fix(ci): resolve ruff lint errors and add libre2-dev for google-re2 build

- Fix 36 ruff lint errors (unused imports, E701, E402, F841, F541)
- Add libre2-dev system dependency and pybind11 build deps to CI workflow
- Use --no-build-isolation with CFLAGS for google-re2 compilation

Author: Brad Kinnard

.github/workflows/ci.yml

@@ -21,10 +21,16 @@ jobs:
       with:
         python-version: ${{ matrix.python-version }}
 
+    - name: Install system dependencies
+      run: |
+        sudo apt-get update
+        sudo apt-get install -y libre2-dev
+
     - name: Install dependencies
       run: |
         python -m pip install --upgrade pip
-        pip install ruff pytest
+        pip install ruff pytest pybind11 setuptools wheel
+        CFLAGS="-I$(python -c 'import pybind11; print(pybind11.get_include())')" pip install google-re2==1.0.0 --no-build-isolation
         pip install .[embed,demo]
         
     - name: Lint with Ruff

benchmarks/run_benchmark.py

@@ -1,9 +1,7 @@
-import sys
 import time
 import json
 import random
 import subprocess
-import os
 from pathlib import Path
 
 # Provide a sample schema for an Apache Access Log
@@ -99,11 +97,11 @@ def extract(text):
     import math
     std_dev = math.sqrt(variance)
 
-    print(f"\n======== WARM CACHE BENCHMARK (1000 lines) ========")
+    print("\n======== WARM CACHE BENCHMARK (1000 lines) ========")
     print(f"Average total wall time: {avg_time:.2f}ms ± {std_dev:.2f}ms")
     print(f"Min: {min_time:.2f}ms, Max: {max_time:.2f}ms")
     print(f"Lines processed per second: {1000 / (avg_time / 1000):.2f} ops/sec")
-    print(f"===================================================")
+    print("===================================================")
 
     schema_path.unlink()
 

benchmarks/run_examples.py

@@ -1,10 +1,8 @@
-import sys
 import time
 import json
 import random
 import subprocess
 import math
-from pathlib import Path
 
 # Schemas must be pre-created in examples/
 SCHEMAS = {
@@ -165,7 +163,7 @@ def run_benchmark(name, schema_path, logs_str, true_log_len):
     print(f"\\n======== {name.upper()} WARM CACHE (1000 elements) ========")
     print(f"Avg Wall Time: {avg_time:.2f}ms ± {std_dev:.2f}ms")
     print(f"Throughput: {1000 / (avg_time / 1000):.2f} ops/sec")
-    print(f"===========================================================\\n")
+    print("===========================================================\\n")
 
 if __name__ == "__main__":
     print("Generating logs (with random.seed(42))...")

symparse/cache_manager.py

@@ -3,7 +3,7 @@
 import logging
 import hashlib
 from pathlib import Path
-from typing import Optional, Any
+from typing import Optional
 import portalocker
 
 logger = logging.getLogger(__name__)
@@ -52,7 +52,8 @@ def _cosine_similarity(self, vec1: list[float], vec2: list[float]) -> float:
         dot = sum(a*b for a, b in zip(vec1, vec2))
         mag1 = math.sqrt(sum(a*a for a in vec1))
         mag2 = math.sqrt(sum(b*b for b in vec2))
-        if mag1 * mag2 == 0: return 0.0
+        if mag1 * mag2 == 0:
+            return 0.0
         return dot / (mag1 * mag2)
 
     def _get_embedding(self, text: str) -> list[float]:

symparse/compiler.py

@@ -2,7 +2,6 @@
 import logging
 import ast
 import re2
-from typing import Any
 from symparse.ai_client import AIClient
 
 logger = logging.getLogger(__name__)

symparse/demo.py

@@ -1,6 +1,5 @@
 import time
 import sys
-import subprocess
 
 def type_text(text, speed=0.03, newline=True):
     for char in text:

symparse/engine.py

@@ -1,5 +1,5 @@
-import json
 import logging
+from dataclasses import dataclass
 from enum import Enum
 from typing import Any, Dict
 
@@ -14,8 +14,6 @@ class GracefulDegradationMode(Enum):
     HALT = "halt"
     PASSTHROUGH = "passthrough"
 
-from dataclasses import dataclass
-
 @dataclass
 class EngineStats:
     fast_path_hits: int = 0

tests/test_ai_client.py

@@ -1,6 +1,5 @@
 import pytest
-from symparse.ai_client import AIClient, ConfidenceDegradationError
-import litellm
+from symparse.ai_client import AIClient
 import os
 
 def test_ai_client_extract_success():

tests/test_cache_manager.py

@@ -1,12 +1,9 @@
-import os
 import json
 import multiprocessing
-import pytest
-from pathlib import Path
 from symparse.cache_manager import CacheManager
 
 def test_cache_init_metadata(tmp_path):
-    cm = CacheManager(cache_dir=tmp_path)
+    CacheManager(cache_dir=tmp_path)
     meta_file = tmp_path / "metadata.json"
     assert meta_file.exists()
 

tests/test_cli.py

@@ -1,12 +1,11 @@
 import pytest
-from unittest.mock import patch
+from unittest.mock import patch, mock_open
 import sys
 import io
 from symparse.cli import main
 
 def test_cli_cache_list_clear(monkeypatch, capsys):
     from symparse.cli import main
-    import json
 
     # We can mock CacheManager.list_cache and clear_cache
     class DummyCache:
@@ -60,9 +59,6 @@ def test_run_command_no_stdin(capsys):
     assert "Error: No data piped into stdin." in captured.err
 
 
-
-from unittest.mock import mock_open
-
 def test_run_command_success(capsys):
     test_args = ["symparse", "run", "--schema", "dummy.json", "--compile"]
     dummy_schema = '{"type": "object", "properties": {"name": {"type": "string"}}}'