Consolidate the Rust and Sqllogictest tests

habedi · habedi · commit 72d3b68d4453 · 2025-09-30T22:45:42.000+02:00
diff --git a/.github/workflows/dist_pipeline.yml b/.github/workflows/dist_pipeline.yml
@@ -1,7 +1,9 @@
-name: Build Extensions
+name: Build Extension Binaries
 on:
-  pull_request:
   workflow_dispatch:
+  push:
+    tags:
+      - 'v*'
 
 permissions:
   contents: read
@@ -11,7 +13,7 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  duckdb-next-build:
+  duckdb-next-stable-build:
     uses: duckdb/extension-ci-tools/.github/workflows/_extension_distribution.yml@main
     with:
       duckdb_version: main
diff --git a/.github/workflows/lints.yml b/.github/workflows/lints.yml
@@ -1,4 +1,4 @@
-name: Run Linters
+name: Run Rust Linters
 
 on:
   workflow_dispatch:
@@ -10,7 +10,7 @@ permissions:
   contents: read
 
 jobs:
-  lint:
+  lints:
     runs-on: ubuntu-latest
 
     steps:
diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
@@ -5,14 +5,19 @@ on:
   pull_request:
     branches:
       - main
+  push:
+    branches:
+      - main
 
 permissions:
   contents: read
 
 jobs:
-  test:
+  rust-tests:
+    name: Rust Unit & Coverage
     runs-on: ubuntu-latest
-
+    env:
+      CARGO_TERM_COLOR: always
     steps:
       - name: Checkout Code
         uses: actions/checkout@v4
@@ -22,17 +27,55 @@ jobs:
       - name: Set up Rust
         uses: actions-rust-lang/setup-rust-toolchain@v1
 
-      - name: Install Dependencies
+      - name: Cache Cargo
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cargo/registry
+            ~/.cargo/git
+            infera/target
+          key: ${{ runner.os }}-cargo-${{ hashFiles('**/Cargo.lock') }}
+          restore-keys: |
+            ${{ runner.os }}-cargo-
+
+      - name: Install System Dependencies
         run: |
-          sudo apt-get update && sudo apt-get upgrade -y
+          sudo apt-get update
           sudo apt-get install -y gcc curl pkg-config libssl-dev make
-          cargo install cargo-tarpaulin
+          cargo install cargo-tarpaulin --locked || true
+
+      - name: Run Rust Tests
+        run: |
+          make rust-test
 
-      - name: Run Tests and Generate Coverage Report
-        run: make rust-coverage # Generates a coverage report for Infera crate
+      - name: Generate Coverage (Tarpaulin)
+        run: |
+          make rust-coverage
 
       - name: Upload Coverage Reports to Codecov
         uses: codecov/codecov-action@v5
         with:
           token: ${{ secrets.CODECOV_TOKEN }}
         continue-on-error: false
+
+  sqllogictest:
+    name: SQLLogic (DuckDB Extension)
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Code
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - name: Set up Rust
+        uses: actions-rust-lang/setup-rust-toolchain@v1
+
+      - name: Install System Dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y gcc curl pkg-config libssl-dev make
+
+      - name: Build Extension & Run SQL Tests
+        run: |
+          make release
+          make test
diff --git a/README.md b/README.md
@@ -7,7 +7,6 @@
 <h2>Infera</h2>
 
 [![Tests](https://img.shields.io/github/actions/workflow/status/CogitatorTech/infera/tests.yml?label=tests&style=flat&labelColor=282c34&logo=github)](https://github.com/CogitatorTech/infera/actions/workflows/tests.yml)
-[![Code Coverage](https://img.shields.io/codecov/c/github/CogitatorTech/infera?label=coverage&style=flat&labelColor=282c34&logo=codecov)](https://codecov.io/gh/CogitatorTech/infera)
 [![Code Quality](https://img.shields.io/codefactor/grade/github/CogitatorTech/infera?label=quality&style=flat&labelColor=282c34&logo=codefactor)](https://www.codefactor.io/repository/github/CogitatorTech/infera)
 [![Examples](https://img.shields.io/badge/examples-view-green?style=flat&labelColor=282c34&logo=github)](https://github.com/CogitatorTech/infera/tree/main/docs/examples)
 [![Docs](https://img.shields.io/badge/docs-view-blue?style=flat&labelColor=282c34&logo=read-the-docs)](https://github.com/CogitatorTech/infera/tree/main/docs)
@@ -23,7 +22,7 @@ Infera is DuckDB extension that allows you use machine learning (ML) models dire
 on data stored in DuckDB tables.
 It is developed in Rust and uses [Tract](https://github.com/snipsco/tract) as the backend inference engine.
 Infera supports loading and running models in [ONNX](https://onnx.ai/) format.
-Check out the [ONNX Model Zoo](https://huggingface.co/onnxmodelzoo) repositors on Hugging Face for a very large
+Check out the [ONNX Model Zoo](https://huggingface.co/onnxmodelzoo) repositors on Hugging Face for a large
 collection of ready-to-use models that can be used with Infera.
 
 ### Motivation
@@ -41,7 +40,7 @@ It simplifies the workflow and speeds up the process for users, and eliminates t
 - Adds ML inference as first-class citizens in SQL queries.
 - Supports loading and using local as well as remote models.
 - Supports using ML models in ONNX format with a simple and flexible API.
-- Supports performing inference on table columns or raw `BLOB` (tensor) data.
+- Supports performing inference on table columns or raw BLOB (tensor) data.
 - Supports both single-value and multi-value model outputs.
 - Supports autoloading all models from a specified directory.
 - Thread-safe, fast, and memory-efficient.
@@ -56,22 +55,42 @@ See the [ROADMAP.md](ROADMAP.md) for the list of implemented and planned feature
 
 ### Quickstart
 
+1. Clone the repository and build Infera extension from source:
+
+```bash
+git clone --recursive https://github.com/CogitatorTech/infera.git
+cd infera
+
+make release
+```
+
+2. Start DuckDB shell (with Infera statically linked to it):
+
+```bash
+./build/release/duckdb
+```
+
+3. Run the following SQL commands in the shell to try Infera out:
+
 ```sql
--- 1. Load the extension
-LOAD
-infera;
+-- 1. Load the extension (optional when building from source)
+load 'build/release/extension/infera/infera.duckdb_extension'
+load infera;
 
 -- 2. Load a simple linear model from a remote URL
-SELECT infera_load_model('linear_model',
+select infera_load_model('linear_model',
                          'https://github.com/CogitatorTech/infera/raw/refs/heads/main/test/models/linear.onnx');
 
 -- 3. Run a prediction using a very simple linear model
 -- Model: y = 2*x1 - 1*x2 + 0.5*x3 + 0.25
-SELECT infera_predict('linear_model', 1.0, 2.0, 3.0);
+select infera_predict('linear_model', 1.0, 2.0, 3.0);
 -- Expected output: 1.75
 
 -- 4. Unload the model when we're done with it
-SELECT infera_unload_model('linear_model');
+select infera_unload_model('linear_model');
+
+-- 5. Check the Infera version
+select infera_get_version();
 ````
 
 ---
diff --git a/docs/README.md b/docs/README.md
@@ -2,17 +2,17 @@
 
 Table below includes the information about all SQL functions exposed by the Infera.
 
-| Function                                                | Return Type      | Description                                                                                                                                 |
-|:--------------------------------------------------------|:-----------------|:--------------------------------------------------------------------------------------------------------------------------------------------|
-| `infera_load_model(name VARCHAR, path_or_url VARCHAR)`  | `BOOLEAN`        | Loads an ONNX model from a local file path or a remote URL and assigns it a unique name. Returns `true` on success.                         |
-| `infera_unload_model(name VARCHAR)`                     | `BOOLEAN`        | Unloads a model, freeing its associated resources. Returns `true` on success.                                                               |
-| `infera_set_autoload_dir(path VARCHAR)`                 | `VARCHAR (JSON)` | Scans a directory for `.onnx` files, loads them automatically, and returns a JSON report of loaded models and any errors.                   |
-| `infera_get_loaded_models()`                            | `VARCHAR (JSON)` | Returns a JSON array containing the names of all currently loaded models.                                                                   |
-| `infera_get_model_info(name VARCHAR)`                   | `VARCHAR (JSON)` | Returns a JSON object with metadata about a specific loaded model, including its name, input/output shapes, and status.                     |
-| `infera_predict(name VARCHAR, features... FLOAT)`       | `FLOAT`          | Performs inference on a batch of data, returning a single float value for each input row.                                                   |
-| `infera_predict_multi(name VARCHAR, features... FLOAT)` | `VARCHAR (JSON)` | Performs inference and returns all outputs as a JSON-encoded array. This is useful for models that produce multiple predictions per sample. |
-| `infera_predict_from_blob(name VARCHAR, data BLOB)`     | `LIST[FLOAT]`    | Performs inference on raw `BLOB` data (for example, used for an image tensor), returning the result as a list of floats.                    |
-| `infera_get_version()`                                  | `VARCHAR (JSON)` | Returns a JSON object with version and build information for the Infera extension.                                                          |
+| # | Function                                                | Return Type      | Description                                                                                                                                 |
+|---|:--------------------------------------------------------|:-----------------|:--------------------------------------------------------------------------------------------------------------------------------------------|
+| 1 | `infera_load_model(name VARCHAR, path_or_url VARCHAR)`  | `BOOLEAN`        | Loads an ONNX model from a local file path or a remote URL and assigns it a unique name. Returns `true` on success.                         |
+| 2 | `infera_unload_model(name VARCHAR)`                     | `BOOLEAN`        | Unloads a model, freeing its associated resources. Returns `true` on success.                                                               |
+| 3 | `infera_set_autoload_dir(path VARCHAR)`                 | `VARCHAR (JSON)` | Scans a directory for `.onnx` files, loads them automatically, and returns a JSON report of loaded models and any errors.                   |
+| 4 | `infera_get_loaded_models()`                            | `VARCHAR (JSON)` | Returns a JSON array containing the names of all currently loaded models.                                                                   |
+| 5 | `infera_get_model_info(name VARCHAR)`                   | `VARCHAR (JSON)` | Returns a JSON object with metadata about a specific loaded model, including its name, input/output shapes, and status.                     |
+| 6 | `infera_predict(name VARCHAR, features... FLOAT)`       | `FLOAT`          | Performs inference on a batch of data, returning a single float value for each input row.                                                   |
+| 7 | `infera_predict_multi(name VARCHAR, features... FLOAT)` | `VARCHAR (JSON)` | Performs inference and returns all outputs as a JSON-encoded array. This is useful for models that produce multiple predictions per sample. |
+| 8 | `infera_predict_from_blob(name VARCHAR, data BLOB)`     | `LIST[FLOAT]`    | Performs inference on raw `BLOB` data (for example, used for an image tensor), returning the result as a list of floats.                    |
+| 9 | `infera_get_version()`                                  | `VARCHAR (JSON)` | Returns a JSON object with version and build information for the Infera extension.                                                          |
 
 ---
 
@@ -24,42 +24,42 @@ This section includes some examples of how to use the Infera functions.
 
 ```sql
 -- Load a model from a local file
-SELECT infera_load_model('local_model', '/path/to/model.onnx');
+select infera_load_model('local_model', '/path/to/model.onnx');
 
 -- Load a model from a remote URL
-SELECT infera_load_model('remote_model', 'https://.../model.onnx');
+select infera_load_model('remote_model', 'https://.../model.onnx');
 
 -- List all loaded models
-SELECT infera_get_loaded_models();
+select infera_get_loaded_models();
 -- Output: ["local_model", "remote_model"]
 
 -- Get information about a specific model
-SELECT infera_get_model_info('local_model');
+select infera_get_model_info('local_model');
 -- Output: {"name":"local_model","input_shape":[-1,3],"output_shape":[-1,1],"loaded":true}
 
 -- Unload a loaded model
-SELECT infera_unload_model('remote_model');
+select infera_unload_model('remote_model');
 ```
 
 #### Inference
 
 ```sql
 -- Predict using literal feature values
-SELECT infera_predict('my_model', 1.0, 2.5, 3.0) as prediction;
+select infera_predict('my_model', 1.0, 2.5, 3.0) as prediction;
 
 -- Predict using columns from a table
-SELECT id,
+select id,
        infera_predict('my_model', feature1, feature2, feature3) as prediction
-FROM features_table;
+from features_table;
 
 -- Get multiple outputs as a JSON array.
 -- This is useful models that return multiple outputs per prediction (like a non-binary classifier)
-SELECT infera_predict_multi('multi_output_model', 1.0, 2.0);
+select infera_predict_multi('multi_output_model', 1.0, 2.0);
 -- Output: [0.85, 0.12, 0.03]
 
 -- Predict using raw BLOB data (like tensor data)
-SELECT infera_predict_from_blob('my_model', my_blob_column)
-FROM my_table;
+select infera_predict_from_blob('my_model', my_blob_column)
+from my_table;
 -- Expected output: [0.1, 0.2, 0.3, ...] (as a LIST<FLOAT>)
 ```
 
@@ -71,11 +71,11 @@ FROM my_table;
 
 ```sql
 -- Get a JSON list of all loaded models
-SELECT infera_get_loaded_models();
+select infera_get_loaded_models();
 -- Output: ["linear_model", "squeezenet"]
 
 -- Get detailed metadata for a specific model
-SELECT infera_get_model_info('squeezenet');
+select infera_get_model_info('squeezenet');
 /* Output:
 {
   "name": "squeezenet",
@@ -86,7 +86,7 @@ SELECT infera_get_model_info('squeezenet');
 */
 
 -- Load all models from the 'models/' directory
-SELECT infera_set_autoload_dir('path/to/your/models');
+select infera_set_autoload_dir('path/to/your/models');
 /* Output:
 {
   "loaded": ["model1", "model2"],
diff --git a/docs/examples/README.md b/docs/examples/README.md
@@ -1,35 +1,24 @@
-# Infera Examples
+## Examples
 
 This directory contains DuckDB SQL scripts that show usage patterns of the Infera extension.
 Each file is self‑contained and can be executed in the DuckDB shell (or via `duckdb < file.sql`).
 
-## Prerequisites
+### Prerequisites
 
-1. Build the extension:
+1. Build the extension
    ```bash
    make release
    ```
-2. Start the DuckDB shell from the project root directory:
+2. Start the DuckDB shell from the project root directory
    ```bash
    ./build/release/duckdb
    ```
-3. Inside the shell, load a script:
+3. Inside the shell, load a script
    ```sql
    .read docs/examples/e1_core_functionality.sql
    ```
 
-## Example Index
-
-| File                            | Topic                                | Functionalities                                                                                           |
-|---------------------------------|--------------------------------------|-----------------------------------------------------------------------------------------------------------|
-| `e1_core_functionality.sql`     | Core lifecycle                       | Version, load, inspect model info, single prediction, unload, autoload directory reuse                    |
-| `e2_advanced_features.sql`      | Remote and BLOB inference            | Remote model load (GitHub), large vision model, constructing a zero‑filled BLOB for inference             |
-| `e3_integration_and_errors.sql` | Integration patterns and errors      | Deterministic batch table, aggregation, null feature detection, missing model handling                    |
-| `e4_multi_output.sql`           | Multi‑output models                  | Uses `multi_output.onnx`, shows `infera_predict_multi` vs (commented) single‑output mismatch              |
-| `e5_autoload_and_json.sql`      | Autoload and lightweight JSON checks | Error JSON from missing dir, loading multiple models via directory scan, simple substring presence checks |
-| `e6_blob_diagnostics.sql`       | BLOB diagnostics                     | Shows correct BLOB sizing for mobilenet; commented invalid case to illustrate error path                  |
-
-## Running All Examples
+### Running All Examples
 
 ```bash
 make examples
diff --git a/docs/examples/e4_multi_output.sql b/docs/examples/e4_multi_output.sql
@@ -19,4 +19,3 @@ select '## predict (single-output API) will raise mismatch error if executed';
 select '## cleanup';
 select infera_unload_model('multi_output') as unloaded_multi;
 .echo off
-
diff --git a/docs/examples/e5_autoload_and_json.sql b/docs/examples/e5_autoload_and_json.sql
@@ -19,4 +19,3 @@ select '## cleanup';
 select infera_unload_model('linear') as unload_linear;
 select infera_unload_model('multi_output') as unload_multi_output;
 .echo off
-
diff --git a/docs/examples/e6_blob_diagnostics.sql b/docs/examples/e6_blob_diagnostics.sql
@@ -21,4 +21,3 @@ select len(infera_predict_from_blob('mobilenet', zero_blob)) as output_len from
 select '## cleanup';
 select infera_unload_model('mobilenet') as mobilenet_unloaded;
 .echo off
-
