Python bindings (#115)

Signed-off-by: Anand Krishnamoorthi <anakrish@microsoft.com>

.github/workflows/publish-python.yml

@@ -0,0 +1,113 @@
+# This file is autogenerated by maturin v1.4.0
+# To update, run
+#
+#    maturin generate-ci github --zig
+#
+name: publish-python
+
+on:
+  workflow_dispatch:
+
+permissions:
+  contents: read
+
+jobs:
+  linux:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        target: [x86_64, x86, aarch64, armv7, s390x, ppc64le]
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4
+        with:
+          python-version: '3.10'
+      - name: Build wheels
+        uses: PyO3/maturin-action@v1
+        with:
+          target: ${{ matrix.target }}
+          args: --release --out dist --find-interpreter --zig
+          sccache: 'true'
+          manylinux: auto
+      - name: Upload wheels
+        uses: actions/upload-artifact@v3
+        with:
+          name: wheels
+          path: dist
+
+  windows:
+    runs-on: windows-latest
+    strategy:
+      matrix:
+        target: [x64, x86]
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4
+        with:
+          python-version: '3.10'
+          architecture: ${{ matrix.target }}
+      - name: Build wheels
+        uses: PyO3/maturin-action@v1
+        with:
+          target: ${{ matrix.target }}
+          args: --release --out dist --find-interpreter
+          sccache: 'true'
+      - name: Upload wheels
+        uses: actions/upload-artifact@v3
+        with:
+          name: wheels
+          path: dist
+
+  macos:
+    runs-on: macos-latest
+    strategy:
+      matrix:
+        target: [x86_64, aarch64]
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions/setup-python@v4
+        with:
+          python-version: '3.10'
+      - name: Build wheels
+        uses: PyO3/maturin-action@v1
+        with:
+          target: ${{ matrix.target }}
+          args: --release --out dist --find-interpreter
+          sccache: 'true'
+      - name: Upload wheels
+        uses: actions/upload-artifact@v3
+        with:
+          name: wheels
+          path: dist
+
+  sdist:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+      - name: Build sdist
+        uses: PyO3/maturin-action@v1
+        with:
+          command: sdist
+          args: --out dist
+      - name: Upload sdist
+        uses: actions/upload-artifact@v3
+        with:
+          name: wheels
+          path: dist
+
+  release:
+    name: Release
+    runs-on: ubuntu-latest
+    if: "startsWith(github.ref, 'refs/tags/')"
+    needs: [linux, windows, macos, sdist]
+    steps:
+      - uses: actions/download-artifact@v3
+        with:
+          name: wheels
+      - name: Publish to PyPI
+        uses: PyO3/maturin-action@v1
+        env:
+          MATURIN_PYPI_TOKEN: ${{ secrets.PYPI_API_TOKEN }}
+        with:
+          command: upload
+          args: --non-interactive --skip-existing *

.github/workflows/publish-wasm.yml

@@ -15,9 +15,17 @@ jobs:
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
+      # Setup .npmrc file to publish to npm
+      - uses: actions/setup-node@v3
+        with:
+          node-version: '20.x'
+          registry-url: 'https://registry.npmjs.org'
       - name: Install wasm-pack
         run: curl https://rustwasm.github.io/wasm-pack/installer/init.sh -sSf | sh
       - name: Build
         run: wasm-pack build --target nodejs -r
       - name: Publish
-        run: # TODO
+        run: wasm-pack publish --target nodejs
+        env:
+          NODE_AUTH_TOKEN: ${{ secrets.NPM_PUBLISH_TOKEN }}
+          

Cargo.toml

@@ -1,4 +1,10 @@
-workspace = { members = ["bindings/wasm"] }
+[workspace]
+
+members = [
+    "bindings/python",
+    "bindings/wasm"
+]
+
 [package]
 name = "regorus"
 description = "A fast, lightweight Rego (OPA policy language) interpreter"

README.md

@@ -65,7 +65,8 @@ builtins. See [OPA Conformance](#opa-conformance) below.
 
 Regorus can be used from a variety of languages:
 
-- Javascript (nodejs): Via npm package `regorus-wasm`. This package is Regorus compiled into WASM.
+- Javascript: Via npm package `regorusjs`. This package is Regorus compiled into WASM.
+- Python: Via `regorus` package.
 
 ## Getting Started
 

bindings/python/.python-version

@@ -0,0 +1 @@
+pyo3

bindings/python/Cargo.toml

@@ -0,0 +1,20 @@
+[package]
+name = "regoruspy"
+version = "0.1.0"
+edition = "2021"
+repository = "https://github.com/microsoft/regorus/bindings/python"
+description = "Python bindings for Regorus - a fast, lightweight Rego interpreter written in Rust"
+keywords = ["interpreter", "opa", "policy-as-code", "rego"]
+
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+[lib]
+crate-type = ["cdylib"]
+
+[dependencies]
+anyhow = "1.0.79"
+ordered-float = "4.2.0"
+pyo3 = {version = "0.20.2", features = ["anyhow", "extension-module"] }
+regorus = { path  = "../.." }
+serde_json = "1.0.112"
+

bindings/python/pyproject.toml

@@ -0,0 +1,16 @@
+[build-system]
+requires = ["maturin>=1.4,<2.0"]
+build-backend = "maturin"
+
+[project]
+name = "regorus"
+requires-python = ">=3.8"
+classifiers = [
+    "Programming Language :: Rust",
+    "Programming Language :: Python :: Implementation :: CPython",
+    "Programming Language :: Python :: Implementation :: PyPy",
+]
+dynamic = ["version"]
+
+[tool.maturin]
+features = ["pyo3/extension-module"]

bindings/python/src/lib.rs

@@ -0,0 +1,279 @@
+// Copyright (c) Microsoft Corporation.
+// Licensed under the MIT License.
+use anyhow::{anyhow, Result};
+use pyo3::exceptions::PyTypeError;
+use pyo3::prelude::*;
+use pyo3::types::*;
+
+use std::collections::{BTreeMap, BTreeSet};
+
+use regorus::Value;
+
+/// Python wrapper for [`regorus::Engine`]
+#[pyclass(unsendable)]
+pub struct Engine {
+    engine: regorus::Engine,
+}
+
+impl Default for Engine {
+    fn default() -> Self {
+        Self::new()
+    }
+}
+
+impl Clone for Engine {
+    /// Clone a [`Engine`]
+    ///
+    /// To avoid having to parse same policy again, the engine can be cloned
+    /// after policies and data have been added.
+    fn clone(&self) -> Self {
+        Self {
+            engine: self.engine.clone(),
+        }
+    }
+}
+
+fn from<'source>(ob: &'source PyAny) -> Result<Value, PyErr> {
+    // dicts
+    Ok(if let Ok(dict) = ob.downcast::<PyDict>() {
+        let mut map = BTreeMap::new();
+        for (k, v) in dict {
+            map.insert(from(k)?, from(v)?);
+        }
+        map.into()
+    }
+    // set
+    else if let Ok(pset) = ob.downcast::<PySet>() {
+        let mut set = BTreeSet::new();
+        for v in pset {
+            set.insert(from(v)?);
+        }
+        set.into()
+    }
+    // frozen set
+    else if let Ok(pfset) = ob.downcast::<PyFrozenSet>() {
+        //
+        let mut set = BTreeSet::new();
+        for v in pfset {
+            set.insert(from(v)?);
+        }
+        set.into()
+    }
+    // lists and tuples
+    else if let Ok(plist) = ob.downcast::<PyList>() {
+        let mut array = Vec::new();
+        for v in plist {
+            array.push(from(v)?);
+        }
+        array.into()
+    } else if let Ok(ptuple) = ob.downcast::<PyTuple>() {
+        let mut array = Vec::new();
+        for v in ptuple {
+            array.push(from(v)?);
+        }
+        array.into()
+    }
+    // String
+    else if let Ok(s) = String::extract(ob) {
+        s.into()
+    }
+    // Numeric
+    else if let Ok(v) = i64::extract(ob) {
+        v.into()
+    } else if let Ok(v) = u64::extract(ob) {
+        v.into()
+    } else if let Ok(v) = f64::extract(ob) {
+        v.into()
+    }
+    // Boolean
+    else if let Ok(b) = bool::extract(ob) {
+        b.into()
+    }
+    // None
+    else if ob.downcast::<PyNone>().is_ok() {
+        Value::Null
+    }
+    // Anything that is a sequence
+    else if let Ok(pseq) = ob.downcast::<PySequence>() {
+        let mut array = Vec::new();
+        for i in 0..pseq.len()? {
+            array.push(from(pseq.get_item(i)?)?);
+        }
+        array.into()
+    }
+    // Anything that is a map
+    else if let Ok(pmap) = ob.downcast::<PyMapping>() {
+        let mut map = BTreeMap::new();
+        let keys = pmap.keys()?;
+        let values = pmap.values()?;
+        for i in 0..keys.len()? {
+            let key = keys.get_item(i)?;
+            let value = values.get_item(i)?;
+            map.insert(from(key)?, from(value)?);
+        }
+        map.into()
+    } else {
+        return Err(PyErr::new::<PyTypeError, _>(
+            "object cannot be converted to RegoValue",
+        ));
+    })
+}
+
+fn to(mut v: Value, py: Python<'_>) -> Result<PyObject> {
+    Ok(match v {
+        Value::Null => None::<u64>.to_object(py),
+
+        // TODO: Revisit this mapping
+        Value::Undefined => None::<u64>.to_object(py),
+
+        Value::Bool(b) => b.to_object(py),
+        Value::String(s) => s.to_object(py),
+
+        Value::Number(_) => {
+            if let Ok(f) = v.as_f64() {
+                f.to_object(py)
+            } else if let Ok(u) = v.as_u64() {
+                u.to_object(py)
+            } else {
+                v.as_i64()?.to_object(py)
+            }
+        }
+
+        Value::Array(_) => {
+            let list = PyList::empty(py);
+            for v in std::mem::replace(v.as_array_mut()?, Vec::new()) {
+                list.append(to(v, py)?)?;
+            }
+            list.into()
+        }
+
+        Value::Set(_) => {
+            let set = PySet::empty(py)?;
+            for v in std::mem::replace(v.as_set_mut()?, BTreeSet::new()) {
+                set.add(to(v, py)?)?;
+            }
+            set.into()
+        }
+
+        Value::Object(_) => {
+            let dict = PyDict::new(py);
+            for (k, v) in std::mem::replace(v.as_object_mut()?, BTreeMap::new()) {
+                dict.set_item(to(k, py)?, to(v, py)?)?;
+            }
+            dict.into()
+        }
+    })
+}
+
+#[pymethods]
+impl Engine {
+    /// Construct a new Engine
+    #[new]
+    pub fn new() -> Self {
+        Self {
+            engine: regorus::Engine::new(),
+        }
+    }
+
+    /// Add a policy
+    ///
+    /// The policy is parsed into AST.
+    ///
+    /// * `path`: A filename to be associated with the policy.
+    /// * `rego`: Rego policy.
+    pub fn add_policy(&mut self, path: String, rego: String) -> Result<()> {
+        self.engine.add_policy(path, rego)?;
+        Ok(())
+    }
+
+    /// Add policy data.
+    ///
+    /// * `data`: Rego value. A Rego value is a number, bool, string, None
+    ///           or a list/set/map whose items themselves are Rego values.
+    pub fn add_data(&mut self, data: &PyAny) -> Result<()> {
+        let data = from(data)?;
+        self.engine.add_data(data)
+    }
+
+    /// Add policy data.
+    ///
+    /// * `data`: JSON encoded value to be used as policy data.
+    pub fn add_data_json(&mut self, data: String) -> Result<()> {
+        let data = regorus::Value::from_json_str(&data)?;
+        self.engine.add_data(data)
+    }
+
+    /// Clear policy data.
+    pub fn clear_data(&mut self) -> Result<()> {
+        self.engine.clear_data();
+        Ok(())
+    }
+
+    /// Set input.
+    ///
+    /// * `input`: Rego value. A Rego value is a number, bool, string, None
+    ///            or a list/set/map whose items themselves are Rego values.
+    pub fn set_input(&mut self, input: &PyAny) -> Result<()> {
+        let input = from(input)?;
+        self.engine.set_input(input);
+        Ok(())
+    }
+
+    /// Set input.
+    ///
+    /// * `input`: JSON encoded value to be used as input to query.
+    pub fn set_input_json(&mut self, input: String) -> Result<()> {
+        let input = regorus::Value::from_json_str(&input)?;
+        self.engine.set_input(input);
+        Ok(())
+    }
+
+    /// Evaluate query.
+    ///
+    /// * `query`: Rego expression to be evaluate.
+    pub fn eval_query(&mut self, query: String, py: Python<'_>) -> Result<PyObject> {
+        let results = self.engine.eval_query(query, false)?;
+
+        let rlist = PyList::empty(py);
+        for result in results.result.into_iter() {
+            let rdict = PyDict::new(py);
+
+            let elist = PyList::empty(py);
+            for expr in result.expressions.into_iter() {
+                let edict = PyDict::new(py);
+                edict.set_item("value".to_object(py), to(expr.value, py)?)?;
+                edict.set_item("text".to_object(py), expr.text.as_ref().to_object(py))?;
+
+                let ldict = PyDict::new(py);
+                ldict.set_item("row".to_object(py), expr.location.row.to_object(py))?;
+                ldict.set_item("col".to_object(py), expr.location.col.to_object(py))?;
+
+                edict.set_item("location".to_object(py), ldict)?;
+                elist.append(edict)?;
+            }
+
+            rdict.set_item("expressions".to_object(py), elist)?;
+            rdict.set_item("bindings".to_object(py), to(result.bindings, py)?)?;
+            rlist.append(rdict)?;
+        }
+        let dict = PyDict::new(py);
+        dict.set_item("result".to_object(py), rlist)?;
+        Ok(dict.into())
+    }
+
+    /// Evaluate query. Returns result as JSON.
+    ///
+    /// * `query`: Rego expression to be evaluate.
+    pub fn eval_query_as_json(&mut self, query: String) -> Result<String> {
+        let results = self.engine.eval_query(query, false)?;
+        serde_json::to_string_pretty(&results).map_err(|e| anyhow!("{e}"))
+    }
+}
+
+mod export {
+    use pyo3::prelude::*;
+    #[pymodule]
+    fn regorus(_py: Python<'_>, m: &PyModule) -> PyResult<()> {
+        m.add_class::<crate::Engine>()
+    }
+}

bindings/wasm/Cargo.toml

@@ -1,5 +1,5 @@
 [package]
-name = "regorus-wasm"
+name = "regorusjs"
 version = "0.1.0"
 edition = "2021"
 repository = "https://github.com/microsoft/regorus/bindings/wasm"

bindings/wasm/README.md

@@ -1,4 +1,4 @@
-# regorus-wasm
+# regorusjs
 
 **Regorus** is
 
@@ -8,21 +8,70 @@
 
 See [Repository](https://github.com/microsoft/regorus).
 
-`regorus-wasm` is Regorus compiled into WASM.
+`regorusjs` is Regorus compiled into WASM.
 
 ## Usage
 
-In nodejs,
+```javascript
 
-``javascript
-
-var regorus = require('regorus-wasm')
+var regorus = require('regorusjs')
 
 // Create an engine.
 var engine = new regorus.Engine();
 
 // Add Rego policy.
-engine.add_policy()
+engine.add_policy(
+    // Associate this file name with policy
+    'hello.rego',
+    
+    // Rego policy
+`
+  package test
+  
+  # Join messages
+  message = concat(", ", [input.message, data.message])
+`)
 
+// Set policy data
+engine.add_data_json(`
+ {
+    "message" : "World!"
+ }
+`)
 
+// Set policy input
+engine.set_input_json(`
+ {
+	"message" : "Hello"
+ }
+`)
+
+// Eval query
+results = engine.eval_query('data.test.message')
+
+// Display
+console.log(results)
+// {
+//   "result": [
+//     {
+//       "expressions": [
+//         {
+//           "value": "Hello, World!",
+//           "text": "data.test.message",
+//           "location": {
+//             "row": 1,
+//             "col": 1
+//           }
+//         }
+//       ]
+//     }
+//   ]
+// }
+
+// Convert results to object
+results = JSON.parse(results)
+
+// Process result
+console.log(results.result[0].expressions[0].value)
+// Hello, World!
 ```

bindings/wasm/src/lib.rs

@@ -63,6 +63,12 @@ impl Engine {
         self.engine.add_data(data).map_err(error_to_jsvalue)
     }
 
+    /// Clear policy data.
+    pub fn clear_data(&mut self) -> Result<(), JsValue> {
+        self.engine.clear_data();
+        Ok(())
+    }
+
     /// Set input.
     ///
     /// See https://docs.rs/regorus/0.1.0-alpha.2/regorus/struct.Engine.html#method.set_input

src/engine.rs

@@ -13,7 +13,7 @@ use crate::QueryResults;
 use std::convert::AsRef;
 use std::path::Path;
 
-use anyhow::Result;
+use anyhow::{bail, Result};
 
 /// The Rego evaluation engine.
 ///
@@ -175,6 +175,9 @@ impl Engine {
     /// # }
     /// ```
     pub fn add_data(&mut self, data: Value) -> Result<()> {
+        if data.as_object().is_err() {
+            bail!("data must be object");
+        }
         self.prepared = false;
         self.interpreter.get_data_mut().merge(data)
     }