Support for set WOQL operators

hoijnet · hoijnet · commit 82bc07ba79da · 2025-12-13T12:16:09.000+01:00
diff --git a/.gitignore b/.gitignore
@@ -35,6 +35,7 @@ terminusdb_client_coverage/
 *~
 
 venv/
+.venv/
 
 # due to using tox and pytest
 .tox
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
@@ -4,19 +4,58 @@ Thanks for interested to contribute to TerminusDB Client, to get started, fork t
 
 ## Setting up dev environment 💻
 
-Make sure you have Python>=3.9 installed. We use [pipenv](https://pipenv-fork.readthedocs.io/en/latest/) for dev environment, to install pipenv:
+Make sure you have Python>=3.9 and <3.13 installed.
 
-`pip3 install pipenv --upgrade`
+[Fork and clone](https://help.github.com/en/github/getting-started-with-github/fork-a-repo) this repo, then set up your development environment using one of the methods below.
 
-[Fork and clone](https://help.github.com/en/github/getting-started-with-github/fork-a-repo) this repo, then in your local repo:
+### Option 1: Using venv (recommended)
 
-`pipenv install --dev --pre` or `make init`
+Create and activate a virtual environment:
 
-To “editable” install the local Terminus Client Python:
+```bash
+# Create venv with Python 3.12 (or any version 3.9-3.12)
+python3.12 -m venv .venv
 
-`pip3 install -e .`
+# Activate the virtual environment
+source .venv/bin/activate  # On macOS/Linux
+# .venv\Scripts\activate   # On Windows
 
-**to be able to run integration tests, local installation of docker is required**
+# Install the package in editable mode with dev dependencies
+pip install -e ".[dev]"
+
+# Install pytest for running tests
+pip install pytest
+```
+
+### Option 2: Using pipenv
+
+We also support [pipenv](https://pipenv-fork.readthedocs.io/en/latest/) for dev environment:
+
+```bash
+pip install pipenv --upgrade
+pipenv install --dev --pre
+```
+
+Or simply run `make init`.
+
+To "editable" install the local Terminus Client Python:
+
+`pip install -e .`
+
+### Running a local TerminusDB server
+
+**To run integration tests, you need either Docker or a local TerminusDB server.**
+
+For integration tests, you can either:
+
+1. **Use Docker** (automatic): Tests will automatically start a Docker container if no server is detected
+2. **Use a local server**: Start the TerminusDB test server from the main terminusdb repository:
+   ```bash
+   cd /path/to/terminusdb
+   ./tests/terminusdb-test-server.sh start
+   ```
+
+The test configuration will automatically detect and use an available server.
 
 We use [shed](https://pypi.org/project/shed/) to lint our code. Although you can do it manually by running `shed`, we highly recommend setting up the pre-commit hook to do the linting automatically.
 
diff --git a/terminusdb_client/tests/integration_tests/test_woql_set_operations.py b/terminusdb_client/tests/integration_tests/test_woql_set_operations.py
@@ -0,0 +1,183 @@
+"""
+Integration tests for WOQL set operations.
+
+These tests verify the new set operations:
+- set_difference
+- set_intersection
+- set_union
+- set_member
+- list_to_set
+"""
+import time
+
+import pytest
+
+from terminusdb_client import Client
+from terminusdb_client.woqlquery.woql_query import WOQLQuery
+
+test_user_agent = "terminusdb-client-python-tests"
+
+
+def extract_values(result_list):
+    """Extract raw values from a list of typed literals."""
+    if not result_list:
+        return []
+    return [item["@value"] if isinstance(item, dict) and "@value" in item else item
+            for item in result_list]
+
+
+class TestWOQLSetOperations:
+    """Tests for WOQL set operations."""
+
+    @pytest.fixture(autouse=True)
+    def setup_teardown(self, docker_url):
+        """Setup and teardown for each test."""
+        self.client = Client(docker_url, user_agent=test_user_agent)
+        self.client.connect()
+        self.db_name = "test_woql_set_operations"
+        
+        # Create database for tests
+        if self.db_name in self.client.list_databases():
+            self.client.delete_database(self.db_name)
+        self.client.create_database(self.db_name)
+        
+        yield
+        
+        # Cleanup
+        self.client.delete_database(self.db_name)
+
+    def test_set_difference_basic(self):
+        """Test set_difference computes difference between two lists."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:ListA", [1, 2, 3, 4]),
+            WOQLQuery().eq("v:ListB", [2, 4]),
+            WOQLQuery().set_difference("v:ListA", "v:ListB", "v:Diff")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+        assert extract_values(result["bindings"][0]["Diff"]) == [1, 3]
+
+    def test_set_difference_subset(self):
+        """Test set_difference returns empty when first list is subset."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:ListA", [1, 2]),
+            WOQLQuery().eq("v:ListB", [1, 2, 3]),
+            WOQLQuery().set_difference("v:ListA", "v:ListB", "v:Diff")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+        assert result["bindings"][0]["Diff"] == []
+
+    def test_set_difference_empty_list(self):
+        """Test set_difference handles empty lists."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:ListA", []),
+            WOQLQuery().eq("v:ListB", [1]),
+            WOQLQuery().set_difference("v:ListA", "v:ListB", "v:Diff")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+        assert result["bindings"][0]["Diff"] == []
+
+    def test_set_intersection_basic(self):
+        """Test set_intersection computes intersection of two lists."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:ListA", [1, 2, 3]),
+            WOQLQuery().eq("v:ListB", [2, 3, 4]),
+            WOQLQuery().set_intersection("v:ListA", "v:ListB", "v:Common")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+        assert extract_values(result["bindings"][0]["Common"]) == [2, 3]
+
+    def test_set_intersection_no_common(self):
+        """Test set_intersection returns empty when no common elements."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:ListA", [1, 2]),
+            WOQLQuery().eq("v:ListB", [3, 4]),
+            WOQLQuery().set_intersection("v:ListA", "v:ListB", "v:Common")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+        assert result["bindings"][0]["Common"] == []
+
+    def test_set_union_basic(self):
+        """Test set_union computes union of two lists."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:ListA", [1, 2]),
+            WOQLQuery().eq("v:ListB", [2, 3]),
+            WOQLQuery().set_union("v:ListA", "v:ListB", "v:All")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+        assert extract_values(result["bindings"][0]["All"]) == [1, 2, 3]
+
+    def test_set_union_removes_duplicates(self):
+        """Test set_union removes duplicates."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:ListA", [1, 1, 2]),
+            WOQLQuery().eq("v:ListB", [2, 2]),
+            WOQLQuery().set_union("v:ListA", "v:ListB", "v:All")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+        assert extract_values(result["bindings"][0]["All"]) == [1, 2]
+
+    def test_set_member_success(self):
+        """Test set_member succeeds for element in set."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:MySet", [1, 2, 3]),
+            WOQLQuery().set_member(2, "v:MySet")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+
+    def test_set_member_failure(self):
+        """Test set_member fails for element not in set."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:MySet", [1, 2, 3]),
+            WOQLQuery().set_member(5, "v:MySet")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 0
+
+    def test_list_to_set(self):
+        """Test list_to_set removes duplicates and sorts."""
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:MyList", [3, 1, 2, 1]),
+            WOQLQuery().list_to_set("v:MyList", "v:MySet")
+        )
+        
+        result = self.client.query(query)
+        assert len(result["bindings"]) == 1
+        assert extract_values(result["bindings"][0]["MySet"]) == [1, 2, 3]
+
+    def test_performance_large_sets(self):
+        """Test set operations handle large sets efficiently."""
+        list_a = list(range(1000))
+        list_b = list(range(500, 1500))
+        
+        query = WOQLQuery().woql_and(
+            WOQLQuery().eq("v:ListA", list_a),
+            WOQLQuery().eq("v:ListB", list_b),
+            WOQLQuery().set_difference("v:ListA", "v:ListB", "v:Diff")
+        )
+        
+        start_time = time.time()
+        result = self.client.query(query)
+        elapsed = time.time() - start_time
+        
+        assert len(result["bindings"]) == 1
+        assert len(result["bindings"][0]["Diff"]) == 500
+        
+        # Should complete in under 1 second with O(n log n) algorithm
+        assert elapsed < 1.0
diff --git a/terminusdb_client/woqlquery/woql_query.py b/terminusdb_client/woqlquery/woql_query.py
@@ -2426,6 +2426,135 @@ def member(self, member, mem_list):
         self._cursor["list"] = self._value_list(mem_list)
         return self
 
+    def set_difference(self, list_a, list_b, result):
+        """Computes the set difference between two lists (elements in list_a but not in list_b)
+
+        Parameters
+        ----------
+        list_a : str or list
+            First list or variable
+        list_b : str or list
+            Second list or variable
+        result : str
+            Variable to store the result
+
+        Returns
+        -------
+        WOQLQuery object
+            query object that can be chained and/or execute
+        """
+        if list_a and list_a == "args":
+            return ["list_a", "list_b", "result"]
+        if self._cursor.get("@type"):
+            self._wrap_cursor_with_and()
+        self._cursor["@type"] = "SetDifference"
+        self._cursor["list_a"] = self._value_list(list_a)
+        self._cursor["list_b"] = self._value_list(list_b)
+        self._cursor["result"] = self._value_list(result)
+        return self
+
+    def set_intersection(self, list_a, list_b, result):
+        """Computes the set intersection of two lists (elements in both list_a and list_b)
+
+        Parameters
+        ----------
+        list_a : str or list
+            First list or variable
+        list_b : str or list
+            Second list or variable
+        result : str
+            Variable to store the result
+
+        Returns
+        -------
+        WOQLQuery object
+            query object that can be chained and/or execute
+        """
+        if list_a and list_a == "args":
+            return ["list_a", "list_b", "result"]
+        if self._cursor.get("@type"):
+            self._wrap_cursor_with_and()
+        self._cursor["@type"] = "SetIntersection"
+        self._cursor["list_a"] = self._value_list(list_a)
+        self._cursor["list_b"] = self._value_list(list_b)
+        self._cursor["result"] = self._value_list(result)
+        return self
+
+    def set_union(self, list_a, list_b, result):
+        """Computes the set union of two lists (all unique elements from both lists)
+
+        Parameters
+        ----------
+        list_a : str or list
+            First list or variable
+        list_b : str or list
+            Second list or variable
+        result : str
+            Variable to store the result
+
+        Returns
+        -------
+        WOQLQuery object
+            query object that can be chained and/or execute
+        """
+        if list_a and list_a == "args":
+            return ["list_a", "list_b", "result"]
+        if self._cursor.get("@type"):
+            self._wrap_cursor_with_and()
+        self._cursor["@type"] = "SetUnion"
+        self._cursor["list_a"] = self._value_list(list_a)
+        self._cursor["list_b"] = self._value_list(list_b)
+        self._cursor["result"] = self._value_list(result)
+        return self
+
+    def set_member(self, element, set_list):
+        """Checks if an element is a member of a set (efficient O(log n) lookup)
+
+        Parameters
+        ----------
+        element : any
+            Element to check
+        set_list : str or list
+            Set (list) to check membership in
+
+        Returns
+        -------
+        WOQLQuery object
+            query object that can be chained and/or execute
+        """
+        if element and element == "args":
+            return ["element", "set"]
+        if self._cursor.get("@type"):
+            self._wrap_cursor_with_and()
+        self._cursor["@type"] = "SetMember"
+        self._cursor["element"] = self._clean_object(element)
+        self._cursor["set"] = self._value_list(set_list)
+        return self
+
+    def list_to_set(self, input_list, result_set):
+        """Converts a list to a set (removes duplicates and sorts)
+
+        Parameters
+        ----------
+        input_list : str or list
+            Input list or variable
+        result_set : str
+            Variable to store the resulting set
+
+        Returns
+        -------
+        WOQLQuery object
+            query object that can be chained and/or execute
+        """
+        if input_list and input_list == "args":
+            return ["list", "set"]
+        if self._cursor.get("@type"):
+            self._wrap_cursor_with_and()
+        self._cursor["@type"] = "ListToSet"
+        self._cursor["list"] = self._value_list(input_list)
+        self._cursor["set"] = self._value_list(result_set)
+        return self
+
     def concat(self, concat_list, result):
         """Concatenates the list of variables into a string and saves the result in v
 

-Original file line number
+Diff line change
 *~
 venv/
 +.venv/
 # due to using tox and pytest
 .tox