sourcecode 1.30.21__tar.gz → 1.30.23__tar.gz

{sourcecode-1.30.21 → sourcecode-1.30.23}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sourcecode
-Version: 1.30.21
+Version: 1.30.23
 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-1.30.21-blue)
+![Version](https://img.shields.io/badge/version-1.30.23-blue)
 ![Python](https://img.shields.io/badge/python-3.10%2B-green)
 
 ---
@@ -257,7 +257,7 @@ pipx install sourcecode
 
 ```bash
 sourcecode version
-# sourcecode 1.30.21
+# sourcecode 1.30.23
 ```
 
 ---

{sourcecode-1.30.21 → sourcecode-1.30.23}/README.md

@@ -2,7 +2,7 @@
 
 **Deterministic, behavior-aware codebase context for AI agents and PR review.**
 
-![Version](https://img.shields.io/badge/version-1.30.21-blue)
+![Version](https://img.shields.io/badge/version-1.30.23-blue)
 ![Python](https://img.shields.io/badge/python-3.10%2B-green)
 
 ---
@@ -38,7 +38,7 @@ pipx install sourcecode
 
 ```bash
 sourcecode version
-# sourcecode 1.30.21
+# sourcecode 1.30.23
 ```
 
 ---

{sourcecode-1.30.21 → sourcecode-1.30.23}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "sourcecode"
-version = "1.30.21"
+version = "1.30.23"
 description = "Deterministic codebase context for AI coding agents"
 readme = "README.md"
 requires-python = ">=3.9"

{sourcecode-1.30.21 → sourcecode-1.30.23}/src/sourcecode/__init__.py

@@ -1,3 +1,3 @@
 """sourcecode — Deterministic codebase context maps for AI coding agents."""
 
-__version__ = "1.30.21"
+__version__ = "1.30.23"

{sourcecode-1.30.21 → sourcecode-1.30.23}/src/sourcecode/cli.py

@@ -1610,11 +1610,10 @@ def _serialize_relevant_file(f: Any) -> dict:
     d = {k: v for k, v in _asdict(f).items() if v != "" and v is not None}
     reason = d.pop("reason", "") or ""
     why = d.pop("why", "") or ""
+    d.pop("score", None)  # score removed from public output (internal ranking only)
     explanation = _make_explanation(reason, why)
     if explanation:
         d["explanation"] = explanation
-    if reason:
-        d["reason"] = reason
     return d
 
 
@@ -1882,9 +1881,14 @@ def prepare_context_cmd(
     def _task_include(field: str) -> bool:
         return _content_filter.get(field, True)
 
+    _run_id = hashlib.sha256(
+        f"{task}:{target}:{since or ''}:{format or ''}:{symptom or ''}".encode()
+    ).hexdigest()[:16]
+
     out: dict[str, Any] = {
         "task": output.task,
         "goal": output.goal,
+        "run_id": _run_id,
     }
     if _task_include("project_summary"):
         out["project_summary"] = output.project_summary
@@ -1935,6 +1939,30 @@ def prepare_context_cmd(
                 _sys.stdout.buffer.write(b"\n")
                 _sys.stdout.buffer.flush()
             raise typer.Exit(code=1)
+        if output.ci_decision == "no_changes":
+            # Early exit: no diff — emit minimal JSON without any analysis fields.
+            # Prevents no_changes from ever being serialized alongside relevant_files > 0.
+            _nc_out: dict[str, Any] = {
+                "task": output.task,
+                "ci_decision": "no_changes",
+                "summary": "No changes detected",
+            }
+            if output.since:
+                _nc_out["since"] = output.since
+            if output.analysis_scope:
+                _nc_out["analysis_scope"] = output.analysis_scope
+            _nc_json = json.dumps(_nc_out, indent=2, ensure_ascii=False)
+            if output_path is not None:
+                output_path.write_text(_nc_json, encoding="utf-8")
+            else:
+                import sys as _sys
+                _sys.stdout.buffer.write(_nc_json.encode("utf-8"))
+                _sys.stdout.buffer.write(b"\n")
+                _sys.stdout.buffer.flush()
+            if copy:
+                _copy_to_clipboard(_nc_json)
+                typer.echo("✓ copied to clipboard", err=True)
+            raise typer.Exit()
         if output.ci_decision:
             out["ci_decision"] = output.ci_decision
         if output.since:

{sourcecode-1.30.21 → sourcecode-1.30.23}/src/sourcecode/prepare_context.py

@@ -25,6 +25,7 @@ class DiffSourceType(str, Enum):
     WORKTREE_STAGED   = "WORKTREE_STAGED"     # git diff --cached
     GIT_SINCE_REF     = "GIT_SINCE_REF"       # git diff ref HEAD
     GIT_RANGE         = "GIT_RANGE"           # git diff refA refB
+    HEAD_MINUS_1      = "HEAD_MINUS_1"        # git diff HEAD~1 HEAD (auto-fallback when tree clean)
 
 
 # ─────────────────────────────────────────────────────────────────────────────
@@ -377,6 +378,25 @@ class TaskOutput:
     analysis_scope: dict = field(default_factory=dict)
 
 
+@dataclass
+class CanonicalAnalysisIR:
+    """Shared intermediate representation produced by _build_delta_impact.
+
+    Both delta and review-pr derive their task-specific output from this IR.
+    Never recalculate logic per command — render views from this single object.
+    """
+    relevant_files: list[RelevantFile]
+    impact_summary: str
+    affected_modules: list[str]
+    risk_areas: list[dict]
+    why_these_files: dict[str, str]
+    analysis_gaps: list[str]
+    system_impact: dict
+    change_type: list[str]
+    dependency_graph_summary: dict
+    impact_score_per_file: dict
+
+
 # ─────────────────────────────────────────────────────────────────────────────
 # Builder
 # ─────────────────────────────────────────────────────────────────────────────
@@ -582,6 +602,83 @@ _FRONTEND_SYMPTOM_MAP: dict[str, list[str]] = {
 }
 
 
+def _build_analysis_scope(
+    *,
+    task_name: str,
+    since: Optional[str],
+    delta_baseline: dict,
+    pr_scope_source: str,
+    pr_uncommitted_files: list,
+) -> dict:
+    """Build analysis_scope metadata from actual git resolution — never hardcoded."""
+    if task_name == "review-pr":
+        sources = pr_scope_source.split(",") if pr_scope_source else []
+        # Derive git_equivalent_command from sources
+        cmds: list[str] = []
+        if DiffSourceType.GIT_RANGE.value in sources or DiffSourceType.GIT_SINCE_REF.value in sources:
+            cmds.append(f"git diff --name-only {since} HEAD")
+        if DiffSourceType.HEAD_MINUS_1.value in sources:
+            cmds.append("git diff --name-only HEAD~1 HEAD")
+        if DiffSourceType.WORKTREE_UNSTAGED.value in sources:
+            cmds.append("git diff --name-only")
+        if DiffSourceType.WORKTREE_STAGED.value in sources:
+            cmds.append("git diff --name-only --cached")
+        git_cmd = " + ".join(cmds) if cmds else "git diff --name-only"
+        return {
+            "sources_used": sources,
+            "git_equivalent_command": git_cmd,
+            "includes_uncommitted": bool(pr_uncommitted_files),
+            "git_resolution": {
+                "source_of_truth": "RANGE" if since else (
+                    "HEAD" if DiffSourceType.HEAD_MINUS_1.value in sources else "WORKTREE"
+                ),
+                "resolved_diff_strategy": pr_scope_source,
+                "commit_count_analyzed": 1 if DiffSourceType.HEAD_MINUS_1.value in sources else (
+                    0 if not since else None
+                ),
+            },
+        }
+
+    # delta
+    rpath = delta_baseline.get("resolution_path", "")
+    rref = delta_baseline.get("resolved_ref", "HEAD")
+
+    _COMMITTED_PATHS = {"exact_local_ref", "remote_tracking_ref", "symbolic_ref"}
+    if rpath in _COMMITTED_PATHS:
+        sources = [DiffSourceType.GIT_SINCE_REF.value]
+        git_cmd = f"git diff --name-only {rref} HEAD"
+        source_of_truth = "RANGE"
+        includes_uncommitted = False
+    elif rpath == "head_minus_1_fallback":
+        sources = [DiffSourceType.HEAD_MINUS_1.value]
+        git_cmd = "git diff --name-only HEAD~1 HEAD"
+        source_of_truth = "HEAD"
+        includes_uncommitted = False
+    elif rpath == "uncommitted_staged":
+        sources = [DiffSourceType.WORKTREE_STAGED.value]
+        git_cmd = "git diff --name-only --cached"
+        source_of_truth = "STAGED"
+        includes_uncommitted = True
+    else:  # uncommitted_unstaged, uncommitted_changes, no_changes_confirmed
+        sources = [DiffSourceType.WORKTREE_UNSTAGED.value]
+        git_cmd = "git diff --name-only"
+        source_of_truth = "WORKTREE"
+        includes_uncommitted = True
+
+    return {
+        "sources_used": sources,
+        "git_equivalent_command": git_cmd,
+        "includes_uncommitted": includes_uncommitted,
+        "git_resolution": {
+            "source_of_truth": source_of_truth,
+            "resolved_diff_strategy": rpath or "unknown",
+            "commit_count_analyzed": 1 if rpath == "head_minus_1_fallback" else (
+                0 if rpath == "no_changes_confirmed" else None
+            ),
+        },
+    }
+
+
 class TaskContextBuilder:
     def __init__(self, root: Path) -> None:
         self.root = root
@@ -851,9 +948,11 @@ class TaskContextBuilder:
         # For delta task, relevant_files must rank only files changed in the
         # specified git range, not the full repo by generic entrypoint scoring.
         _delta_files: Optional[set[str]] = None
+        _delta_baseline: dict = {}  # resolution metadata threaded into analysis_scope
         if task_name == "delta":
-            _delta_raw = self._get_git_changed_files(since=since)
-            if _delta_raw is None:
+            _baseline = self._resolve_git_baseline(since=since)
+            _delta_baseline = _baseline
+            if _baseline["error"]:
                 # Explicit --since ref couldn't be resolved — hard error, no fallback
                 _avail_branches, _suggested = self._get_available_refs(since or "")
                 _hints: list[str] = []
@@ -880,8 +979,11 @@ class TaskContextBuilder:
                     error_hints=_hints,
                     gaps=[f"Cannot compute delta: git ref '{since}' not found."] + _hints,
                     ci_decision="git_ref_error",
+                    resolution_path=_baseline.get("resolution_path"),
+                    diff_validation_status=_baseline.get("diff_validation_status"),
                 )
-            elif _delta_raw:
+            _delta_raw = _baseline["files"]
+            if _delta_raw:
                 _delta_files = set(_delta_raw)
 
         # ── 5d. review-pr: set _delta_files from pre-resolved git scope ──────────
@@ -934,7 +1036,8 @@ class TaskContextBuilder:
         test_set = {p for p in all_paths if self._is_test(p)}
         source_set = {p for p in all_paths if not self._is_test(p) and self._is_source(p)}
 
-        # Delta uses a dedicated impact-analysis path — never the generic ranker.
+        # Delta and review-pr share CanonicalAnalysisIR — computed once, rendered per task.
+        _ir: Optional[CanonicalAnalysisIR] = None
         _delta_impact_summary: Optional[str] = None
         _delta_affected_modules: list[str] = []
         _delta_risk_areas: list[dict] = []
@@ -947,23 +1050,22 @@ class TaskContextBuilder:
 
         if task_name in ("delta", "review-pr"):
             _delta_changed_list: list[str] = sorted(_delta_files) if _delta_files else []
-            (
-                relevant_files,
-                _delta_impact_summary,
-                _delta_affected_modules,
-                _delta_risk_areas,
-                _delta_why,
-                _delta_analysis_gaps,
-                _delta_system_impact,
-                _delta_change_type,
-                _delta_dep_graph_summary,
-                _delta_impact_score_per_file,
-            ) = self._build_delta_impact(
+            _ir = self._build_delta_impact(
                 changed_files=_delta_changed_list,
                 all_paths=all_paths,
                 entry_points=entry_points,
                 since=since,
             )
+            relevant_files = _ir.relevant_files
+            _delta_impact_summary = _ir.impact_summary
+            _delta_affected_modules = _ir.affected_modules
+            _delta_risk_areas = _ir.risk_areas
+            _delta_why = _ir.why_these_files
+            _delta_analysis_gaps = _ir.analysis_gaps
+            _delta_system_impact = _ir.system_impact
+            _delta_change_type = _ir.change_type
+            _delta_dep_graph_summary = _ir.dependency_graph_summary
+            _delta_impact_score_per_file = _ir.impact_score_per_file
         else:
             relevant_files = self._rank_files(
                 task_name, spec, all_paths, entry_set, test_set,
@@ -1496,7 +1598,11 @@ class TaskContextBuilder:
         changed_files: list[str] = []
         affected_entry_points: list[str] = []
         if task_name in ("delta", "review-pr"):
-            changed_files = sorted(_delta_files) if _delta_files else (self._get_git_changed_files(since=since) or [])
+            # For delta: _delta_files already resolved via _resolve_git_baseline — no second git call.
+            # For review-pr: _get_git_changed_files fallback still valid as last resort.
+            changed_files = sorted(_delta_files) if _delta_files else (
+                [] if task_name == "delta" else (self._get_git_changed_files(since=since) or [])
+            )
             _ep_set = {ep.path for ep in entry_points}
             # include framework-detected entry points AND files classified as
             # entrypoint/controller/security by artifact taxonomy
@@ -1563,21 +1669,16 @@ class TaskContextBuilder:
             committed_changes=_pr_committed_changes,
             uncommitted_changes=_pr_uncommitted_changes,
             # transparency: explicit diff scope
-            analysis_scope={
-                "sources_used": _pr_scope_source.split(",") if task_name == "review-pr" and _pr_scope_source else (
-                    [DiffSourceType.GIT_SINCE_REF.value] if task_name == "delta" and since else
-                    [DiffSourceType.WORKTREE_UNSTAGED.value] if task_name == "delta" else
-                    []
-                ),
-                "git_equivalent_command": (
-                    f"git diff --name-only {since} HEAD" if since else
-                    "git diff --name-only"
-                ) if task_name in ("delta", "review-pr") else None,
-                "includes_uncommitted": bool(
-                    (task_name == "review-pr" and _pr_uncommitted_files) or
-                    (task_name == "delta" and not since)
-                ),
-            } if task_name in ("delta", "review-pr") else {},
+            analysis_scope=_build_analysis_scope(
+                task_name=task_name,
+                since=since,
+                delta_baseline=_delta_baseline,
+                pr_scope_source=_pr_scope_source,
+                pr_uncommitted_files=_pr_uncommitted_files,
+            ) if task_name in ("delta", "review-pr") else {},
+            resolved_since_ref=_delta_baseline.get("resolved_ref") if task_name == "delta" else None,
+            resolution_path=_delta_baseline.get("resolution_path") if task_name == "delta" else None,
+            diff_validation_status=_delta_baseline.get("diff_validation_status") if task_name == "delta" else None,
         )
 
     def render_prompt(self, output: TaskOutput) -> str:
@@ -1943,6 +2044,17 @@ class TaskContextBuilder:
             if DiffSourceType.WORKTREE_STAGED.value not in sources:
                 sources.append(DiffSourceType.WORKTREE_STAGED.value)
 
+        # ── HEAD~1 fallback — working tree clean, surface last commit ────────
+        # Without --since: if no unstaged/staged changes exist, fall back to the
+        # last committed diff so a clean tree never silently returns no_changes.
+        if since is None and not committed_files and not uncommitted_files:
+            head_ok = _run("git", "rev-parse", "--verify", "HEAD~1")
+            if head_ok is not None:  # HEAD~1 exists
+                head_committed = _run("git", "diff", "--name-only", "--relative", "HEAD~1", "HEAD")
+                if head_committed:
+                    committed_files = head_committed
+                    sources.append(DiffSourceType.HEAD_MINUS_1.value)
+
         # ── Drop paths outside self.root ──────────────────────────────────────
         def _drop_outside(lst: list[str]) -> list[str]:
             return [f for f in lst if not f.startswith("../") and not f.startswith("..\\")]
@@ -2297,13 +2409,10 @@ class TaskContextBuilder:
         all_paths: list[str],
         entry_points: list,
         since: Optional[str],
-    ) -> tuple[list[RelevantFile], str, list[str], list[dict[str, Any]], dict[str, str], list[str]]:
+    ) -> "CanonicalAnalysisIR":
         """Build incremental impact analysis for changed files.
 
-        Returns:
-            (relevant_files, impact_summary, affected_modules, risk_areas,
-             why_these_files, analysis_gaps)
-
+        Returns CanonicalAnalysisIR — the shared IR consumed by delta and review-pr views.
         Changed files are always included in relevant_files (never dropped by score).
         Related files are expanded type-aware: controller→service→repository→mapper chain.
         Scoring is hierarchical by artifact_type, not by heuristic impact_level.
@@ -2471,6 +2580,18 @@ class TaskContextBuilder:
                 return "data_access"
             if atype == "service":
                 return "service"   # annotation-confirmed — not "core_service" (overclaim)
+            # Name-pattern heuristics — only when atype alone gave no verdict.
+            # These are INFERRED (LOW CONFIDENCE) — stem match, not annotation evidence.
+            if any(kw in stem_lower for kw in ("validator", "validation")):
+                return "validation_component"
+            if any(kw in stem_lower for kw in ("controller", "resource", "endpoint", "rest")):
+                return "external_interface"
+            if any(kw in stem_lower for kw in ("service", "svc", "usecase", "facade")):
+                return "service"
+            if any(kw in stem_lower for kw in ("repository", "repo", "dao", "store")):
+                return "data_access"
+            if any(kw in stem_lower for kw in ("config", "configuration", "settings", "properties")):
+                return "configuration"
             return "unclassified"   # no role claim without evidence
 
         def _structured_why(path: str, atype: str, module: str, role: str, risk_areas: list[str], cls_confidence: str = "low") -> str:
@@ -2491,17 +2612,17 @@ class TaskContextBuilder:
             return " | ".join(parts)
 
         if not changed_files:
-            return (
-                [],
-                "No changes detected — verify the git ref passed to --since",
-                [],
-                [],
-                {},
-                ["No changed files found. Check that --since ref exists and the diff is non-empty."],
-                {},   # system_impact
-                [],   # change_type
-                {"edges": [], "propagation_depth": 0},  # dependency_graph_summary
-                {},   # impact_score_per_file
+            return CanonicalAnalysisIR(
+                relevant_files=[],
+                impact_summary="No changes detected — verify the git ref passed to --since",
+                affected_modules=[],
+                risk_areas=[],
+                why_these_files={},
+                analysis_gaps=["No changed files found. Check that --since ref exists and the diff is non-empty."],
+                system_impact={},
+                change_type=[],
+                dependency_graph_summary={"edges": [], "propagation_depth": 0},
+                impact_score_per_file={},
             )
 
         ep_paths = {ep.path for ep in entry_points}
@@ -3026,17 +3147,17 @@ class TaskContextBuilder:
                 " — related file expansion uses directory proximity only"
             )
 
-        return (
-            relevant,
-            impact_summary,
-            sorted(affected_modules_set),
-            risk_areas_out,
-            why,
-            analysis_gaps,
-            system_impact,
-            aggregate_change_type,
-            dependency_graph_summary,
-            impact_score_per_file,
+        return CanonicalAnalysisIR(
+            relevant_files=relevant,
+            impact_summary=impact_summary,
+            affected_modules=sorted(affected_modules_set),
+            risk_areas=risk_areas_out,
+            why_these_files=why,
+            analysis_gaps=analysis_gaps,
+            system_impact=system_impact,
+            change_type=aggregate_change_type,
+            dependency_graph_summary=dependency_graph_summary,
+            impact_score_per_file=impact_score_per_file,
         )
 
     def _resolve_git_baseline(self, since: Optional[str]) -> dict[str, Any]:
@@ -3136,25 +3257,33 @@ class TaskContextBuilder:
             }
 
         else:
-            # No since: uncommitted changes first
+            # No since: unstaged → staged → HEAD~1 — never error, clean tree is valid
             ok, out = _run("diff", "--name-only", "--relative", timeout=10)
             if ok:
                 files = [line.strip() for line in out.splitlines() if line.strip()]
                 if files:
-                    return _make(files, "HEAD", "uncommitted_changes")
+                    return _make(files, "HEAD", "uncommitted_unstaged")
 
-            # HEAD~1 fallback
+            # Staged (committed to index but not yet to history)
+            ok, out = _run("diff", "--name-only", "--cached", "--relative", timeout=10)
+            if ok:
+                files = [line.strip() for line in out.splitlines() if line.strip()]
+                if files:
+                    return _make(files, "HEAD", "uncommitted_staged")
+
+            # HEAD~1 fallback — working tree clean, surface last commit
             if _verify("HEAD~1"):
                 files = _diff("HEAD~1")
                 if files is not None:
                     return _make(files or [], "HEAD~1", "head_minus_1_fallback")
 
+            # Confirmed no changes: first commit or empty repo
             return {
                 "files": [],
                 "resolved_ref": "HEAD",
-                "resolution_path": "unresolvable",
-                "diff_validation_status": "invalid_ref",
-                "error": True,
+                "resolution_path": "no_changes_confirmed",
+                "diff_validation_status": "valid_empty",
+                "error": False,
             }
 
     def _get_uncommitted_changed_files(self) -> list[str]: