai": "^4.3.4",
         "@types/mocha": "^10.0.1",
         "@types/node": "^18.16.2",
+        "@types/sinon": "^10.0.15",
         "@types/temp": "^0.9.1",
         "@typescript-eslint/eslint-plugin": "^5.59.1",
         "cargo-cp-artifact": "^0.1",
@@ -26,6 +27,8 @@
         "eslint-plugin-n": "^15.7.0",
         "eslint-plugin-promise": "^6.1.1",
         "mocha": "^10.2.0",
+        "openai": "^3.2.1",
+        "sinon": "^15.1.0",
         "temp": "^0.9.4",
         "ts-node": "^10.9.1",
         "ts-node-dev": "^2.0.0",
@@ -232,6 +235,50 @@
         "node": ">= 8"
       }
     },
+    "node_modules/@sinonjs/commons": {
+      "version": "3.0.0",
+      "resolved": "https://registry.npmjs.org/@sinonjs/commons/-/commons-3.0.0.tgz",
+      "integrity": "sha512-jXBtWAF4vmdNmZgD5FoKsVLv3rPgDnLgPbU84LIJ3otV44vJlDRokVng5v8NFJdCf/da9legHcKaRuZs4L7faA==",
+      "dev": true,
+      "dependencies": {
+        "type-detect": "4.0.8"
+      }
+    },
+    "node_modules/@sinonjs/fake-timers": {
+      "version": "10.2.0",
+      "resolved": "https://registry.npmjs.org/@sinonjs/fake-timers/-/fake-timers-10.2.0.tgz",
+      "integrity": "sha512-OPwQlEdg40HAj5KNF8WW6q2KG4Z+cBCZb3m4ninfTZKaBmbIJodviQsDBoYMPHkOyJJMHnOJo5j2+LKDOhOACg==",
+      "dev": true,
+      "dependencies": {
+        "@sinonjs/commons": "^3.0.0"
+      }
+    },
+    "node_modules/@sinonjs/samsam": {
+      "version": "8.0.0",
+      "resolved": "https://registry.npmjs.org/@sinonjs/samsam/-/samsam-8.0.0.tgz",
+      "integrity": "sha512-Bp8KUVlLp8ibJZrnvq2foVhP0IVX2CIprMJPK0vqGqgrDa0OHVKeZyBykqskkrdxV6yKBPmGasO8LVjAKR3Gew==",
+      "dev": true,
+      "dependencies": {
+        "@sinonjs/commons": "^2.0.0",
+        "lodash.get": "^4.4.2",
+        "type-detect": "^4.0.8"
+      }
+    },
+    "node_modules/@sinonjs/samsam/node_modules/@sinonjs/commons": {
+      "version": "2.0.0",
+      "resolved": "https://registry.npmjs.org/@sinonjs/commons/-/commons-2.0.0.tgz",
+      "integrity": "sha512-uLa0j859mMrg2slwQYdO/AkrOfmH+X6LTVmNTS9CqexuE2IvVORIkSpJLqePAbEnKJ77aMmCwr1NUZ57120Xcg==",
+      "dev": true,
+      "dependencies": {
+        "type-detect": "4.0.8"
+      }
+    },
+    "node_modules/@sinonjs/text-encoding": {
+      "version": "0.7.2",
+      "resolved": "https://registry.npmjs.org/@sinonjs/text-encoding/-/text-encoding-0.7.2.tgz",
+      "integrity": "sha512-sXXKG+uL9IrKqViTtao2Ws6dy0znu9sOaP1di/jKGW1M6VssO8vlpXCQcpZ+jisQ1tTFAC5Jo/EOzFbggBagFQ==",
+      "dev": true
+    },
     "node_modules/@tsconfig/node10": {
       "version": "1.0.9",
       "resolved": "https://registry.npmjs.org/@tsconfig/node10/-/node10-1.0.9.tgz",
@@ -307,6 +354,21 @@
       "integrity": "sha512-21cFJr9z3g5dW8B0CVI9g2O9beqaThGQ6ZFBqHfwhzLDKUxaqTIy3vnfah/UPkfOiF2pLq+tGz+W8RyCskuslw==",
       "dev": true
     },
+    "node_modules/@types/sinon": {
+      "version": "10.0.15",
+      "resolved": "https://registry.npmjs.org/@types/sinon/-/sinon-10.0.15.tgz",
+      "integrity": "sha512-3lrFNQG0Kr2LDzvjyjB6AMJk4ge+8iYhQfdnSwIwlG88FUOV43kPcQqDZkDa/h3WSZy6i8Fr0BSjfQtB1B3xuQ==",
+      "dev": true,
+      "dependencies": {
+        "@types/sinonjs__fake-timers": "*"
+      }
+    },
+    "node_modules/@types/sinonjs__fake-timers": {
+      "version": "8.1.2",
+      "resolved": "https://registry.npmjs.org/@types/sinonjs__fake-timers/-/sinonjs__fake-timers-8.1.2.tgz",
+      "integrity": "sha512-9GcLXF0/v3t80caGs5p2rRfkB+a8VBGLJZVih6CNFkx8IZ994wiKKLSRs9nuFwk1HevWs/1mnUmkApGrSGsShA==",
+      "dev": true
+    },
     "node_modules/@types/strip-bom": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/@types/strip-bom/-/strip-bom-3.0.0.tgz",
@@ -744,6 +806,12 @@
         "node": "*"
       }
     },
+    "node_modules/asynckit": {
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/asynckit/-/asynckit-0.4.0.tgz",
+      "integrity": "sha512-Oei9OH4tRh0YqU3GxhX79dM/mwVgvbZJaSNaRk+bshkj0S5cfHcgYakreBjrHwatXKbz+IoIdYLxrKim2MjW0Q==",
+      "dev": true
+    },
     "node_modules/available-typed-arrays": {
       "version": "1.0.5",
       "resolved": "https://registry.npmjs.org/available-typed-arrays/-/available-typed-arrays-1.0.5.tgz",
@@ -756,6 +824,15 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/axios": {
+      "version": "0.26.1",
+      "resolved": "https://registry.npmjs.org/axios/-/axios-0.26.1.tgz",
+      "integrity": "sha512-fPwcX4EvnSHuInCMItEhAGnaSEXRBjtzh9fOtsE6E1G6p7vl7edEeZe11QHf18+6+9gR5PbKV/sGKNaD8YaMeA==",
+      "dev": true,
+      "dependencies": {
+        "follow-redirects": "^1.14.8"
+      }
+    },
     "node_modules/balanced-match": {
       "version": "1.0.2",
       "resolved": "https://registry.npmjs.org/balanced-match/-/balanced-match-1.0.2.tgz",
@@ -968,6 +1045,18 @@
       "integrity": "sha512-dOy+3AuW3a2wNbZHIuMZpTcgjGuLU/uBL/ubcZF9OXbDo8ff4O8yVp5Bf0efS8uEoYo5q4Fx7dY9OgQGXgAsQA==",
       "dev": true
     },
+    "node_modules/combined-stream": {
+      "version": "1.0.8",
+      "resolved": "https://registry.npmjs.org/combined-stream/-/combined-stream-1.0.8.tgz",
+      "integrity": "sha512-FQN4MRfuJeHf7cBbBMJFXhKSDq+2kAArBlmRBvcvFE5BB1HZKXtSFASDhdlz9zOYwxh8lDdnvmMOe/+5cdoEdg==",
+      "dev": true,
+      "dependencies": {
+        "delayed-stream": "~1.0.0"
+      },
+      "engines": {
+        "node": ">= 0.8"
+      }
+    },
     "node_modules/command-line-args": {
       "version": "5.2.1",
       "resolved": "https://registry.npmjs.org/command-line-args/-/command-line-args-5.2.1.tgz",
@@ -1179,6 +1268,15 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/delayed-stream": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/delayed-stream/-/delayed-stream-1.0.0.tgz",
+      "integrity": "sha512-ZySD7Nf91aLB0RxL4KGrKHBXl7Eds1DAmEdcoVawXnLD7SDhpNgtuII2aAkg7a7QS41jxPSZ17p4VdGnMHk3MQ==",
+      "dev": true,
+      "engines": {
+        "node": ">=0.4.0"
+      }
+    },
     "node_modules/diff": {
       "version": "4.0.2",
       "resolved": "https://registry.npmjs.org/diff/-/diff-4.0.2.tgz",
@@ -1942,6 +2040,26 @@
       "integrity": "sha512-5nqDSxl8nn5BSNxyR3n4I6eDmbolI6WT+QqR547RwxQapgjQBmtktdP+HTBb/a/zLsbzERTONyUB5pefh5TtjQ==",
       "dev": true
     },
+    "node_modules/follow-redirects": {
+      "version": "1.15.2",
+      "resolved": "https://registry.npmjs.org/follow-redirects/-/follow-redirects-1.15.2.tgz",
+      "integrity": "sha512-VQLG33o04KaQ8uYi2tVNbdrWp1QWxNNea+nmIB4EVM28v0hmP17z7aG1+wAkNzVq4KeXTq3221ye5qTJP91JwA==",
+      "dev": true,
+      "funding": [
+        {
+          "type": "individual",
+          "url": "https://github.com/sponsors/RubenVerborgh"
+        }
+      ],
+      "engines": {
+        "node": ">=4.0"
+      },
+      "peerDependenciesMeta": {
+        "debug": {
+          "optional": true
+        }
+      }
+    },
     "node_modules/for-each": {
       "version": "0.3.3",
       "resolved": "https://registry.npmjs.org/for-each/-/for-each-0.3.3.tgz",
@@ -1951,6 +2069,20 @@
         "is-callable": "^1.1.3"
       }
     },
+    "node_modules/form-data": {
+      "version": "4.0.0",
+      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.0.tgz",
+      "integrity": "sha512-ETEklSGi5t0QMZuiXoA/Q6vcnxcLQP5vdugSpuAyi6SVGi2clPPp+xgEhuMaHC+zGgn31Kd235W35f7Hykkaww==",
+      "dev": true,
+      "dependencies": {
+        "asynckit": "^0.4.0",
+        "combined-stream": "^1.0.8",
+        "mime-types": "^2.1.12"
+      },
+      "engines": {
+        "node": ">= 6"
+      }
+    },
     "node_modules/fs.realpath": {
       "version": "1.0.0",
       "resolved": "https://registry.npmjs.org/fs.realpath/-/fs.realpath-1.0.0.tgz",
@@ -2584,6 +2716,12 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/isarray": {
+      "version": "0.0.1",
+      "resolved": "https://registry.npmjs.org/isarray/-/isarray-0.0.1.tgz",
+      "integrity": "sha512-D2S+3GLxWH+uhrNEcoh/fnmYeP8E8/zHl644d/jdA0g2uyXvy3sb0qxotE+ne0LtccHknQzWwZEzhak7oJ0COQ==",
+      "dev": true
+    },
     "node_modules/isexe": {
       "version": "2.0.0",
       "resolved": "https://registry.npmjs.org/isexe/-/isexe-2.0.0.tgz",
@@ -2644,6 +2782,12 @@
         "json5": "lib/cli.js"
       }
     },
+    "node_modules/just-extend": {
+      "version": "4.2.1",
+      "resolved": "https://registry.npmjs.org/just-extend/-/just-extend-4.2.1.tgz",
+      "integrity": "sha512-g3UB796vUFIY90VIv/WX3L2c8CS2MdWUww3CNrYmqza1Fg0DURc2K/O4YrnklBdQarSJ/y8JnJYDGc+1iumQjg==",
+      "dev": true
+    },
     "node_modules/levn": {
       "version": "0.4.1",
       "resolved": "https://registry.npmjs.org/levn/-/levn-0.4.1.tgz",
@@ -2677,6 +2821,12 @@
       "resolved": "https://registry.npmjs.org/lodash.camelcase/-/lodash.camelcase-4.3.0.tgz",
       "integrity": "sha512-TwuEnCnxbc3rAvhf/LbG7tJUDzhqXyFnv3dtzLOPgCG/hODL7WFnsbwktkD7yUV0RrreP/l1PALq/YSg6VvjlA=="
     },
+    "node_modules/lodash.get": {
+      "version": "4.4.2",
+      "resolved": "https://registry.npmjs.org/lodash.get/-/lodash.get-4.4.2.tgz",
+      "integrity": "sha512-z+Uw/vLuy6gQe8cfaFWD7p0wVv8fJl3mbzXh33RS+0oW2wvUqiRXiQ69gLWSLpgB5/6sU+r6BlQR0MBILadqTQ==",
+      "dev": true
+    },
     "node_modules/lodash.merge": {
       "version": "4.6.2",
       "resolved": "https://registry.npmjs.org/lodash.merge/-/lodash.merge-4.6.2.tgz",
@@ -2748,6 +2898,27 @@
         "node": ">=8.6"
       }
     },
+    "node_modules/mime-db": {
+      "version": "1.52.0",
+      "resolved": "https://registry.npmjs.org/mime-db/-/mime-db-1.52.0.tgz",
+      "integrity": "sha512-sPU4uV7dYlvtWJxwwxHD0PuihVNiE7TyAbQ5SWxDCB9mUYvOgroQOwYQQOKPJ8CIbE+1ETVlOoK1UC2nU3gYvg==",
+      "dev": true,
+      "engines": {
+        "node": ">= 0.6"
+      }
+    },
+    "node_modules/mime-types": {
+      "version": "2.1.35",
+      "resolved": "https://registry.npmjs.org/mime-types/-/mime-types-2.1.35.tgz",
+      "integrity": "sha512-ZDY+bPm5zTTF+YpCrAU9nK0UgICYPT0QtT1NZWFv4s++TNkcgVaT0g6+4R2uI4MjQjzysHB1zxuWL50hzaeXiw==",
+      "dev": true,
+      "dependencies": {
+        "mime-db": "1.52.0"
+      },
+      "engines": {
+        "node": ">= 0.6"
+      }
+    },
     "node_modules/minimatch": {
       "version": "3.1.2",
       "resolved": "https://registry.npmjs.org/minimatch/-/minimatch-3.1.2.tgz",
@@ -2925,6 +3096,28 @@
       "integrity": "sha512-Tj+HTDSJJKaZnfiuw+iaF9skdPpTo2GtEly5JHnWV/hfv2Qj/9RKsGISQtLh2ox3l5EAGw487hnBee0sIJ6v2g==",
       "dev": true
     },
+    "node_modules/nise": {
+      "version": "5.1.4",
+      "resolved": "https://registry.npmjs.org/nise/-/nise-5.1.4.tgz",
+      "integrity": "sha512-8+Ib8rRJ4L0o3kfmyVCL7gzrohyDe0cMFTBa2d364yIrEGMEoetznKJx899YxjybU6bL9SQkYPSBBs1gyYs8Xg==",
+      "dev": true,
+      "dependencies": {
+        "@sinonjs/commons": "^2.0.0",
+        "@sinonjs/fake-timers": "^10.0.2",
+        "@sinonjs/text-encoding": "^0.7.1",
+        "just-extend": "^4.0.2",
+        "path-to-regexp": "^1.7.0"
+      }
+    },
+    "node_modules/nise/node_modules/@sinonjs/commons": {
+      "version": "2.0.0",
+      "resolved": "https://registry.npmjs.org/@sinonjs/commons/-/commons-2.0.0.tgz",
+      "integrity": "sha512-uLa0j859mMrg2slwQYdO/AkrOfmH+X6LTVmNTS9CqexuE2IvVORIkSpJLqePAbEnKJ77aMmCwr1NUZ57120Xcg==",
+      "dev": true,
+      "dependencies": {
+        "type-detect": "4.0.8"
+      }
+    },
     "node_modules/normalize-path": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/normalize-path/-/normalize-path-3.0.0.tgz",
@@ -2996,6 +3189,16 @@
         "wrappy": "1"
       }
     },
+    "node_modules/openai": {
+      "version": "3.2.1",
+      "resolved": "https://registry.npmjs.org/openai/-/openai-3.2.1.tgz",
+      "integrity": "sha512-762C9BNlJPbjjlWZi4WYK9iM2tAVAv0uUp1UmI34vb0CN5T2mjB/qM6RYBmNKMh/dN9fC+bxqPwWJZUTWW052A==",
+      "dev": true,
+      "dependencies": {
+        "axios": "^0.26.0",
+        "form-data": "^4.0.0"
+      }
+    },
     "node_modules/optionator": {
       "version": "0.9.1",
       "resolved": "https://registry.npmjs.org/optionator/-/optionator-0.9.1.tgz",
@@ -3099,6 +3302,15 @@
       "integrity": "sha512-LDJzPVEEEPR+y48z93A0Ed0yXb8pAByGWo/k5YYdYgpY2/2EsOsksJrq7lOHxryrVOn1ejG6oAp8ahvOIQD8sw==",
       "dev": true
     },
+    "node_modules/path-to-regexp": {
+      "version": "1.8.0",
+      "resolved": "https://registry.npmjs.org/path-to-regexp/-/path-to-regexp-1.8.0.tgz",
+      "integrity": "sha512-n43JRhlUKUAlibEJhPeir1ncUID16QnEjNpwzNdO3Lm4ywrBpBZ5oLD0I6br9evr1Y9JTqwRtAh7JLoOzAQdVA==",
+      "dev": true,
+      "dependencies": {
+        "isarray": "0.0.1"
+      }
+    },
     "node_modules/path-type": {
       "version": "4.0.0",
       "resolved": "https://registry.npmjs.org/path-type/-/path-type-4.0.0.tgz",
@@ -3406,6 +3618,45 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/sinon": {
+      "version": "15.1.0",
+      "resolved": "https://registry.npmjs.org/sinon/-/sinon-15.1.0.tgz",
+      "integrity": "sha512-cS5FgpDdE9/zx7no8bxROHymSlPLZzq0ChbbLk1DrxBfc+eTeBK3y8nIL+nu/0QeYydhhbLIr7ecHJpywjQaoQ==",
+      "dev": true,
+      "dependencies": {
+        "@sinonjs/commons": "^3.0.0",
+        "@sinonjs/fake-timers": "^10.2.0",
+        "@sinonjs/samsam": "^8.0.0",
+        "diff": "^5.1.0",
+        "nise": "^5.1.4",
+        "supports-color": "^7.2.0"
+      },
+      "funding": {
+        "type": "opencollective",
+        "url": "https://opencollective.com/sinon"
+      }
+    },
+    "node_modules/sinon/node_modules/diff": {
+      "version": "5.1.0",
+      "resolved": "https://registry.npmjs.org/diff/-/diff-5.1.0.tgz",
+      "integrity": "sha512-D+mk+qE8VC/PAUrlAU34N+VfXev0ghe5ywmpqrawphmVZc1bEfn56uo9qpyGp1p4xpzOHkSW4ztBd6L7Xx4ACw==",
+      "dev": true,
+      "engines": {
+        "node": ">=0.3.1"
+      }
+    },
+    "node_modules/sinon/node_modules/supports-color": {
+      "version": "7.2.0",
+      "resolved": "https://registry.npmjs.org/supports-color/-/supports-color-7.2.0.tgz",
+      "integrity": "sha512-qpCAvRl9stuOHveKsn7HncJRvv501qIacKzQlO/+Lwxc9+0q2wLyv4Dfvt80/DPn2pqOBsJdDiogXGR9+OvwRw==",
+      "dev": true,
+      "dependencies": {
+        "has-flag": "^4.0.0"
+      },
+      "engines": {
+        "node": ">=8"
+      }
+    },
     "node_modules/slash": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/slash/-/slash-3.0.0.tgz",
@@ -4217,6 +4468,52 @@
         "fastq": "^1.6.0"
       }
     },
+    "@sinonjs/commons": {
+      "version": "3.0.0",
+      "resolved": "https://registry.npmjs.org/@sinonjs/commons/-/commons-3.0.0.tgz",
+      "integrity": "sha512-jXBtWAF4vmdNmZgD5FoKsVLv3rPgDnLgPbU84LIJ3otV44vJlDRokVng5v8NFJdCf/da9legHcKaRuZs4L7faA==",
+      "dev": true,
+      "requires": {
+        "type-detect": "4.0.8"
+      }
+    },
+    "@sinonjs/fake-timers": {
+      "version": "10.2.0",
+      "resolved": "https://registry.npmjs.org/@sinonjs/fake-timers/-/fake-timers-10.2.0.tgz",
+      "integrity": "sha512-OPwQlEdg40HAj5KNF8WW6q2KG4Z+cBCZb3m4ninfTZKaBmbIJodviQsDBoYMPHkOyJJMHnOJo5j2+LKDOhOACg==",
+      "dev": true,
+      "requires": {
+        "@sinonjs/commons": "^3.0.0"
+      }
+    },
+    "@sinonjs/samsam": {
+      "version": "8.0.0",
+      "resolved": "https://registry.npmjs.org/@sinonjs/samsam/-/samsam-8.0.0.tgz",
+      "integrity": "sha512-Bp8KUVlLp8ibJZrnvq2foVhP0IVX2CIprMJPK0vqGqgrDa0OHVKeZyBykqskkrdxV6yKBPmGasO8LVjAKR3Gew==",
+      "dev": true,
+      "requires": {
+        "@sinonjs/commons": "^2.0.0",
+        "lodash.get": "^4.4.2",
+        "type-detect": "^4.0.8"
+      },
+      "dependencies": {
+        "@sinonjs/commons": {
+          "version": "2.0.0",
+          "resolved": "https://registry.npmjs.org/@sinonjs/commons/-/commons-2.0.0.tgz",
+          "integrity": "sha512-uLa0j859mMrg2slwQYdO/AkrOfmH+X6LTVmNTS9CqexuE2IvVORIkSpJLqePAbEnKJ77aMmCwr1NUZ57120Xcg==",
+          "dev": true,
+          "requires": {
+            "type-detect": "4.0.8"
+          }
+        }
+      }
+    },
+    "@sinonjs/text-encoding": {
+      "version": "0.7.2",
+      "resolved": "https://registry.npmjs.org/@sinonjs/text-encoding/-/text-encoding-0.7.2.tgz",
+      "integrity": "sha512-sXXKG+uL9IrKqViTtao2Ws6dy0znu9sOaP1di/jKGW1M6VssO8vlpXCQcpZ+jisQ1tTFAC5Jo/EOzFbggBagFQ==",
+      "dev": true
+    },
     "@tsconfig/node10": {
       "version": "1.0.9",
       "resolved": "https://registry.npmjs.org/@tsconfig/node10/-/node10-1.0.9.tgz",
@@ -4292,6 +4589,21 @@
       "integrity": "sha512-21cFJr9z3g5dW8B0CVI9g2O9beqaThGQ6ZFBqHfwhzLDKUxaqTIy3vnfah/UPkfOiF2pLq+tGz+W8RyCskuslw==",
       "dev": true
     },
+    "@types/sinon": {
+      "version": "10.0.15",
+      "resolved": "https://registry.npmjs.org/@types/sinon/-/sinon-10.0.15.tgz",
+      "integrity": "sha512-3lrFNQG0Kr2LDzvjyjB6AMJk4ge+8iYhQfdnSwIwlG88FUOV43kPcQqDZkDa/h3WSZy6i8Fr0BSjfQtB1B3xuQ==",
+      "dev": true,
+      "requires": {
+        "@types/sinonjs__fake-timers": "*"
+      }
+    },
+    "@types/sinonjs__fake-timers": {
+      "version": "8.1.2",
+      "resolved": "https://registry.npmjs.org/@types/sinonjs__fake-timers/-/sinonjs__fake-timers-8.1.2.tgz",
+      "integrity": "sha512-9GcLXF0/v3t80caGs5p2rRfkB+a8VBGLJZVih6CNFkx8IZ994wiKKLSRs9nuFwk1HevWs/1mnUmkApGrSGsShA==",
+      "dev": true
+    },
     "@types/strip-bom": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/@types/strip-bom/-/strip-bom-3.0.0.tgz",
@@ -4579,12 +4891,27 @@
       "integrity": "sha512-jgsaNduz+ndvGyFt3uSuWqvy4lCnIJiovtouQN5JZHOKCS2QuhEdbcQHFhVksz2N2U9hXJo8odG7ETyWlEeuDw==",
       "dev": true
     },
+    "asynckit": {
+      "version": "0.4.0",
+      "resolved": "https://registry.npmjs.org/asynckit/-/asynckit-0.4.0.tgz",
+      "integrity": "sha512-Oei9OH4tRh0YqU3GxhX79dM/mwVgvbZJaSNaRk+bshkj0S5cfHcgYakreBjrHwatXKbz+IoIdYLxrKim2MjW0Q==",
+      "dev": true
+    },
     "available-typed-arrays": {
       "version": "1.0.5",
       "resolved": "https://registry.npmjs.org/available-typed-arrays/-/available-typed-arrays-1.0.5.tgz",
       "integrity": "sha512-DMD0KiN46eipeziST1LPP/STfDU0sufISXmjSgvVsoU2tqxctQeASejWcfNtxYKqETM1UxQ8sp2OrSBWpHY6sw==",
       "dev": true
     },
+    "axios": {
+      "version": "0.26.1",
+      "resolved": "https://registry.npmjs.org/axios/-/axios-0.26.1.tgz",
+      "integrity": "sha512-fPwcX4EvnSHuInCMItEhAGnaSEXRBjtzh9fOtsE6E1G6p7vl7edEeZe11QHf18+6+9gR5PbKV/sGKNaD8YaMeA==",
+      "dev": true,
+      "requires": {
+        "follow-redirects": "^1.14.8"
+      }
+    },
     "balanced-match": {
       "version": "1.0.2",
       "resolved": "https://registry.npmjs.org/balanced-match/-/balanced-match-1.0.2.tgz",
@@ -4749,6 +5076,15 @@
       "integrity": "sha512-dOy+3AuW3a2wNbZHIuMZpTcgjGuLU/uBL/ubcZF9OXbDo8ff4O8yVp5Bf0efS8uEoYo5q4Fx7dY9OgQGXgAsQA==",
       "dev": true
     },
+    "combined-stream": {
+      "version": "1.0.8",
+      "resolved": "https://registry.npmjs.org/combined-stream/-/combined-stream-1.0.8.tgz",
+      "integrity": "sha512-FQN4MRfuJeHf7cBbBMJFXhKSDq+2kAArBlmRBvcvFE5BB1HZKXtSFASDhdlz9zOYwxh8lDdnvmMOe/+5cdoEdg==",
+      "dev": true,
+      "requires": {
+        "delayed-stream": "~1.0.0"
+      }
+    },
     "command-line-args": {
       "version": "5.2.1",
       "resolved": "https://registry.npmjs.org/command-line-args/-/command-line-args-5.2.1.tgz",
@@ -4908,6 +5244,12 @@
         "object-keys": "^1.1.1"
       }
     },
+    "delayed-stream": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/delayed-stream/-/delayed-stream-1.0.0.tgz",
+      "integrity": "sha512-ZySD7Nf91aLB0RxL4KGrKHBXl7Eds1DAmEdcoVawXnLD7SDhpNgtuII2aAkg7a7QS41jxPSZ17p4VdGnMHk3MQ==",
+      "dev": true
+    },
     "diff": {
       "version": "4.0.2",
       "resolved": "https://registry.npmjs.org/diff/-/diff-4.0.2.tgz",
@@ -5475,6 +5817,12 @@
       "integrity": "sha512-5nqDSxl8nn5BSNxyR3n4I6eDmbolI6WT+QqR547RwxQapgjQBmtktdP+HTBb/a/zLsbzERTONyUB5pefh5TtjQ==",
       "dev": true
     },
+    "follow-redirects": {
+      "version": "1.15.2",
+      "resolved": "https://registry.npmjs.org/follow-redirects/-/follow-redirects-1.15.2.tgz",
+      "integrity": "sha512-VQLG33o04KaQ8uYi2tVNbdrWp1QWxNNea+nmIB4EVM28v0hmP17z7aG1+wAkNzVq4KeXTq3221ye5qTJP91JwA==",
+      "dev": true
+    },
     "for-each": {
       "version": "0.3.3",
       "resolved": "https://registry.npmjs.org/for-each/-/for-each-0.3.3.tgz",
@@ -5484,6 +5832,17 @@
         "is-callable": "^1.1.3"
       }
     },
+    "form-data": {
+      "version": "4.0.0",
+      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.0.tgz",
+      "integrity": "sha512-ETEklSGi5t0QMZuiXoA/Q6vcnxcLQP5vdugSpuAyi6SVGi2clPPp+xgEhuMaHC+zGgn31Kd235W35f7Hykkaww==",
+      "dev": true,
+      "requires": {
+        "asynckit": "^0.4.0",
+        "combined-stream": "^1.0.8",
+        "mime-types": "^2.1.12"
+      }
+    },
     "fs.realpath": {
       "version": "1.0.0",
       "resolved": "https://registry.npmjs.org/fs.realpath/-/fs.realpath-1.0.0.tgz",
@@ -5912,6 +6271,12 @@
         "call-bind": "^1.0.2"
       }
     },
+    "isarray": {
+      "version": "0.0.1",
+      "resolved": "https://registry.npmjs.org/isarray/-/isarray-0.0.1.tgz",
+      "integrity": "sha512-D2S+3GLxWH+uhrNEcoh/fnmYeP8E8/zHl644d/jdA0g2uyXvy3sb0qxotE+ne0LtccHknQzWwZEzhak7oJ0COQ==",
+      "dev": true
+    },
     "isexe": {
       "version": "2.0.0",
       "resolved": "https://registry.npmjs.org/isexe/-/isexe-2.0.0.tgz",
@@ -5959,6 +6324,12 @@
         "minimist": "^1.2.0"
       }
     },
+    "just-extend": {
+      "version": "4.2.1",
+      "resolved": "https://registry.npmjs.org/just-extend/-/just-extend-4.2.1.tgz",
+      "integrity": "sha512-g3UB796vUFIY90VIv/WX3L2c8CS2MdWUww3CNrYmqza1Fg0DURc2K/O4YrnklBdQarSJ/y8JnJYDGc+1iumQjg==",
+      "dev": true
+    },
     "levn": {
       "version": "0.4.1",
       "resolved": "https://registry.npmjs.org/levn/-/levn-0.4.1.tgz",
@@ -5983,6 +6354,12 @@
       "resolved": "https://registry.npmjs.org/lodash.camelcase/-/lodash.camelcase-4.3.0.tgz",
       "integrity": "sha512-TwuEnCnxbc3rAvhf/LbG7tJUDzhqXyFnv3dtzLOPgCG/hODL7WFnsbwktkD7yUV0RrreP/l1PALq/YSg6VvjlA=="
     },
+    "lodash.get": {
+      "version": "4.4.2",
+      "resolved": "https://registry.npmjs.org/lodash.get/-/lodash.get-4.4.2.tgz",
+      "integrity": "sha512-z+Uw/vLuy6gQe8cfaFWD7p0wVv8fJl3mbzXh33RS+0oW2wvUqiRXiQ69gLWSLpgB5/6sU+r6BlQR0MBILadqTQ==",
+      "dev": true
+    },
     "lodash.merge": {
       "version": "4.6.2",
       "resolved": "https://registry.npmjs.org/lodash.merge/-/lodash.merge-4.6.2.tgz",
@@ -6039,6 +6416,21 @@
         "picomatch": "^2.3.1"
       }
     },
+    "mime-db": {
+      "version": "1.52.0",
+      "resolved": "https://registry.npmjs.org/mime-db/-/mime-db-1.52.0.tgz",
+      "integrity": "sha512-sPU4uV7dYlvtWJxwwxHD0PuihVNiE7TyAbQ5SWxDCB9mUYvOgroQOwYQQOKPJ8CIbE+1ETVlOoK1UC2nU3gYvg==",
+      "dev": true
+    },
+    "mime-types": {
+      "version": "2.1.35",
+      "resolved": "https://registry.npmjs.org/mime-types/-/mime-types-2.1.35.tgz",
+      "integrity": "sha512-ZDY+bPm5zTTF+YpCrAU9nK0UgICYPT0QtT1NZWFv4s++TNkcgVaT0g6+4R2uI4MjQjzysHB1zxuWL50hzaeXiw==",
+      "dev": true,
+      "requires": {
+        "mime-db": "1.52.0"
+      }
+    },
     "minimatch": {
       "version": "3.1.2",
       "resolved": "https://registry.npmjs.org/minimatch/-/minimatch-3.1.2.tgz",
@@ -6172,6 +6564,30 @@
       "integrity": "sha512-Tj+HTDSJJKaZnfiuw+iaF9skdPpTo2GtEly5JHnWV/hfv2Qj/9RKsGISQtLh2ox3l5EAGw487hnBee0sIJ6v2g==",
       "dev": true
     },
+    "nise": {
+      "version": "5.1.4",
+      "resolved": "https://registry.npmjs.org/nise/-/nise-5.1.4.tgz",
+      "integrity": "sha512-8+Ib8rRJ4L0o3kfmyVCL7gzrohyDe0cMFTBa2d364yIrEGMEoetznKJx899YxjybU6bL9SQkYPSBBs1gyYs8Xg==",
+      "dev": true,
+      "requires": {
+        "@sinonjs/commons": "^2.0.0",
+        "@sinonjs/fake-timers": "^10.0.2",
+        "@sinonjs/text-encoding": "^0.7.1",
+        "just-extend": "^4.0.2",
+        "path-to-regexp": "^1.7.0"
+      },
+      "dependencies": {
+        "@sinonjs/commons": {
+          "version": "2.0.0",
+          "resolved": "https://registry.npmjs.org/@sinonjs/commons/-/commons-2.0.0.tgz",
+          "integrity": "sha512-uLa0j859mMrg2slwQYdO/AkrOfmH+X6LTVmNTS9CqexuE2IvVORIkSpJLqePAbEnKJ77aMmCwr1NUZ57120Xcg==",
+          "dev": true,
+          "requires": {
+            "type-detect": "4.0.8"
+          }
+        }
+      }
+    },
     "normalize-path": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/normalize-path/-/normalize-path-3.0.0.tgz",
@@ -6222,6 +6638,16 @@
         "wrappy": "1"
       }
     },
+    "openai": {
+      "version": "3.2.1",
+      "resolved": "https://registry.npmjs.org/openai/-/openai-3.2.1.tgz",
+      "integrity": "sha512-762C9BNlJPbjjlWZi4WYK9iM2tAVAv0uUp1UmI34vb0CN5T2mjB/qM6RYBmNKMh/dN9fC+bxqPwWJZUTWW052A==",
+      "dev": true,
+      "requires": {
+        "axios": "^0.26.0",
+        "form-data": "^4.0.0"
+      }
+    },
     "optionator": {
       "version": "0.9.1",
       "resolved": "https://registry.npmjs.org/optionator/-/optionator-0.9.1.tgz",
@@ -6295,6 +6721,15 @@
       "integrity": "sha512-LDJzPVEEEPR+y48z93A0Ed0yXb8pAByGWo/k5YYdYgpY2/2EsOsksJrq7lOHxryrVOn1ejG6oAp8ahvOIQD8sw==",
       "dev": true
     },
+    "path-to-regexp": {
+      "version": "1.8.0",
+      "resolved": "https://registry.npmjs.org/path-to-regexp/-/path-to-regexp-1.8.0.tgz",
+      "integrity": "sha512-n43JRhlUKUAlibEJhPeir1ncUID16QnEjNpwzNdO3Lm4ywrBpBZ5oLD0I6br9evr1Y9JTqwRtAh7JLoOzAQdVA==",
+      "dev": true,
+      "requires": {
+        "isarray": "0.0.1"
+      }
+    },
     "path-type": {
       "version": "4.0.0",
       "resolved": "https://registry.npmjs.org/path-type/-/path-type-4.0.0.tgz",
@@ -6484,6 +6919,37 @@
         "object-inspect": "^1.9.0"
       }
     },
+    "sinon": {
+      "version": "15.1.0",
+      "resolved": "https://registry.npmjs.org/sinon/-/sinon-15.1.0.tgz",
+      "integrity": "sha512-cS5FgpDdE9/zx7no8bxROHymSlPLZzq0ChbbLk1DrxBfc+eTeBK3y8nIL+nu/0QeYydhhbLIr7ecHJpywjQaoQ==",
+      "dev": true,
+      "requires": {
+        "@sinonjs/commons": "^3.0.0",
+        "@sinonjs/fake-timers": "^10.2.0",
+        "@sinonjs/samsam": "^8.0.0",
+        "diff": "^5.1.0",
+        "nise": "^5.1.4",
+        "supports-color": "^7.2.0"
+      },
+      "dependencies": {
+        "diff": {
+          "version": "5.1.0",
+          "resolved": "https://registry.npmjs.org/diff/-/diff-5.1.0.tgz",
+          "integrity": "sha512-D+mk+qE8VC/PAUrlAU34N+VfXev0ghe5ywmpqrawphmVZc1bEfn56uo9qpyGp1p4xpzOHkSW4ztBd6L7Xx4ACw==",
+          "dev": true
+        },
+        "supports-color": {
+          "version": "7.2.0",
+          "resolved": "https://registry.npmjs.org/supports-color/-/supports-color-7.2.0.tgz",
+          "integrity": "sha512-qpCAvRl9stuOHveKsn7HncJRvv501qIacKzQlO/+Lwxc9+0q2wLyv4Dfvt80/DPn2pqOBsJdDiogXGR9+OvwRw==",
+          "dev": true,
+          "requires": {
+            "has-flag": "^4.0.0"
+          }
+        }
+      }
+    },
     "slash": {
       "version": "3.0.0",
       "resolved": "https://registry.npmjs.org/slash/-/slash-3.0.0.tgz",

node/package.json

@@ -1,6 +1,6 @@
 {
   "name": "vectordb",
-  "version": "0.1.1",
+  "version": "0.1.3",
   "description": " Serverless, low-latency vector database for AI applications",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
@@ -27,6 +27,7 @@
     "@types/chai": "^4.3.4",
     "@types/mocha": "^10.0.1",
     "@types/node": "^18.16.2",
+    "@types/sinon": "^10.0.15",
     "@types/temp": "^0.9.1",
     "@typescript-eslint/eslint-plugin": "^5.59.1",
     "cargo-cp-artifact": "^0.1",
@@ -37,6 +38,8 @@
     "eslint-plugin-n": "^15.7.0",
     "eslint-plugin-promise": "^6.1.1",
     "mocha": "^10.2.0",
+    "sinon": "^15.1.0",
+    "openai": "^3.2.1",
     "temp": "^0.9.4",
     "ts-node": "^10.9.1",
     "ts-node-dev": "^2.0.0",

node/src/arrow.ts

@@ -15,15 +15,16 @@
 import {
   Field,
   Float32,
-  List,
+  List, type ListBuilder,
   makeBuilder,
   RecordBatchFileWriter,
   Table, Utf8,
   type Vector,
   vectorFromArray
 } from 'apache-arrow'
+import { type EmbeddingFunction } from './index'
 
-export function convertToTable (data: Array<Record<string, unknown>>): Table {
+export async function convertToTable<T> (data: Array<Record<string, unknown>>, embeddings?: EmbeddingFunction<T>): Promise<Table> {
   if (data.length === 0) {
     throw new Error('At least one record needs to be provided')
   }
@@ -33,11 +34,7 @@ export function convertToTable (data: Array<Record<string, unknown>>): Table {
 
   for (const columnsKey of columns) {
     if (columnsKey === 'vector') {
-      const children = new Field<Float32>('item', new Float32())
+      const listBuilder = newVectorListBuilder()
-      const list = new List(children)
-      const listBuilder = makeBuilder({
-        type: list
-      })
       const vectorSize = (data[0].vector as any[]).length
       for (const datum of data) {
         if ((datum[columnsKey] as any[]).length !== vectorSize) {
@@ -52,6 +49,14 @@ export function convertToTable (data: Array<Record<string, unknown>>): Table {
       for (const datum of data) {
         values.push(datum[columnsKey])
       }
+
+      if (columnsKey === embeddings?.sourceColumn) {
+        const vectors = await embeddings.embed(values as T[])
+        const listBuilder = newVectorListBuilder()
+        vectors.map(v => listBuilder.append(v))
+        records.vector = listBuilder.finish().toVector()
+      }
+
       if (typeof values[0] === 'string') {
         // `vectorFromArray` converts strings into dictionary vectors, forcing it back to a string column
         records[columnsKey] = vectorFromArray(values, new Utf8())
@@ -64,8 +69,17 @@ export function convertToTable (data: Array<Record<string, unknown>>): Table {
   return new Table(records)
 }
 
-export async function fromRecordsToBuffer (data: Array<Record<string, unknown>>): Promise<Buffer> {
+// Creates a new Arrow ListBuilder that stores a Vector column
-  const table = convertToTable(data)
+function newVectorListBuilder (): ListBuilder<Float32, any> {
+  const children = new Field<Float32>('item', new Float32())
+  const list = new List(children)
+  return makeBuilder({
+    type: list
+  })
+}
+
+export async function fromRecordsToBuffer<T> (data: Array<Record<string, unknown>>, embeddings?: EmbeddingFunction<T>): Promise<Buffer> {
+  const table = await convertToTable(data, embeddings)
   const writer = RecordBatchFileWriter.writeAll(table)
   return Buffer.from(await writer.toUint8Array())
 }

node/src/embedding/embedding_function.ts

@@ -0,0 +1,28 @@
+// Copyright 2023 Lance Developers.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+/**
+ * An embedding function that automatically creates vector representation for a given column.
+ */
+export interface EmbeddingFunction<T> {
+  /**
+     * The name of the column that will be used as input for the Embedding Function.
+     */
+  sourceColumn: string
+
+  /**
+     * Creates a vector representation for the given values.
+     */
+  embed: (data: T[]) => Promise<number[][]>
+}

node/src/embedding/openai.ts

@@ -0,0 +1,51 @@
+// Copyright 2023 Lance Developers.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+import { type EmbeddingFunction } from '../index'
+
+export class OpenAIEmbeddingFunction implements EmbeddingFunction<string> {
+  private readonly _openai: any
+  private readonly _modelName: string
+
+  constructor (sourceColumn: string, openAIKey: string, modelName: string = 'text-embedding-ada-002') {
+    let openai
+    try {
+      // eslint-disable-next-line @typescript-eslint/no-var-requires
+      openai = require('openai')
+    } catch {
+      throw new Error('please install openai using npm install openai')
+    }
+
+    this.sourceColumn = sourceColumn
+    const configuration = new openai.Configuration({
+      apiKey: openAIKey
+    })
+    this._openai = new openai.OpenAIApi(configuration)
+    this._modelName = modelName
+  }
+
+  async embed (data: string[]): Promise<number[][]> {
+    const response = await this._openai.createEmbedding({
+      model: this._modelName,
+      input: data
+    })
+    const embeddings: number[][] = []
+    for (let i = 0; i < response.data.data.length; i++) {
+      embeddings.push(response.data.data[i].embedding as number[])
+    }
+    return embeddings
+  }
+
+  sourceColumn: string
+}

node/src/index.ts

@@ -19,16 +19,21 @@ import {
   Vector
 } from 'apache-arrow'
 import { fromRecordsToBuffer } from './arrow'
+import type { EmbeddingFunction } from './embedding/embedding_function'
 
 // eslint-disable-next-line @typescript-eslint/no-var-requires
 const { databaseNew, databaseTableNames, databaseOpenTable, tableCreate, tableSearch, tableAdd, tableCreateVectorIndex } = require('../native.js')
 
+export type { EmbeddingFunction }
+export { OpenAIEmbeddingFunction } from './embedding/openai'
+
 /**
  * Connect to a LanceDB instance at the given URI
  * @param uri The uri of the database.
  */
 export async function connect (uri: string): Promise<Connection> {
-  return new Connection(uri)
+  const db = await databaseNew(uri)
+  return new Connection(db, uri)
 }
 
 /**
@@ -38,9 +43,9 @@ export class Connection {
   private readonly _uri: string
   private readonly _db: any
 
-  constructor (uri: string) {
+  constructor (db: any, uri: string) {
     this._uri = uri
-    this._db = databaseNew(uri)
+    this._db = db
   }
 
   get uri (): string {
@@ -56,16 +61,49 @@ export class Connection {
 
   /**
    * Open a table in the database.
+   *
    * @param name The name of the table.
    */
-  async openTable (name: string): Promise<Table> {
+  async openTable (name: string): Promise<Table>
+  /**
+   * Open a table in the database.
+   *
+   * @param name The name of the table.
+   * @param embeddings An embedding function to use on this Table
+   */
+  async openTable<T> (name: string, embeddings: EmbeddingFunction<T>): Promise<Table<T>>
+  async openTable<T> (name: string, embeddings?: EmbeddingFunction<T>): Promise<Table<T>> {
     const tbl = await databaseOpenTable.call(this._db, name)
+    if (embeddings !== undefined) {
+      return new Table(tbl, name, embeddings)
+    } else {
       return new Table(tbl, name)
     }
+  }
 
-  async createTable (name: string, data: Array<Record<string, unknown>>): Promise<Table> {
+  /**
-    await tableCreate.call(this._db, name, await fromRecordsToBuffer(data))
+   * Creates a new Table and initialize it with new data.
-    return await this.openTable(name)
+   *
+   * @param name The name of the table.
+   * @param data Non-empty Array of Records to be inserted into the Table
+   */
+
+  async createTable (name: string, data: Array<Record<string, unknown>>): Promise<Table>
+  /**
+   * Creates a new Table and initialize it with new data.
+   *
+   * @param name The name of the table.
+   * @param data Non-empty Array of Records to be inserted into the Table
+   * @param embeddings An embedding function to use on this Table
+   */
+  async createTable<T> (name: string, data: Array<Record<string, unknown>>, embeddings: EmbeddingFunction<T>): Promise<Table<T>>
+  async createTable<T> (name: string, data: Array<Record<string, unknown>>, embeddings?: EmbeddingFunction<T>): Promise<Table<T>> {
+    const tbl = await tableCreate.call(this._db, name, await fromRecordsToBuffer(data, embeddings))
+    if (embeddings !== undefined) {
+      return new Table(tbl, name, embeddings)
+    } else {
+      return new Table(tbl, name)
+    }
   }
 
   async createTableArrow (name: string, table: ArrowTable): Promise<Table> {
@@ -75,16 +113,22 @@ export class Connection {
   }
 }
 
-/**
+export class Table<T = number[]> {
- * A table in a LanceDB database.
- */
-export class Table {
   private readonly _tbl: any
   private readonly _name: string
+  private readonly _embeddings?: EmbeddingFunction<T>
 
-  constructor (tbl: any, name: string) {
+  constructor (tbl: any, name: string)
+  /**
+   * @param tbl
+   * @param name
+   * @param embeddings An embedding function to use when interacting with this table
+   */
+  constructor (tbl: any, name: string, embeddings: EmbeddingFunction<T>)
+  constructor (tbl: any, name: string, embeddings?: EmbeddingFunction<T>) {
     this._tbl = tbl
     this._name = name
+    this._embeddings = embeddings
   }
 
   get name (): string {
@@ -92,11 +136,11 @@ export class Table {
   }
 
   /**
-     * Create a search query to find the nearest neighbors of the given query vector.
+   * Creates a search query to find the nearest neighbors of the given search term
-     * @param queryVector The query vector.
+   * @param query The query search term
    */
-  search (queryVector: number[]): Query {
+  search (query: T): Query<T> {
-    return new Query(this._tbl, queryVector)
+    return new Query(this._tbl, query, this._embeddings)
   }
 
   /**
@@ -106,7 +150,7 @@ export class Table {
    * @return The number of rows added to the table
    */
   async add (data: Array<Record<string, unknown>>): Promise<number> {
-    return tableAdd.call(this._tbl, await fromRecordsToBuffer(data), WriteMode.Append.toString())
+    return tableAdd.call(this._tbl, await fromRecordsToBuffer(data, this._embeddings), WriteMode.Append.toString())
   }
 
   /**
@@ -116,9 +160,14 @@ export class Table {
    * @return The number of rows added to the table
    */
   async overwrite (data: Array<Record<string, unknown>>): Promise<number> {
-    return tableAdd.call(this._tbl, await fromRecordsToBuffer(data), WriteMode.Overwrite.toString())
+    return tableAdd.call(this._tbl, await fromRecordsToBuffer(data, this._embeddings), WriteMode.Overwrite.toString())
   }
 
+  /**
+   * Create an ANN index on this Table vector index.
+   *
+   * @param indexParams The parameters of this Index, @see VectorIndexParams.
+   */
   async create_index (indexParams: VectorIndexParams): Promise<any> {
     return tableCreateVectorIndex.call(this._tbl, indexParams)
   }
@@ -177,32 +226,35 @@ export type VectorIndexParams = IvfPQIndexConfig
 /**
  * A builder for nearest neighbor queries for LanceDB.
  */
-export class Query {
+export class Query<T = number[]> {
   private readonly _tbl: any
-  private readonly _queryVector: number[]
+  private readonly _query: T
+  private _queryVector?: number[]
   private _limit: number
   private _refineFactor?: number
   private _nprobes: number
   private readonly _columns?: string[]
   private _filter?: string
   private _metricType?: MetricType
+  private readonly _embeddings?: EmbeddingFunction<T>
 
-  constructor (tbl: any, queryVector: number[]) {
+  constructor (tbl: any, query: T, embeddings?: EmbeddingFunction<T>) {
     this._tbl = tbl
-    this._queryVector = queryVector
+    this._query = query
     this._limit = 10
     this._nprobes = 20
     this._refineFactor = undefined
     this._columns = undefined
     this._filter = undefined
     this._metricType = undefined
+    this._embeddings = embeddings
   }
 
   /***
    * Sets the number of results that will be returned
    * @param value number of results
    */
-  limit (value: number): Query {
+  limit (value: number): Query<T> {
     this._limit = value
     return this
   }
@@ -211,7 +263,7 @@ export class Query {
    * Refine the results by reading extra elements and re-ranking them in memory.
    * @param value refine factor to use in this query.
    */
-  refineFactor (value: number): Query {
+  refineFactor (value: number): Query<T> {
     this._refineFactor = value
     return this
   }
@@ -220,7 +272,7 @@ export class Query {
    * The number of probes used. A higher number makes search more accurate but also slower.
    * @param value The number of probes used.
    */
-  nprobes (value: number): Query {
+  nprobes (value: number): Query<T> {
     this._nprobes = value
     return this
   }
@@ -229,7 +281,7 @@ export class Query {
    * A filter statement to be applied to this query.
    * @param value A filter in the same format used by a sql WHERE clause.
    */
-  filter (value: string): Query {
+  filter (value: string): Query<T> {
     this._filter = value
     return this
   }
@@ -238,7 +290,7 @@ export class Query {
    * The MetricType used for this Query.
    * @param value The metric to the. @see MetricType for the different options
    */
-  metricType (value: MetricType): Query {
+  metricType (value: MetricType): Query<T> {
     this._metricType = value
     return this
   }
@@ -247,6 +299,12 @@ export class Query {
    * Execute the query and return the results as an Array of Objects
    */
   async execute<T = Record<string, unknown>> (): Promise<T[]> {
+    if (this._embeddings !== undefined) {
+      this._queryVector = (await this._embeddings.embed([this._query]))[0]
+    } else {
+      this._queryVector = this._query as number[]
+    }
+
     const buffer = await tableSearch.call(this._tbl, this)
     const data = tableFromIPC(buffer)
     return data.toArray().map((entry: Record<string, unknown>) => {

node/src/test/embedding/openai.ts

@@ -0,0 +1,50 @@
+// Copyright 2023 Lance Developers.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+import { describe } from 'mocha'
+import { assert } from 'chai'
+
+import { OpenAIEmbeddingFunction } from '../../embedding/openai'
+
+// eslint-disable-next-line @typescript-eslint/no-var-requires
+const { OpenAIApi } = require('openai')
+// eslint-disable-next-line @typescript-eslint/no-var-requires
+const { stub } = require('sinon')
+
+describe('OpenAPIEmbeddings', function () {
+  const stubValue = {
+    data: {
+      data: [
+        {
+          embedding: Array(1536).fill(1.0)
+        },
+        {
+          embedding: Array(1536).fill(2.0)
+        }
+      ]
+    }
+  }
+
+  describe('#embed', function () {
+    it('should create vector embeddings', async function () {
+      const openAIStub = stub(OpenAIApi.prototype, 'createEmbedding').returns(stubValue)
+      const f = new OpenAIEmbeddingFunction('text', 'sk-key')
+      const vectors = await f.embed(['abc', 'def'])
+      assert.isTrue(openAIStub.calledOnce)
+      assert.equal(vectors.length, 2)
+      assert.deepEqual(vectors[0], stubValue.data.data[0].embedding)
+      assert.deepEqual(vectors[1], stubValue.data.data[1].embedding)
+    })
+  })
+})

node/src/test/io.ts

@@ -0,0 +1,52 @@
+// Copyright 2023 Lance Developers.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+// IO tests
+
+import { describe } from 'mocha'
+import { assert } from 'chai'
+
+import * as lancedb from '../index'
+
+describe('LanceDB S3 client', function () {
+  if (process.env.TEST_S3_BASE_URL != null) {
+    const baseUri = process.env.TEST_S3_BASE_URL
+    it('should have a valid url', async function () {
+      const uri = `${baseUri}/valid_url`
+      const table = await createTestDB(uri, 2, 20)
+      const con = await lancedb.connect(uri)
+      assert.equal(con.uri, uri)
+
+      const results = await table.search([0.1, 0.3]).limit(5).execute()
+      assert.equal(results.length, 5)
+    })
+  } else {
+    describe.skip('Skip S3 test', function () {})
+  }
+})
+
+async function createTestDB (uri: string, numDimensions: number = 2, numRows: number = 2): Promise<lancedb.Table> {
+  const con = await lancedb.connect(uri)
+
+  const data = []
+  for (let i = 0; i < numRows; i++) {
+    const vector = []
+    for (let j = 0; j < numDimensions; j++) {
+      vector.push(i + (j * 0.1))
+    }
+    data.push({ id: i + 1, name: `name_${i}`, price: i + 10, is_active: (i % 2 === 0), vector })
+  }
+
+  return await con.createTable('vectors', data)
+}

node/src/test/test.ts

@@ -17,7 +17,7 @@ import { assert } from 'chai'
 import { track } from 'temp'
 
 import * as lancedb from '../index'
-import { MetricType, Query } from '../index'
+import { type EmbeddingFunction, MetricType, Query } from '../index'
 
 describe('LanceDB client', function () {
   describe('when creating a connection to lancedb', function () {
@@ -140,6 +140,39 @@ describe('LanceDB client', function () {
       await table.create_index({ type: 'ivf_pq', column: 'vector', num_partitions: 2, max_iters: 2 })
     }).timeout(10_000) // Timeout is high partially because GH macos runner is pretty slow
   })
+
+  describe('when using a custom embedding function', function () {
+    class TextEmbedding implements EmbeddingFunction<string> {
+      sourceColumn: string
+
+      constructor (targetColumn: string) {
+        this.sourceColumn = targetColumn
+      }
+
+      _embedding_map = new Map<string, number[]>([
+        ['foo', [2.1, 2.2]],
+        ['bar', [3.1, 3.2]]
+      ])
+
+      async embed (data: string[]): Promise<number[][]> {
+        return data.map(datum => this._embedding_map.get(datum) ?? [0.0, 0.0])
+      }
+    }
+
+    it('should encode the original data into embeddings', async function () {
+      const dir = await track().mkdir('lancejs')
+      const con = await lancedb.connect(dir)
+      const embeddings = new TextEmbedding('name')
+
+      const data = [
+        { price: 10, name: 'foo' },
+        { price: 50, name: 'bar' }
+      ]
+      const table = await con.createTable('vectors', data, embeddings)
+      const results = await table.search('foo').execute()
+      assert.equal(results.length, 2)
+    })
+  })
 })
 
 describe('Query object', function () {

notebooks/diffusiondb/datagen.py

@@ -0,0 +1,108 @@
+#!/usr/bin/env python
+#
+#  Copyright 2023 LanceDB Developers
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+
+"""Dataset hf://poloclub/diffusiondb
+"""
+
+import io
+from argparse import ArgumentParser
+from multiprocessing import Pool
+
+import lance
+import lancedb
+import pyarrow as pa
+from datasets import load_dataset
+from PIL import Image
+from transformers import CLIPModel, CLIPProcessor, CLIPTokenizerFast
+
+MODEL_ID = "openai/clip-vit-base-patch32"
+
+device = "cuda"
+
+tokenizer = CLIPTokenizerFast.from_pretrained(MODEL_ID)
+model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32").to(device)
+processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch32")
+
+schema = pa.schema(
+    [
+        pa.field("prompt", pa.string()),
+        pa.field("seed", pa.uint32()),
+        pa.field("step", pa.uint16()),
+        pa.field("cfg", pa.float32()),
+        pa.field("sampler", pa.string()),
+        pa.field("width", pa.uint16()),
+        pa.field("height", pa.uint16()),
+        pa.field("timestamp", pa.timestamp("s")),
+        pa.field("image_nsfw", pa.float32()),
+        pa.field("prompt_nsfw", pa.float32()),
+        pa.field("vector", pa.list_(pa.float32(), 512)),
+        pa.field("image", pa.binary()),
+    ]
+)
+
+
+def pil_to_bytes(img) -> list[bytes]:
+    buf = io.BytesIO()
+    img.save(buf, format="PNG")
+    return buf.getvalue()
+
+
+def generate_clip_embeddings(batch) -> pa.RecordBatch:
+    image = processor(text=None, images=batch["image"], return_tensors="pt")[
+        "pixel_values"
+    ].to(device)
+    img_emb = model.get_image_features(image)
+    batch["vector"] = img_emb.cpu().tolist()
+
+    with Pool() as p:
+        batch["image_bytes"] = p.map(pil_to_bytes, batch["image"])
+    return batch
+
+
+def datagen(args):
+    """Generate DiffusionDB dataset, and use CLIP model to generate image embeddings."""
+    dataset = load_dataset("poloclub/diffusiondb", args.subset)
+    data = []
+    for b in dataset.map(
+        generate_clip_embeddings, batched=True, batch_size=256, remove_columns=["image"]
+    )["train"]:
+        b["image"] = b["image_bytes"]
+        del b["image_bytes"]
+        data.append(b)
+    tbl = pa.Table.from_pylist(data, schema=schema)
+    return tbl
+
+
+def main():
+    parser = ArgumentParser()
+    parser.add_argument(
+        "-o", "--output", metavar="DIR", help="Output lance directory", required=True
+    )
+    parser.add_argument(
+        "-s",
+        "--subset",
+        choices=["2m_all", "2m_first_10k", "2m_first_100k"],
+        default="2m_first_10k",
+        help="subset of the hg dataset",
+    )
+
+    args = parser.parse_args()
+
+    batches = datagen(args)
+    lance.write_dataset(batches, args.output)
+
+
+if __name__ == "__main__":
+    main()

notebooks/diffusiondb/requirements.txt

@@ -0,0 +1,9 @@
+datasets
+Pillow
+lancedb
+isort
+black
+transformers
+--index-url https://download.pytorch.org/whl/cu118
+torch
+torchvision

notebooks/multimodal_search.ipynb

@@ -0,0 +1,240 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip available: \u001b[0m\u001b[31;49m22.3.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.1.2\u001b[0m\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
+      "\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip available: \u001b[0m\u001b[31;49m22.3.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.1.2\u001b[0m\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "!pip install --quiet -U lancedb\n",
+    "!pip install --quiet gradio transformers torch torchvision"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 60,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import io\n",
+    "import PIL\n",
+    "import duckdb\n",
+    "import lancedb"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## First run setup: Download data and pre-process"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "<lance.dataset.LanceDataset at 0x3045db590>"
+      ]
+     },
+     "execution_count": 30,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# remove null prompts\n",
+    "import lance\n",
+    "import pyarrow.compute as pc\n",
+    "\n",
+    "# download s3://eto-public/datasets/diffusiondb/small_10k.lance to this uri\n",
+    "data = lance.dataset(\"~/datasets/rawdata.lance\").to_table()\n",
+    "\n",
+    "# First data processing and full-text-search index\n",
+    "db = lancedb.connect(\"~/datasets/demo\")\n",
+    "tbl = db.create_table(\"diffusiondb\", data.filter(~pc.field(\"prompt\").is_null()))\n",
+    "tbl = tbl.create_fts_index([\"prompt\"])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Create / Open LanceDB Table"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "db = lancedb.connect(\"~/datasets/demo\")\n",
+    "tbl = db.open_table(\"diffusiondb\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Create CLIP embedding function for the text"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from transformers import CLIPModel, CLIPProcessor, CLIPTokenizerFast\n",
+    "\n",
+    "MODEL_ID = \"openai/clip-vit-base-patch32\"\n",
+    "\n",
+    "tokenizer = CLIPTokenizerFast.from_pretrained(MODEL_ID)\n",
+    "model = CLIPModel.from_pretrained(MODEL_ID)\n",
+    "processor = CLIPProcessor.from_pretrained(MODEL_ID)\n",
+    "\n",
+    "def embed_func(query):\n",
+    "    inputs = tokenizer([query], padding=True, return_tensors=\"pt\")\n",
+    "    text_features = model.get_text_features(**inputs)\n",
+    "    return text_features.detach().numpy()[0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Search functions for Gradio"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 64,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def find_image_vectors(query):\n",
+    "    emb = embed_func(query)\n",
+    "    return _extract(tbl.search(emb).limit(9).to_df())\n",
+    "\n",
+    "def find_image_keywords(query):\n",
+    "    return _extract(tbl.search(query).limit(9).to_df())\n",
+    "\n",
+    "def find_image_sql(query):\n",
+    "    diffusiondb = tbl.to_lance()\n",
+    "    return _extract(duckdb.query(query).to_df())\n",
+    "\n",
+    "def _extract(df):\n",
+    "    image_col = \"image\"\n",
+    "    return [(PIL.Image.open(io.BytesIO(row[image_col])), row[\"prompt\"]) for _, row in df.iterrows()]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Setup Gradio interface"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 65,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Running on local URL:  http://127.0.0.1:7867\n",
+      "\n",
+      "To create a public link, set `share=True` in `launch()`.\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<div><iframe src=\"http://127.0.0.1:7867/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": []
+     },
+     "execution_count": 65,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import gradio as gr\n",
+    "\n",
+    "\n",
+    "with gr.Blocks() as demo:\n",
+    "\n",
+    "    with gr.Row():\n",
+    "        with gr.Tab(\"Embeddings\"):\n",
+    "            vector_query = gr.Textbox(value=\"portraits of a person\", show_label=False)\n",
+    "            b1 = gr.Button(\"Submit\")\n",
+    "        with gr.Tab(\"Keywords\"):\n",
+    "            keyword_query = gr.Textbox(value=\"ninja turtle\", show_label=False)\n",
+    "            b2 = gr.Button(\"Submit\")\n",
+    "        with gr.Tab(\"SQL\"):\n",
+    "            sql_query = gr.Textbox(value=\"SELECT * from diffusiondb WHERE image_nsfw >= 2 LIMIT 9\", show_label=False)\n",
+    "            b3 = gr.Button(\"Submit\")\n",
+    "    with gr.Row():\n",
+    "        gallery = gr.Gallery(\n",
+    "                label=\"Found images\", show_label=False, elem_id=\"gallery\"\n",
+    "            ).style(columns=[3], rows=[3], object_fit=\"contain\", height=\"auto\")   \n",
+    "        \n",
+    "    b1.click(find_image_vectors, inputs=vector_query, outputs=gallery)\n",
+    "    b2.click(find_image_keywords, inputs=keyword_query, outputs=gallery)\n",
+    "    b3.click(find_image_sql, inputs=sql_query, outputs=gallery)\n",
+    "    \n",
+    "demo.launch()"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}

python/lancedb/db.py

@@ -13,13 +13,16 @@
 
 from __future__ import annotations
 
+import os
 from pathlib import Path
+import os
 
 import pyarrow as pa
+from pyarrow import fs
 
 from .common import DATA, URI
 from .table import LanceTable
-from .util import get_uri_scheme
+from .util import get_uri_scheme, get_uri_location
 
 
 class LanceDBConnection:
@@ -47,12 +50,21 @@ class LanceDBConnection:
         -------
         A list of table names.
         """
-        if get_uri_scheme(self.uri) == "file":
+        try:
-            return [p.stem for p in Path(self.uri).glob("*.lance")]
+            filesystem, path = fs.FileSystem.from_uri(self.uri)
+        except pa.ArrowInvalid:
             raise NotImplementedError(
-            "List table_names is only supported for local filesystem for now"
+                "Unsupported scheme: " + self.uri
             )
 
+        try:
+            paths = filesystem.get_file_info(fs.FileSelector(get_uri_location(self.uri)))
+        except FileNotFoundError:
+            # It is ok if the file does not exist since it will be created
+            paths = []
+        tables = [os.path.splitext(file_info.base_name)[0] for file_info in paths if file_info.extension == 'lance']
+        return tables
+
     def __len__(self) -> int:
         return len(self.table_names())
 
@@ -112,3 +124,15 @@ class LanceDBConnection:
         A LanceTable object representing the table.
         """
         return LanceTable(self, name)
+
+    def drop_table(self, name: str):
+        """Drop a table from the database.
+
+        Parameters
+        ----------
+        name: str
+            The name of the table.
+        """
+        filesystem, path = pa.fs.FileSystem.from_uri(self.uri)
+        table_path = os.path.join(path, name + ".lance")
+        filesystem.delete_dir(table_path)

python/lancedb/fts.py

@@ -16,7 +16,13 @@ import os
 from typing import List, Tuple
 
 import pyarrow as pa
+
+try:
     import tantivy
+except ImportError:
+    raise ImportError(
+        "Please install tantivy-py `pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985` to use the full text search feature."
+    )
 
 from .table import LanceTable
 

python/lancedb/query.py

@@ -153,7 +153,7 @@ class LanceFtsQueryBuilder(LanceQueryBuilder):
             import tantivy
         except ImportError:
             raise ImportError(
-                "You need to install the `lancedb[fts]` extra to use this method."
+                "Please install tantivy-py `pip install tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985` to use the full text search feature."
             )
 
         from .fts import search_index

python/lancedb/table.py

@@ -253,8 +253,7 @@ def _sanitize_vector_column(data: pa.Table, vector_column_name: str) -> pa.Table
     vector_column_name: str
         The name of the vector column.
     """
-    i = data.column_names.index(vector_column_name)
+    if vector_column_name not in data.column_names:
-    if i < 0:
         raise ValueError(f"Missing vector column: {vector_column_name}")
     vec_arr = data[vector_column_name].combine_chunks()
     if pa.types.is_fixed_size_list(vec_arr.type):
@@ -266,4 +265,4 @@ def _sanitize_vector_column(data: pa.Table, vector_column_name: str) -> pa.Table
         values = values.cast(pa.float32())
     list_size = len(values) / len(data)
     vec_arr = pa.FixedSizeListArray.from_arrays(values, list_size)
-    return data.set_column(i, vector_column_name, vec_arr)
+    return data.set_column(data.column_names.index(vector_column_name), vector_column_name, vec_arr)

python/lancedb/util.py

@@ -41,3 +41,23 @@ def get_uri_scheme(uri: str) -> str:
         # So we add special handling here for schemes that are a single character
         scheme = "file"
     return scheme
+
+
+def get_uri_location(uri: str) -> str:
+    """
+    Get the location of a URI. If the parameter is not a url, assumes it is just a path
+
+    Parameters
+    ----------
+    uri : str
+        The URI to parse.
+
+    Returns
+    -------
+    str: Location part of the URL, without scheme
+    """
+    parsed = urlparse(uri)
+    if not parsed.netloc:
+        return parsed.path
+    else:
+        return parsed.netloc + parsed.path

python/pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "lancedb"
-version = "0.1.2"
+version = "0.1.5"
-dependencies = ["pylance>=0.4.6", "ratelimiter", "retry", "tqdm"]
+dependencies = ["pylance>=0.4.17", "ratelimiter", "retry", "tqdm"]
 description = "lancedb"
 authors = [
     { name = "LanceDB Devs", email = "dev@lancedb.com" },
@@ -45,10 +45,6 @@ dev = [
 docs = [
     "mkdocs", "mkdocs-jupyter", "mkdocs-material", "mkdocstrings[python]"
 ]
-fts = [
-    # tantivy 0.19.2
-    "tantivy@git+https://github.com/quickwit-oss/tantivy-py#164adc87e1a033117001cf70e38c82a53014d985"
-]
 
 [build-system]
 requires = [

python/tests/test_db.py

@@ -97,3 +97,26 @@ def test_create_mode(tmp_path):
     )
     tbl = db.create_table("test", data=new_data, mode="overwrite")
     assert tbl.to_pandas().item.tolist() == ["fizz", "buzz"]
+
+
+def test_delete_table(tmp_path):
+    db = lancedb.connect(tmp_path)
+    data = pd.DataFrame(
+        {
+            "vector": [[3.1, 4.1], [5.9, 26.5]],
+            "item": ["foo", "bar"],
+            "price": [10.0, 20.0],
+        }
+    )
+    db.create_table("test", data=data)
+
+    with pytest.raises(Exception):
+        db.create_table("test", data=data)
+
+    assert db.table_names() == ["test"]
+
+    db.drop_table("test")
+    assert db.table_names() == []
+
+    db.create_table("test", data=data)
+    assert db.table_names() == ["test"]

python/tests/test_embeddings.py

@@ -14,7 +14,6 @@ import sys
 
 import numpy as np
 import pyarrow as pa
-
 from lancedb.embeddings import with_embeddings
 
 

python/tests/test_fts.py

@@ -13,13 +13,13 @@
 import os
 import random
 
+import lancedb.fts
 import numpy as np
 import pandas as pd
 import pytest
 import tantivy
 
 import lancedb as ldb
-import lancedb.fts
 
 
 @pytest.fixture

python/tests/test_io.py

@@ -0,0 +1,49 @@
+#  Copyright 2023 LanceDB Developers
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
+
+import os
+import pytest
+
+import lancedb
+
+# You need to setup AWS credentials an a base path to run this test. Example
+#    AWS_PROFILE=default TEST_S3_BASE_URL=s3://my_bucket/dataset pytest tests/test_io.py
+
+@pytest.mark.skipif(
+    (os.environ.get("TEST_S3_BASE_URL") is None),
+    reason="please setup s3 base url",
+)
+def test_s3_io():
+    db = lancedb.connect(os.environ.get("TEST_S3_BASE_URL"))
+    assert db.table_names() == []
+
+    table = db.create_table(
+        "test",
+        data=[
+            {"vector": [3.1, 4.1], "item": "foo", "price": 10.0},
+            {"vector": [5.9, 26.5], "item": "bar", "price": 20.0},
+        ],
+    )
+    rs = table.search([100, 100]).limit(1).to_df()
+    assert len(rs) == 1
+    assert rs["item"].iloc[0] == "bar"
+
+    rs = table.search([100, 100]).where("price < 15").limit(2).to_df()
+    assert len(rs) == 1
+    assert rs["item"].iloc[0] == "foo"
+
+    assert db.table_names() == ["test"]
+    assert "test" in db
+    assert len(db) == 1
+
+    assert db.open_table("test").name == db["test"].name

python/tests/test_query.py

@@ -17,7 +17,6 @@ import pandas as pd
 import pandas.testing as tm
 import pyarrow as pa
 import pytest
-
 from lancedb.query import LanceQueryBuilder
 
 

python/tests/test_table.py

@@ -16,7 +16,6 @@ from pathlib import Path
 import pandas as pd
 import pyarrow as pa
 import pytest
-
 from lancedb.table import LanceTable
 
 

rust/ffi/node/Cargo.toml

@@ -15,7 +15,7 @@ arrow-ipc = "37.0"
 arrow-schema = "37.0"
 once_cell = "1"
 futures = "0.3"
-lance = "0.4.3"
+lance = "0.4.17"
 vectordb = { path = "../../vectordb" }
 tokio = { version = "1.23", features = ["rt-multi-thread"] }
 neon = {version = "0.10.1", default-features = false, features = ["channel-api", "napi-6", "promise-api", "task-api"] }

rust/ffi/node/src/index/vector.rs

@@ -39,7 +39,7 @@ pub(crate) fn table_create_vector_index(mut cx: FunctionContext) -> JsResult<JsP
         let add_result = table
             .lock()
             .unwrap()
-            .create_idx(&index_params_builder)
+            .create_index(&index_params_builder)
             .await;
 
         deferred.settle_with(&channel, move |mut cx| {

rust/ffi/node/src/lib.rs

@@ -56,23 +56,46 @@ fn runtime<'a, C: Context<'a>>(cx: &mut C) -> NeonResult<&'static Runtime> {
     RUNTIME.get_or_try_init(|| Runtime::new().or_else(|err| cx.throw_error(err.to_string())))
 }
 
-fn database_new(mut cx: FunctionContext) -> JsResult<JsBox<JsDatabase>> {
+fn database_new(mut cx: FunctionContext) -> JsResult<JsPromise> {
     let path = cx.argument::<JsString>(0)?.value(&mut cx);
+
+    let rt = runtime(&mut cx)?;
+    let channel = cx.channel();
+    let (deferred, promise) = cx.promise();
+
+    rt.spawn(async move {
+        let database = Database::connect(&path).await;
+
+        deferred.settle_with(&channel, move |mut cx| {
             let db = JsDatabase {
-        database: Arc::new(Database::connect(path).or_else(|err| cx.throw_error(err.to_string()))?),
+                database: Arc::new(database.or_else(|err| cx.throw_error(err.to_string()))?),
             };
             Ok(cx.boxed(db))
+        });
+    });
+    Ok(promise)
 }
 
-fn database_table_names(mut cx: FunctionContext) -> JsResult<JsArray> {
+fn database_table_names(mut cx: FunctionContext) -> JsResult<JsPromise> {
     let db = cx
         .this()
         .downcast_or_throw::<JsBox<JsDatabase>, _>(&mut cx)?;
-    let tables = db
+
-        .database
+    let rt = runtime(&mut cx)?;
-        .table_names()
+    let (deferred, promise) = cx.promise();
-        .or_else(|err| cx.throw_error(err.to_string()))?;
+    let channel = cx.channel();
-    convert::vec_str_to_array(&tables, &mut cx)
+    let database = db.database.clone();
+
+    rt.spawn(async move {
+        let tables_rst = database.table_names().await;
+
+        deferred.settle_with(&channel, move |mut cx| {
+            let tables = tables_rst.or_else(|err| cx.throw_error(err.to_string()))?;
+            let table_names = convert::vec_str_to_array(&tables, &mut cx);
+            table_names
+        });
+    });
+    Ok(promise)
 }
 
 fn database_open_table(mut cx: FunctionContext) -> JsResult<JsPromise> {
@@ -87,7 +110,7 @@ fn database_open_table(mut cx: FunctionContext) -> JsResult<JsPromise> {
 
     let (deferred, promise) = cx.promise();
     rt.spawn(async move {
-        let table_rst = database.open_table(table_name).await;
+        let table_rst = database.open_table(&table_name).await;
 
         deferred.settle_with(&channel, move |mut cx| {
             let table = Arc::new(Mutex::new(
@@ -186,7 +209,7 @@ fn table_create(mut cx: FunctionContext) -> JsResult<JsPromise> {
 
     rt.block_on(async move {
         let batch_reader: Box<dyn RecordBatchReader> = Box::new(RecordBatchBuffer::new(batches));
-        let table_rst = database.create_table(table_name, batch_reader).await;
+        let table_rst = database.create_table(&table_name, batch_reader).await;
 
         deferred.settle_with(&channel, move |mut cx| {
             let table = Arc::new(Mutex::new(

rust/vectordb/Cargo.toml

@@ -12,7 +12,9 @@ repository = "https://github.com/lancedb/lancedb"
 arrow-array = "37.0"
 arrow-data = "37.0"
 arrow-schema = "37.0"
-lance = "0.4.3"
+object_store = "0.5.6"
+snafu = "0.7.4"
+lance = "0.4.17"
 tokio = { version = "1.23", features = ["rt-multi-thread"] }
 
 [dev-dependencies]

rust/vectordb/src/database.rs

@@ -12,16 +12,20 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use arrow_array::RecordBatchReader;
 use std::fs::create_dir_all;
-use std::path::{Path, PathBuf};
+use std::path::Path;
-use std::sync::Arc;
 
-use crate::error::Result;
+use arrow_array::RecordBatchReader;
+use lance::io::object_store::ObjectStore;
+use snafu::prelude::*;
+
+use crate::error::{CreateDirSnafu, Result};
 use crate::table::Table;
 
 pub struct Database {
-    pub(crate) path: Arc<PathBuf>,
+    object_store: ObjectStore,
+
+    pub(crate) uri: String,
 }
 
 const LANCE_EXTENSION: &str = "lance";
@@ -37,26 +41,38 @@ impl Database {
     /// # Returns
     ///
     /// * A [Database] object.
-    pub fn connect<P: AsRef<Path>>(path: P) -> Result<Database> {
+    pub async fn connect(uri: &str) -> Result<Database> {
-        if !path.as_ref().try_exists()? {
+        let object_store = ObjectStore::new(uri).await?;
-            create_dir_all(&path)?;
+        if object_store.is_local() {
+            Self::try_create_dir(uri).context(CreateDirSnafu { path: uri })?;
         }
         Ok(Database {
-            path: Arc::new(path.as_ref().to_path_buf()),
+            uri: uri.to_string(),
+            object_store,
         })
     }
 
+    /// Try to create a local directory to store the lancedb dataset
+    fn try_create_dir(path: &str) -> core::result::Result<(), std::io::Error> {
+        let path = Path::new(path);
+        if !path.try_exists()? {
+            create_dir_all(&path)?;
+        }
+        Ok(())
+    }
+
     /// Get the names of all tables in the database.
     ///
     /// # Returns
     ///
     /// * A [Vec<String>] with all table names.
-    pub fn table_names(&self) -> Result<Vec<String>> {
+    pub async fn table_names(&self) -> Result<Vec<String>> {
         let f = self
-            .path
+            .object_store
-            .read_dir()?
+            .read_dir("/")
-            .flatten()
+            .await?
-            .map(|dir_entry| dir_entry.path())
+            .iter()
+            .map(|fname| Path::new(fname))
             .filter(|path| {
                 let is_lance = path
                     .extension()
@@ -76,10 +92,10 @@ impl Database {
 
     pub async fn create_table(
         &self,
-        name: String,
+        name: &str,
         batches: Box<dyn RecordBatchReader>,
     ) -> Result<Table> {
-        Table::create(self.path.clone(), name, batches).await
+        Table::create(&self.uri, name, batches).await
     }
 
     /// Open a table in the database.
@@ -90,8 +106,8 @@ impl Database {
     /// # Returns
     ///
     /// * A [Table] object.
-    pub async fn open_table(&self, name: String) -> Result<Table> {
+    pub async fn open_table(&self, name: &str) -> Result<Table> {
-        Table::open(self.path.clone(), name).await
+        Table::open(&self.uri, name).await
     }
 }
 
@@ -105,10 +121,10 @@ mod tests {
     #[tokio::test]
     async fn test_connect() {
         let tmp_dir = tempdir().unwrap();
-        let path_buf = tmp_dir.into_path();
+        let uri = tmp_dir.path().to_str().unwrap();
-        let db = Database::connect(&path_buf);
+        let db = Database::connect(uri).await.unwrap();
 
-        assert_eq!(db.unwrap().path.as_path(), path_buf.as_path())
+        assert_eq!(db.uri, uri);
     }
 
     #[tokio::test]
@@ -118,10 +134,16 @@ mod tests {
         create_dir_all(tmp_dir.path().join("table2.lance")).unwrap();
         create_dir_all(tmp_dir.path().join("invalidlance")).unwrap();
 
-        let db = Database::connect(&tmp_dir.into_path()).unwrap();
+        let uri = tmp_dir.path().to_str().unwrap();
-        let tables = db.table_names().unwrap();
+        let db = Database::connect(uri).await.unwrap();
+        let tables = db.table_names().await.unwrap();
         assert_eq!(tables.len(), 2);
         assert!(tables.contains(&String::from("table1")));
         assert!(tables.contains(&String::from("table2")));
     }
+
+    #[tokio::test]
+    async fn test_connect_s3() {
+        // let db = Database::connect("s3://bucket/path/to/database").await.unwrap();
+    }
 }

rust/vectordb/src/error.rs

@@ -12,32 +12,50 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-#[derive(Debug)]
+use snafu::Snafu;
-pub enum Error {
-    IO(String),
-    Lance(String),
-}
 
-impl std::fmt::Display for Error {
+#[derive(Debug, Snafu)]
-    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+#[snafu(visibility(pub(crate)))]
-        let (catalog, message) = match self {
+pub enum Error {
-            Self::IO(s) => ("I/O", s.as_str()),
+    #[snafu(display("LanceDBError: Invalid table name: {name}"))]
-            Self::Lance(s) => ("Lance", s.as_str()),
+    InvalidTableName { name: String },
-        };
+    #[snafu(display("LanceDBError: Table '{name}' was not found"))]
-        write!(f, "LanceDBError({catalog}): {message}")
+    TableNotFound { name: String },
-    }
+    #[snafu(display("LanceDBError: Table '{name}' already exists"))]
+    TableAlreadyExists { name: String },
+    #[snafu(display("LanceDBError: Unable to created lance dataset at {path}: {source}"))]
+    CreateDir {
+        path: String,
+        source: std::io::Error,
+    },
+    #[snafu(display("LanceDBError: {message}"))]
+    Store { message: String },
+    #[snafu(display("LanceDBError: {message}"))]
+    Lance { message: String },
 }
 
 pub type Result<T> = std::result::Result<T, Error>;
 
-impl From<std::io::Error> for Error {
+impl From<lance::Error> for Error {
-    fn from(e: std::io::Error) -> Self {
+    fn from(e: lance::Error) -> Self {
-        Self::IO(e.to_string())
+        Self::Lance {
+            message: e.to_string(),
+        }
     }
 }
 
-impl From<lance::Error> for Error {
+impl From<object_store::Error> for Error {
-    fn from(e: lance::Error) -> Self {
+    fn from(e: object_store::Error) -> Self {
-        Self::Lance(e.to_string())
+        Self::Store {
+            message: e.to_string(),
+        }
+    }
+}
+
+impl From<object_store::path::Error> for Error {
+    fn from(e: object_store::path::Error) -> Self {
+        Self::Store {
+            message: e.to_string(),
+        }
     }
 }

rust/vectordb/src/table.rs

@@ -12,28 +12,35 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use std::path::PathBuf;
+use std::path::Path;
 use std::sync::Arc;
 
 use arrow_array::{Float32Array, RecordBatchReader};
 use lance::dataset::{Dataset, WriteMode, WriteParams};
 use lance::index::IndexType;
+use snafu::prelude::*;
 
-use crate::error::{Error, Result};
+use crate::error::{Error, InvalidTableNameSnafu, Result};
 use crate::index::vector::VectorIndexBuilder;
 use crate::query::Query;
 
 pub const VECTOR_COLUMN_NAME: &str = "vector";
-
 pub const LANCE_FILE_EXTENSION: &str = "lance";
 
 /// A table in a LanceDB database.
+#[derive(Debug)]
 pub struct Table {
     name: String,
-    path: String,
+    uri: String,
     dataset: Arc<Dataset>,
 }
 
+impl std::fmt::Display for Table {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        write!(f, "Table({})", self.name)
+    }
+}
+
 impl Table {
     /// Opens an existing Table
     ///
@@ -45,18 +52,28 @@ impl Table {
     /// # Returns
     ///
     /// * A [Table] object.
-    pub async fn open(base_path: Arc<PathBuf>, name: String) -> Result<Self> {
+    pub async fn open(base_uri: &str, name: &str) -> Result<Self> {
-        let ds_path = base_path.join(format!("{}.{}", name, LANCE_FILE_EXTENSION));
+        let path = Path::new(base_uri);
-        let ds_uri = ds_path
+
+        let table_uri = path.join(format!("{}.{}", name, LANCE_FILE_EXTENSION));
+        let uri = table_uri
+            .as_path()
             .to_str()
-            .ok_or(Error::IO(format!("Unable to find table {}", name)))?;
+            .context(InvalidTableNameSnafu { name })?;
-        let dataset = Dataset::open(ds_uri).await?;
+
-        let table = Table {
+        let dataset = Dataset::open(&uri).await.map_err(|e| match e {
-            name,
+            lance::Error::DatasetNotFound { .. } => Error::TableNotFound {
-            path: ds_uri.to_string(),
+                name: name.to_string(),
+            },
+            e => Error::Lance {
+                message: e.to_string(),
+            },
+        })?;
+        Ok(Table {
+            name: name.to_string(),
+            uri: uri.to_string(),
             dataset: Arc::new(dataset),
-        };
+        })
-        Ok(table)
     }
 
     /// Creates a new Table
@@ -71,25 +88,36 @@ impl Table {
     ///
     /// * A [Table] object.
     pub async fn create(
-        base_path: Arc<PathBuf>,
+        base_uri: &str,
-        name: String,
+        name: &str,
         mut batches: Box<dyn RecordBatchReader>,
     ) -> Result<Self> {
-        let ds_path = base_path.join(format!("{}.{}", name, LANCE_FILE_EXTENSION));
+        let base_path = Path::new(base_uri);
-        let path = ds_path
+        let table_uri = base_path.join(format!("{}.{}", name, LANCE_FILE_EXTENSION));
+        let uri = table_uri
+            .as_path()
             .to_str()
-            .ok_or(Error::IO(format!("Unable to find table {}", name)))?;
+            .context(InvalidTableNameSnafu { name })?
-
+            .to_string();
-        let dataset =
+        let dataset = Dataset::write(&mut batches, &uri, Some(WriteParams::default()))
-            Arc::new(Dataset::write(&mut batches, path, Some(WriteParams::default())).await?);
+            .await
+            .map_err(|e| match e {
+                lance::Error::DatasetAlreadyExists { .. } => Error::TableAlreadyExists {
+                    name: name.to_string(),
+                },
+                e => Error::Lance {
+                    message: e.to_string(),
+                },
+            })?;
         Ok(Table {
-            name,
+            name: name.to_string(),
-            path: path.to_string(),
+            uri,
-            dataset,
+            dataset: Arc::new(dataset),
         })
     }
 
-    pub async fn create_idx(&mut self, index_builder: &impl VectorIndexBuilder) -> Result<()> {
+    /// Create index on the table.
+    pub async fn create_index(&mut self, index_builder: &impl VectorIndexBuilder) -> Result<()> {
         use lance::index::DatasetIndexExt;
 
         let dataset = self
@@ -125,8 +153,7 @@ impl Table {
         let mut params = WriteParams::default();
         params.mode = write_mode.unwrap_or(WriteMode::Append);
 
-        self.dataset =
+        self.dataset = Arc::new(Dataset::write(&mut batches, &self.uri, Some(params)).await?);
-            Arc::new(Dataset::write(&mut batches, self.path.as_str(), Some(params)).await?);
         Ok(batches.count())
     }
 
@@ -151,6 +178,8 @@ impl Table {
 
 #[cfg(test)]
 mod tests {
+    use std::sync::Arc;
+
     use arrow_array::{
         Array, FixedSizeListArray, Float32Array, Int32Array, RecordBatch, RecordBatchReader,
     };
@@ -161,53 +190,68 @@ mod tests {
     use lance::index::vector::ivf::IvfBuildParams;
     use lance::index::vector::pq::PQBuildParams;
     use rand::Rng;
-    use std::sync::Arc;
     use tempfile::tempdir;
 
-    use crate::error::Result;
+    use super::*;
     use crate::index::vector::IvfPQIndexBuilder;
-    use crate::table::Table;
-
-    #[tokio::test]
-    async fn test_new_table_not_exists() {
-        let tmp_dir = tempdir().unwrap();
-        let path_buf = tmp_dir.into_path();
-
-        let table = Table::open(Arc::new(path_buf), "test".to_string()).await;
-        assert!(table.is_err());
-    }
 
     #[tokio::test]
     async fn test_open() {
         let tmp_dir = tempdir().unwrap();
-        let path_buf = tmp_dir.into_path();
+        let dataset_path = tmp_dir.path().join("test.lance");
+        let uri = tmp_dir.path().to_str().unwrap();
 
         let mut batches: Box<dyn RecordBatchReader> = Box::new(make_test_batches());
-        Dataset::write(
+        Dataset::write(&mut batches, dataset_path.to_str().unwrap(), None)
-            &mut batches,
-            path_buf.join("test.lance").to_str().unwrap(),
-            None,
-        )
             .await
             .unwrap();
 
-        let table = Table::open(Arc::new(path_buf), "test".to_string())
+        let table = Table::open(uri, "test").await.unwrap();
-            .await
-            .unwrap();
 
         assert_eq!(table.name, "test")
     }
 
     #[tokio::test]
-    async fn test_add() {
+    async fn test_open_not_found() {
         let tmp_dir = tempdir().unwrap();
-        let path_buf = tmp_dir.into_path();
+        let uri = tmp_dir.path().to_str().unwrap();
+        let table = Table::open(uri, "test").await;
+        assert!(matches!(table.unwrap_err(), Error::TableNotFound { .. }));
+    }
+
+    #[test]
+    fn test_object_store_path() {
+        use std::path::Path as StdPath;
+        let p = StdPath::new("s3://bucket/path/to/file");
+        let c = p.join("subfile");
+        assert_eq!(c.to_str().unwrap(), "s3://bucket/path/to/file/subfile");
+    }
+
+    #[tokio::test]
+    async fn test_create_already_exists() {
+        let tmp_dir = tempdir().unwrap();
+        let uri = tmp_dir.path().to_str().unwrap();
 
         let batches: Box<dyn RecordBatchReader> = Box::new(make_test_batches());
         let schema = batches.schema().clone();
-        let mut table = Table::create(Arc::new(path_buf), "test".to_string(), batches)
+        Table::create(&uri, "test", batches).await.unwrap();
-            .await
+
-            .unwrap();
+        let batches: Box<dyn RecordBatchReader> = Box::new(make_test_batches());
+        let result = Table::create(&uri, "test", batches).await;
+        assert!(matches!(
+            result.unwrap_err(),
+            Error::TableAlreadyExists { .. }
+        ));
+    }
+
+    #[tokio::test]
+    async fn test_add() {
+        let tmp_dir = tempdir().unwrap();
+        let uri = tmp_dir.path().to_str().unwrap();
+
+        let batches: Box<dyn RecordBatchReader> = Box::new(make_test_batches());
+        let schema = batches.schema().clone();
+        let mut table = Table::create(&uri, "test", batches).await.unwrap();
         assert_eq!(table.count_rows().await.unwrap(), 10);
 
         let new_batches: Box<dyn RecordBatchReader> =
@@ -225,13 +269,11 @@ mod tests {
     #[tokio::test]
     async fn test_add_overwrite() {
         let tmp_dir = tempdir().unwrap();
-        let path_buf = tmp_dir.into_path();
+        let uri = tmp_dir.path().to_str().unwrap();
 
         let batches: Box<dyn RecordBatchReader> = Box::new(make_test_batches());
         let schema = batches.schema().clone();
-        let mut table = Table::create(Arc::new(path_buf), "test".to_string(), batches)
+        let mut table = Table::create(uri, "test", batches).await.unwrap();
-            .await
-            .unwrap();
         assert_eq!(table.count_rows().await.unwrap(), 10);
 
         let new_batches: Box<dyn RecordBatchReader> =
@@ -252,20 +294,15 @@ mod tests {
     #[tokio::test]
     async fn test_search() {
         let tmp_dir = tempdir().unwrap();
-        let path_buf = tmp_dir.into_path();
+        let dataset_path = tmp_dir.path().join("test.lance");
+        let uri = tmp_dir.path().to_str().unwrap();
 
         let mut batches: Box<dyn RecordBatchReader> = Box::new(make_test_batches());
-        Dataset::write(
+        Dataset::write(&mut batches, dataset_path.to_str().unwrap(), None)
-            &mut batches,
-            path_buf.join("test.lance").to_str().unwrap(),
-            None,
-        )
             .await
             .unwrap();
 
-        let table = Table::open(Arc::new(path_buf), "test".to_string())
+        let table = Table::open(uri, "test").await.unwrap();
-            .await
-            .unwrap();
 
         let vector = Float32Array::from_iter_values([0.1, 0.2]);
         let query = table.search(vector.clone());
@@ -291,7 +328,7 @@ mod tests {
         use arrow_array::Float32Array;
 
         let tmp_dir = tempdir().unwrap();
-        let path_buf = tmp_dir.into_path();
+        let uri = tmp_dir.path().to_str().unwrap();
 
         let dimension = 16;
         let schema = Arc::new(ArrowSchema::new(vec![Field::new(
@@ -318,9 +355,7 @@ mod tests {
         .unwrap()]);
 
         let reader: Box<dyn RecordBatchReader + Send> = Box::new(batches);
-        let mut table = Table::create(Arc::new(path_buf), "test".to_string(), reader)
+        let mut table = Table::create(uri, "test", reader).await.unwrap();
-            .await
-            .unwrap();
 
         let mut i = IvfPQIndexBuilder::new();
 
@@ -330,7 +365,7 @@ mod tests {
             .ivf_params(IvfBuildParams::new(256))
             .pq_params(PQBuildParams::default());
 
-        table.create_idx(index_builder).await.unwrap();
+        table.create_index(index_builder).await.unwrap();
 
         assert_eq!(table.dataset.load_indices().await.unwrap().len(), 1);
         assert_eq!(table.count_rows().await.unwrap(), 512);