code-analyser 1.1.0__tar.gz → 1.3.0__tar.gz

{code_analyser-1.1.0 → code_analyser-1.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-analyser
-Version: 1.1.0
+Version: 1.3.0
 Summary: Source code analyser — part of the analyser family
 License-File: LICENSE
 Requires-Python: >=3.10
@@ -8,6 +8,7 @@ Requires-Dist: esprima>=4.0.0
 Requires-Dist: fastapi>=0.109.0
 Requires-Dist: html5lib>=1.1
 Requires-Dist: httpx>=0.27.0
+Requires-Dist: lens-contract>=0.2.0
 Requires-Dist: pydantic-settings>=2.1.0
 Requires-Dist: pydantic>=2.5.0
 Requires-Dist: python-multipart>=0.0.6
@@ -21,6 +22,8 @@ Requires-Dist: httpx>=0.27.0; extra == 'dev'
 Requires-Dist: pytest-cov>=4.1.0; extra == 'dev'
 Requires-Dist: pytest>=7.4.0; extra == 'dev'
 Requires-Dist: ruff>=0.4.0; extra == 'dev'
+Provides-Extra: embeddings
+Requires-Dist: lens-embed[text]>=0.1.1; extra == 'embeddings'
 Provides-Extra: llm
 Requires-Dist: anthropic>=0.7.0; extra == 'llm'
 Description-Content-Type: text/markdown
@@ -29,7 +32,7 @@ Description-Content-Type: text/markdown
 
 Analyses source code files and returns style violations, complexity metrics, and quality indicators. Designed as a low-level tool — feed it a file, get back structured JSON.
 
-Part of the [analyser family](#the-analyser-family).
+Part of the [analyser family](https://github.com/michael-borck/lens-analysers).
 
 > **Status**: Early development. Currently supports Python via ruff and basic AST metrics. Multi-language support and alignment with the family API pattern is in progress.
 

{code_analyser-1.1.0 → code_analyser-1.3.0}/README.md

@@ -2,7 +2,7 @@
 
 Analyses source code files and returns style violations, complexity metrics, and quality indicators. Designed as a low-level tool — feed it a file, get back structured JSON.
 
-Part of the [analyser family](#the-analyser-family).
+Part of the [analyser family](https://github.com/michael-borck/lens-analysers).
 
 > **Status**: Early development. Currently supports Python via ruff and basic AST metrics. Multi-language support and alignment with the family API pattern is in progress.
 

code_analyser-1.3.0/examples/basic_usage.md

@@ -0,0 +1,43 @@
+# code-analyser — basic usage
+
+Analyse a source file (`.py`, `.js`, `.ts`, `.html`, `.css`, `.sql`, `.ipynb`, or a `.zip` of them) for structural code signals.
+
+## Install
+
+```bash
+pip install code-analyser
+```
+
+## CLI
+
+```bash
+# Human-readable table
+code-analyser app.py
+
+# JSON output
+code-analyser app.py --json
+
+# Include LLM quality signals (requires the [llm] extra)
+code-analyser app.py --llm
+```
+
+## Python
+
+```python
+from code_analyser import CodeAnalyser
+
+result = CodeAnalyser().analyse("app.py")
+print(result.languages_detected, result.file_count)
+for f in result.files:
+    print(f.filename, f.language, f.metrics)
+```
+
+## HTTP
+
+```bash
+# Start the server (default port 8004)
+code-analyser serve
+
+# Analyse a source file via multipart upload
+curl -F file=@app.py http://localhost:8004/analyse
+```

{code_analyser-1.1.0 → code_analyser-1.3.0}/pyproject.toml

@@ -4,11 +4,12 @@ build-backend = "hatchling.build"
 
 [project]
 name = "code-analyser"
-version = "1.1.0"
+version = "1.3.0"
 description = "Source code analyser — part of the analyser family"
 readme = "README.md"
 requires-python = ">=3.10"
 dependencies = [
+    "lens-contract>=0.2.0",
     "pydantic>=2.5.0",
     "pydantic-settings>=2.1.0",
     "fastapi>=0.109.0",
@@ -25,6 +26,9 @@ dependencies = [
 
 [project.optional-dependencies]
 llm = ["anthropic>=0.7.0"]
+embeddings = [
+    "lens-embed[text]>=0.1.1",
+]
 dev = [
     "pytest>=7.4.0",
     "pytest-cov>=4.1.0",
@@ -32,6 +36,12 @@ dev = [
     "ruff>=0.4.0",
 ]
 
+# Local dev: resolve lens-contract from the sibling checkout. uv strips this from
+# the published wheel, which keeps the plain `lens-contract>=0.2.0` PyPI pin.
+[tool.uv.sources]
+lens-contract = { path = "../lens-contract", editable = true }
+lens-embed = { path = "../lens-embed", editable = true }
+
 [project.scripts]
 code-analyser = "code_analyser.cli:main"
 

code_analyser-1.3.0/src/code_analyser/api.py

@@ -0,0 +1,32 @@
+from __future__ import annotations
+
+from fastapi import FastAPI, File, HTTPException, UploadFile
+from lens_contract import add_contract_routes, add_cors, upload_tempfile
+
+from .manifest import MANIFEST
+from .models import CodeAnalysis
+from .pipeline import CodeAnalyser
+
+app = FastAPI(title="code-analyser", version=MANIFEST["version"])
+
+# GET /health and GET /manifest (the family contract, via lens-contract).
+add_contract_routes(app, MANIFEST)
+# CORS — env-driven: CODE_ANALYSER_MODE=desktop (Electron) or CODE_ANALYSER_ALLOWED_ORIGINS.
+add_cors(app, env_prefix="CODE_ANALYSER")
+
+_analyser = CodeAnalyser()
+
+
+@app.post("/analyse", response_model=CodeAnalysis)
+async def analyse(file: UploadFile = File(...)) -> CodeAnalysis:
+    content = await file.read()
+    if not content:
+        raise HTTPException(status_code=422, detail="Empty file")
+
+    with upload_tempfile(content, file.filename) as tmp_path:
+        try:
+            return _analyser.analyse(tmp_path)
+        except ValueError as e:
+            raise HTTPException(status_code=422, detail=str(e)) from e
+        except Exception as e:
+            raise HTTPException(status_code=500, detail=str(e)) from e

{code_analyser-1.1.0 → code_analyser-1.3.0}/src/code_analyser/cli.py

@@ -1,6 +1,5 @@
 from __future__ import annotations
 import argparse
-import os
 import sys
 from pathlib import Path
 
@@ -8,14 +7,17 @@ from .pipeline import CodeAnalyser
 
 
 def main() -> None:
-    if len(sys.argv) > 1 and sys.argv[1] == "serve":
-        _serve(sys.argv[2:])
-        return
+    from lens_contract import run_contract_subcommands
+
+    from .manifest import MANIFEST
 
-    if len(sys.argv) > 1 and sys.argv[1] == "manifest":
-        import json
-        from .manifest import MANIFEST
-        print(json.dumps(MANIFEST, indent=2))
+    # `serve` and `manifest` are the family's shared subcommands (lens-contract).
+    if run_contract_subcommands(
+        MANIFEST,
+        app_path="code_analyser.api:app",
+        default_port=8004,
+        env_prefix="CODE_ANALYSER",
+    ):
         return
 
     parser = argparse.ArgumentParser(
@@ -105,10 +107,5 @@ def _summarise_metrics(metrics, language: str) -> str:
     return ""
 
 
-def _serve(argv: list[str]) -> None:
-    import uvicorn
-    parser = argparse.ArgumentParser(prog="code-analyser serve")
-    parser.add_argument("--port", type=int, default=int(os.getenv("CODE_ANALYSER_PORT", "8004")))
-    parser.add_argument("--host", default=os.getenv("CODE_ANALYSER_HOST", "127.0.0.1"))
-    args = parser.parse_args(argv)
-    uvicorn.run("code_analyser.api:app", host=args.host, port=args.port)
+if __name__ == "__main__":
+    main()

code_analyser-1.3.0/src/code_analyser/embedding.py

@@ -0,0 +1,25 @@
+"""Source-code embedding via the family's shared helper (lens-embed).
+
+A single pinned model across the family means this vector is comparable to
+other members' vectors — the basis for cross-artefact and cohort-distinctiveness
+signals computed downstream. Opt-in and degradable: install the [embeddings]
+extra to populate it; without it (or on any failure) this returns None.
+"""
+
+from __future__ import annotations
+
+
+def embed_document(text: str) -> list[float] | None:
+    """Pooled, L2-normalised vector, or None if embeddings are off."""
+    if not text or not text.strip():
+        return None
+    try:
+        from lens_embed import backend_available, embed_long_text
+    except ImportError:
+        return None
+    if not backend_available("text"):
+        return None
+    try:
+        return embed_long_text(text)
+    except Exception:
+        return None

code_analyser-1.3.0/src/code_analyser/manifest.py

@@ -0,0 +1,12 @@
+"""Capability manifest for the lens family (consumed by auto-analyser)."""
+from __future__ import annotations
+
+from lens_contract import make_manifest
+
+MANIFEST = make_manifest(
+    name="code-analyser",
+    accepts=["code"],
+    extensions=[".py", ".js", ".ts", ".tsx", ".jsx", ".html", ".css", ".scss", ".sql", ".ipynb"],
+    auto_routable=True,
+    produces="CodeAnalysis",
+)

{code_analyser-1.1.0 → code_analyser-1.3.0}/src/code_analyser/models.py

@@ -191,3 +191,6 @@ class CodeAnalysis(BaseModel):
     files: list[FileAnalysis]
     cross_file: CrossFileSignals
     llm_signals: TopLevelLLMSignals | None = None
+    # Pooled, L2-normalised source vector from lens-embed (pinned all-MiniLM-L6-v2).
+    # Comparable across members; None unless [embeddings] installed.
+    embedding: list[float] | None = None

{code_analyser-1.1.0 → code_analyser-1.3.0}/src/code_analyser/pipeline.py

@@ -3,6 +3,7 @@ import zipfile
 from pathlib import Path
 
 from .detect import detect_language
+from .embedding import embed_document
 from .models import (
     CodeAnalysis, CrossFileSignals, FileAnalysis,
     FileLLMSignals, TopLevelLLMSignals,
@@ -79,6 +80,9 @@ class CodeAnalyser:
             frameworks_detected=sorted(all_frameworks),
         )
 
+        combined_source = "\n\n".join(_decode(content) for _, content in pairs)
+        embedding = embed_document(combined_source)
+
         return CodeAnalysis(
             input=path.name,
             file_count=len(files),
@@ -86,6 +90,7 @@ class CodeAnalyser:
             files=files,
             cross_file=cross,
             llm_signals=llm_top,
+            embedding=embedding,
         )
 
 

code_analyser-1.3.0/tests/test_embedding.py

@@ -0,0 +1,25 @@
+"""Code embedding wiring — field presence + graceful degradation."""
+from __future__ import annotations
+import importlib.util
+import pytest
+from code_analyser.embedding import embed_document
+from code_analyser.models import CodeAnalysis
+
+_TEXT = importlib.util.find_spec("lens_embed") is not None and importlib.util.find_spec("sentence_transformers") is not None
+
+def test_field_default_none():
+    assert "embedding" in CodeAnalysis.model_fields
+    assert CodeAnalysis.model_fields["embedding"].default is None
+
+def test_empty_is_none():
+    assert embed_document("") is None
+    assert embed_document("  \n ") is None
+
+@pytest.mark.skipif(_TEXT, reason="embeddings extra installed")
+def test_none_without_backend():
+    assert embed_document("def add(a, b):\n    return a + b\n") is None
+
+@pytest.mark.skipif(not _TEXT, reason="needs [embeddings]")
+def test_vector_with_backend():
+    v = embed_document("def add(a, b):\n    return a + b\n\n" * 5)
+    assert isinstance(v, list) and len(v) == 384