Remove bank analyzer document service circular import

app/modules/bank_statement_analyzer/service.py

@@ -11,8 +11,6 @@ from typing import Iterable
 
 from fastapi import UploadFile
 
-from app.modules.documents.services import DEFAULT_STORAGE_ROOT
-
 from .analyzer import analyze_files, export_excel
 
 ALLOWED_ROLES = {"Partner", "Manager", "Branch Manager", "Staff", "Employee", "Consultant"}
@@ -22,18 +20,34 @@ RETENTION_HOURS = int(os.getenv("BANK_ANALYZER_FAILED_RETENTION_HOURS", "24"))
 _SAFE = re.compile(r"[^A-Za-z0-9._-]+")
 
 
+def _resolve_document_storage_root() -> Path:
+    """Resolve the same automatic document root without importing documents.services.
+
+    Importing app.modules.documents.services here creates a circular import through
+    clients.__init__ -> clients.ui -> documents.services. This local resolver
+    intentionally mirrors the existing document module's storage-root rules while
+    keeping the bank analyzer independent of document models and UI modules.
+    """
+    configured = (os.getenv("DOCUMENT_STORAGE_ROOT") or "").strip()
+    if configured:
+        configured_path = Path(configured).expanduser()
+        if configured_path.is_absolute():
+            return configured_path
+        return (Path.cwd() / configured_path).resolve()
+
+    if os.name == "nt":
+        anchor = Path.cwd().anchor or (os.environ.get("SystemDrive", "C:") + "\\")
+        return Path(anchor) / "AuditFirmERPDocuments" / "engagement_documents"
+
+    return (Path.cwd() / "documents" / "engagement_documents").resolve()
+
+
 def _now() -> datetime:
     return datetime.now(timezone.utc)
 
 
 def _root() -> Path:
-    """Return the automatic ERP work-storage root.
-
-    The document module already resolves its storage location for the current
-    deployment.  Bank-statement jobs use a sibling ``Work`` folder so no new
-    environment variable or separate path configuration is required.
-    """
-    return DEFAULT_STORAGE_ROOT.parent / "Work"
+    return _resolve_document_storage_root().parent / "Work"
 
 
 def _segment(value: object, default: str = "NA") -> str:
@@ -92,7 +106,7 @@ def _validate_pdf_header(data: bytes) -> None:
 
 
 async def save_uploads(files: list[UploadFile], input_dir: Path) -> list[Path]:
-    usable = [f for f in files if f and (f.filename or "").strip()]
+    usable = [file for file in files if file and (file.filename or "").strip()]
     if not usable:
         raise ValueError("Please select at least one PDF bank statement.")
     if len(usable) > MAX_FILES:
@@ -122,10 +136,36 @@ async def save_uploads(files: list[UploadFile], input_dir: Path) -> list[Path]:
     return saved
 
 
-def analyze_job(*, user, roles: Iterable[str], job_id: str, paths: list[Path], output_dir: Path, customer_override: str = "", account_override: str = "") -> dict:
-    metas, all_df, unique_df = analyze_files(paths, customer_override.strip(), account_override.strip())
+def analyze_job(
+    *,
+    user,
+    roles: Iterable[str],
+    job_id: str,
+    paths: list[Path],
+    output_dir: Path,
+    customer_override: str = "",
+    account_override: str = "",
+    bank_selection: str = "auto",
+    financial_year: str = "",
+    classification_enabled: bool = True,
+) -> dict:
+    metas, all_df, unique_df = analyze_files(
+        paths,
+        customer_override.strip(),
+        account_override.strip(),
+        bank_hint=bank_selection,
+        classification_enabled=classification_enabled,
+    )
     output = output_dir / "Bank_Statement_Analysis.xlsx"
-    export_excel(output, metas, all_df, unique_df)
+    export_excel(
+        output,
+        metas,
+        all_df,
+        unique_df,
+        financial_year=financial_year.strip(),
+        selected_bank=bank_selection,
+        classification_enabled=classification_enabled,
+    )
     summary = {
         "job_id": job_id,
         "owner_user_id": int(user.id),
@@ -135,9 +175,13 @@ def analyze_job(*, user, roles: Iterable[str], job_id: str, paths: list[Path], o
         "unique_transactions": int(len(unique_df)),
         "exact_duplicate_rows": int(all_df.exact_duplicate.sum()) if not all_df.empty else 0,
         "possible_duplicate_rows": int(all_df.possible_duplicate.sum()) if not all_df.empty else 0,
-        "banks": sorted({m.bank_name for m in metas}),
-        "customer_name": next((m.customer_name for m in metas if m.customer_name), ""),
-        "account_number": next((m.account_number for m in metas if m.account_number), ""),
+        "review_items": int(unique_df.review_note.fillna("").ne("").sum()) if not unique_df.empty else 0,
+        "categories": int(unique_df.category.nunique()) if not unique_df.empty else 0,
+        "banks": sorted({meta.bank_name for meta in metas}),
+        "customer_name": next((meta.customer_name for meta in metas if meta.customer_name), ""),
+        "account_number": next((meta.account_number for meta in metas if meta.account_number), ""),
+        "financial_year": financial_year.strip(),
+        "classification_enabled": classification_enabled,
         "output_file": output.name,
     }
     (output_dir.parent / "job.json").write_text(json.dumps(summary, indent=2), encoding="utf-8")