sourcecode 1.30.10__tar.gz → 1.30.12__tar.gz

sourcecode-1.30.12/.continue-here.md

@@ -0,0 +1,166 @@
+# Continue Here — atlas-cli sesión 11
+
+**Paused:** 2026-05-16 (sesión 11)
+**Repo:** `/Users/user/Downloads/atlas-cli`
+**Branch:** master
+**Último commit:** `dc157fa` — (fix) mejorando honestidad prepare-context pr
+**Working tree:** `prepare_context.py` modificado (semantic engine fixes — NO committed)
+
+---
+
+## Trabajo de esta sesión
+
+### Bloque 1: Strict Diff Source Model
+
+**Qué se hizo:**
+- `DiffSourceType` enum añadido (`WORKTREE_UNSTAGED`, `WORKTREE_STAGED`, `GIT_SINCE_REF`, `GIT_RANGE`)
+- `_get_pr_scope_files` reescrito → retorna 4-tuple `(all_files, scope_source, committed_files, uncommitted_files)`. Sin fallback a HEAD~1. Committed solo si `--since` explícito.
+- `_get_git_changed_files` → sin default HEAD~1. Sin `--since`: WORKTREE_UNSTAGED (`git diff --name-only`). Con `--since`: GIT_SINCE_REF.
+- `_FOLDER_TYPE_MAP` eliminado — path/folder no es evidencia.
+- `TaskOutput` nuevos campos: `committed_changes`, `uncommitted_changes`, `analysis_scope`
+- Cada entry `runtime_changes` incluye `diff_source` (GIT_RANGE vs WORKTREE_UNSTAGED)
+- `changed-only` en cli.py: excluye untracked, solo staged+unstaged
+- Tests `test_v1_10_regressions.py`: mocks actualizados a 4-tuple
+
+**Committed en:** `dc157fa`
+
+---
+
+### Bloque 2: Evidence-Gated Semantic Engine (NO COMMITTED)
+
+**Fichero con cambios:** `src/sourcecode/prepare_context.py` (unstaged)
+
+**Qué se cambió:**
+
+1. **`_ARTIFACT_CHANGE_TYPES["source"]` → `[]`**
+   - Antes: `["behavioral_change"]`. Ahora: ninguno.
+   - `behavioral_change` solo se emite si `diff_severities` devuelve `api_change` o `security_change`
+
+2. **`_DIFF_SEVERITY_CHANGE_TYPES` nuevo dict**
+   - `api_change` → `["behavioral_change"]`
+   - `security_change` → `["behavioral_change", "security_change"]`
+   - `field_change` → `["structural_change"]`
+   - `unknown/trivial` → `[]`
+
+3. **`_ATYPE_IMPACT_AREA["source"]` → `"application_layer"` (antes: `"transaction_boundary"`)**
+   - `"service"` también → `"application_layer"` (sin evidencia transaccional)
+   - `"transaction_boundary"` ya no se emite para archivos no clasificados
+
+4. **`_role_in_system` default → `"unclassified"` (antes: `"core_service"`)**
+   - `"service"` anotado → `"service"` (no `"core_service"`)
+
+5. **`_structured_why` evidence-gated**
+   - Solo emite `role_in_system`, `impact_area`, `propagation_risk` si tipo no es `source`/`unclassified`
+   - Añade `classification_confidence`
+
+6. **`_score_to_confidence(score)` helper**
+   - `>= 0.60` → `"high"`, `>= 0.40` → `"medium"`, else → `"low"`
+
+7. **`impact_score_per_file` ahora es dict estructurado**
+   ```json
+   {
+     "score": 0.145,
+     "confidence": "low",
+     "score_breakdown": { "artifact_base": 0.125, ... },
+     "change_types": [],
+     "diff_severity": "trivial"
+   }
+   ```
+
+8. **`behavioral_changes`** — gateado en `diff_severities`, no en artifact type
+
+9. **`changed_subsystems`** — excluye `source`/`unclassified`; eliminado `"transaction_layer"` del mapa
+
+10. **`dependency_graph_summary.propagation_depth`** → `None` cuando no hay graph edges (no `0`)
+
+11. **`_runtime_impact`** — eliminado mensaje de "verify transaction scope" para service sin evidencia anotacional
+
+**Resultado verificado:**
+```
+changed_subsystems: []
+behavioral_changes: []
+role: unclassified
+why: artifact_type: source | classification_confidence: low | change_effect: ...
+# Sin: core_service, transaction_boundary, transaction_layer, propagation_risk
+```
+
+**Tests:** 765 passed, 3 skipped (sin nuevos fallos)
+
+---
+
+## Estado de tests
+
+```
+765 passed, 3 skipped
+```
+
+Pre-existing failures (no relacionados):
+- `tests/test_block2_coverage.py` — DocRecord bug en `doc_analyzer.py`
+- `tests/test_dependency_analyzer_node_python.py::test_python_requirements_without_lockfile_keeps_declared_versions`
+
+---
+
+## PENDIENTE (antes de próximo commit)
+
+1. **Commit semantic engine fixes** en `prepare_context.py`
+   ```bash
+   git add src/sourcecode/prepare_context.py
+   git commit -m "fix: evidence-gated semantic engine — remove unsupported behavioral/transaction claims"
+   ```
+
+2. **Smoke test en repo spring-boot real**
+   ```bash
+   cd ~/Documents/workspace/spring-boot-realworld-example-app
+   sourcecode prepare-context delta --since HEAD~1 2>/dev/null | python3 -c "
+   import json, sys
+   d = json.load(sys.stdin)
+   si = d.get('system_impact', {})
+   print('subsystems:', si.get('changed_subsystems'))
+   print('behavioral:', si.get('behavioral_changes'))
+   for f, v in d.get('impact_score_per_file', {}).items():
+       print(f'{f}: conf={v[\"confidence\"]} sev={v[\"diff_severity\"]}')
+   "
+   ```
+
+3. **Smoke test behavioral_impact** (pendiente de sesión anterior):
+   ```bash
+   cd ~/Documents/workspace/spring-boot-realworld-example-app
+   sourcecode prepare-context review-pr . 2>/dev/null | python3 -c "
+   import json, sys
+   d = json.load(sys.stdin)
+   for bi in d.get('behavioral_impact', []):
+       print('entry:', bi['entry_point'])
+       print('confidence:', bi['confidence'])
+       print('trace:', bi['trace'])
+       print()
+   "
+   ```
+
+4. **Tests unitarios para `_classify_diff_severity`** — aún sin cobertura
+
+5. **`User.java` clasificado como `"source"` no `"domain_model"`** — fix opcional
+
+6. **Evaluar si `execution_paths` sigue siendo útil** o si `behavioral_impact` lo reemplaza
+
+---
+
+## Para retomar
+
+```bash
+cd /Users/user/Downloads/atlas-cli
+git status  # debe mostrar: prepare_context.py modified
+
+python3 -m pytest tests/ \
+  --ignore=tests/test_block2_coverage.py \
+  --ignore=tests/test_dependency_analyzer_node_python.py \
+  -q
+# Expected: 765 passed, 3 skipped
+
+# Luego commit el semantic engine:
+git add src/sourcecode/prepare_context.py
+git commit -m "fix: evidence-gated semantic engine — remove unsupported behavioral/transaction claims"
+```
+
+---
+
+*Pausado 2026-05-16 sesión 11 — gsd:pause-work*

{sourcecode-1.30.10 → sourcecode-1.30.12}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sourcecode
-Version: 1.30.10
+Version: 1.30.12
 Summary: Deterministic codebase context for AI coding agents
 License:                                  Apache License
                                    Version 2.0, January 2004
@@ -221,7 +221,7 @@ Description-Content-Type: text/markdown
 
 **Deterministic, behavior-aware codebase context for AI agents and PR review.**
 
-![Version](https://img.shields.io/badge/version-1.30.10-blue)
+![Version](https://img.shields.io/badge/version-1.30.12-blue)
 ![Python](https://img.shields.io/badge/python-3.10%2B-green)
 
 ---
@@ -257,7 +257,7 @@ pipx install sourcecode
 
 ```bash
 sourcecode version
-# sourcecode 1.30.10
+# sourcecode 1.30.12
 ```
 
 ---

{sourcecode-1.30.10 → sourcecode-1.30.12}/README.md

@@ -2,7 +2,7 @@
 
 **Deterministic, behavior-aware codebase context for AI agents and PR review.**
 
-![Version](https://img.shields.io/badge/version-1.30.10-blue)
+![Version](https://img.shields.io/badge/version-1.30.12-blue)
 ![Python](https://img.shields.io/badge/python-3.10%2B-green)
 
 ---
@@ -38,7 +38,7 @@ pipx install sourcecode
 
 ```bash
 sourcecode version
-# sourcecode 1.30.10
+# sourcecode 1.30.12
 ```
 
 ---

{sourcecode-1.30.10 → sourcecode-1.30.12}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "sourcecode"
-version = "1.30.10"
+version = "1.30.12"
 description = "Deterministic codebase context for AI coding agents"
 readme = "README.md"
 requires-python = ">=3.9"

{sourcecode-1.30.10 → sourcecode-1.30.12}/src/sourcecode/__init__.py

@@ -1,3 +1,3 @@
 """sourcecode — Deterministic codebase context maps for AI coding agents."""
 
-__version__ = "1.30.10"
+__version__ = "1.30.12"

{sourcecode-1.30.10 → sourcecode-1.30.12}/src/sourcecode/cli.py

@@ -1400,9 +1400,8 @@ def main(
             if _gc_early and not (_bad_gc & set(_gc_early.limitations)):
                 _uc = _gc_early.uncommitted_changes
                 if _uc:
-                    _allowed_changed_files = (
-                        set(_uc.staged) | set(_uc.unstaged) | set(_uc.untracked)
-                    )
+                    # WORKTREE_UNSTAGED + WORKTREE_STAGED only; untracked excluded
+                    _allowed_changed_files = set(_uc.staged) | set(_uc.unstaged)
             if not _allowed_changed_files:
                 typer.echo(
                     "[changed-only] git unavailable or no uncommitted changes — falling back to full scan.",
@@ -1953,6 +1952,89 @@ def telemetry_disable() -> None:
     typer.echo("Re-enable at any time: sourcecode telemetry enable")
 
 
+# ── repo-ir ───────────────────────────────────────────────────────────────────
+
+@app.command("repo-ir")
+def repo_ir_cmd(
+    path: Path = typer.Argument(
+        Path("."),
+        help="Repository root to analyze (default: current directory)",
+    ),
+    since: Optional[str] = typer.Option(
+        None,
+        "--since",
+        help="Git ref for symbol-level diff (e.g. HEAD~1, main)",
+    ),
+    files: Optional[str] = typer.Option(
+        None,
+        "--files",
+        help="Comma-separated list of Java files (relative to path) to analyze",
+    ),
+    output_path: Optional[Path] = typer.Option(
+        None,
+        "--output",
+        "-o",
+        help="Write output to a file instead of stdout",
+    ),
+    include_tests: bool = typer.Option(
+        False,
+        "--include-tests",
+        help="Include test files in analysis (excluded by default)",
+    ),
+) -> None:
+    """Deterministic symbol-level IR for Java repositories.
+
+    \b
+    Extracts symbols, relations, Spring roles, and (with --since) symbol-level diffs.
+    Output is JSON: symbols[], relations[], changed_symbols[], spring_summary, graph_metadata.
+
+    \b
+    Examples:
+      sourcecode repo-ir
+      sourcecode repo-ir /path/to/repo --since HEAD~1
+      sourcecode repo-ir --files src/main/java/UserService.java
+      sourcecode repo-ir --since main --output ir.json
+    """
+    import json as _json
+    import sys as _sys
+
+    from sourcecode.repository_ir import build_repo_ir, find_java_files
+
+    root = path.resolve()
+    if not root.is_dir():
+        typer.echo(f"Error: {root} is not a directory", err=True)
+        raise typer.Exit(1)
+
+    if files:
+        file_list = [f.strip() for f in files.split(",") if f.strip()]
+    else:
+        file_list = find_java_files(root)
+        if not include_tests:
+            file_list = [f for f in file_list if "/test/" not in f and "/tests/" not in f]
+
+    if not file_list:
+        typer.echo(
+            _json.dumps({
+                "symbols": [],
+                "relations": [],
+                "changed_symbols": [],
+                "spring_summary": {},
+                "graph_metadata": {"node_count": 0, "edge_count": 0, "has_call_graph": False},
+            }, indent=2)
+        )
+        return
+
+    ir = build_repo_ir(file_list, root, since=since)
+    output = _json.dumps(ir, indent=2, ensure_ascii=False)
+
+    if output_path:
+        output_path.write_text(output, encoding="utf-8")
+        typer.echo(f"IR written to {output_path}", err=True)
+    else:
+        _sys.stdout.write(output)
+        _sys.stdout.write("\n")
+
+
 # ── version ───────────────────────────────────────────────────────────────────
 
 @app.command("version")