tabularmapper 1.0.0__tar.gz → 1.0.1__tar.gz

{tabularmapper-1.0.0/src/tabularmapper.egg-info → tabularmapper-1.0.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tabularmapper
-Version: 1.0.0
+Version: 1.0.1
 Summary: Map any spreadsheet (.xlsx) to a schema you define — deterministic column mapping with an optional AI matcher
 Author-email: Karthikeyan Duraisamy <karthikeyanduraisamy@kultivateindia.com>
 License-Expression: MIT

{tabularmapper-1.0.0 → tabularmapper-1.0.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "tabularmapper"
-version = "1.0.0"
+version = "1.0.1"
 description = "Map any spreadsheet (.xlsx) to a schema you define — deterministic column mapping with an optional AI matcher"
 readme = "README.md"
 requires-python = ">=3.9"

{tabularmapper-1.0.0 → tabularmapper-1.0.1}/src/tabularmapper/__init__.py

@@ -44,7 +44,7 @@ from .schema import (
 )
 from .stores import open_store
 
-__version__ = "1.0.0"
+__version__ = "1.0.1"
 
 __all__ = [
     "process_file",

{tabularmapper-1.0.0 → tabularmapper-1.0.1}/src/tabularmapper/api.py

@@ -32,14 +32,17 @@ import os
 from contextlib import asynccontextmanager
 from typing import Any, Optional
 
-from fastapi import APIRouter, FastAPI, File, HTTPException, UploadFile
+from fastapi import APIRouter, FastAPI, File, HTTPException, Query, UploadFile
 from fastapi.concurrency import run_in_threadpool
+from fastapi.responses import Response
 from pydantic import BaseModel
 
 from . import engine                    # imported as a module so OUTPUT_SCHEMA is read
-from .engine import process_stream  # dynamically (after configure), never a stale copy
+from .engine import OutputResult, process_stream  # dynamically (after configure), never a stale copy
 from .mapping_cache import MappingCache
 
+_XLSX_MIME = "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
+
 
 # --------------------------------------------------------------------------
 # Shared singletons (built once at startup)
@@ -105,6 +108,9 @@ class MapResponse(BaseModel):
     schema_columns: list[str]
     columns: list[ColumnMapOut]
     transactions: list[dict]
+    # Populated only when ?format=base64 — a base64-encoded .xlsx of the mapped
+    # rows, ready to decode and save client-side. None otherwise.
+    file_base64: Optional[str] = None
 
 
 # --------------------------------------------------------------------------
@@ -114,8 +120,23 @@ async def health() -> dict:
     return {"status": "ok", "ai_enabled": state.matcher is not None}
 
 
-async def map_statement(file: UploadFile = File(...)) -> MapResponse:
-    """Upload a spreadsheet (.xlsx); get the standardized mapping + rows."""
+async def map_statement(
+    file: UploadFile = File(...),
+    format: str = Query(
+        "json",
+        pattern="^(json|base64|file)$",
+        description="json = rows inline (default); base64 = rows inline + an "
+                    ".xlsx encoded in file_base64; file = download the .xlsx "
+                    "directly (binary, no JSON body).",
+    ),
+):
+    """Upload a spreadsheet (.xlsx); get the standardized mapping + rows.
+
+    `format` controls what comes back:
+      * json    -> MapResponse with the rows in `transactions`
+      * base64  -> same MapResponse, plus a mapped .xlsx in `file_base64`
+      * file    -> the mapped .xlsx as a downloadable attachment
+    """
     name = (file.filename or "").lower()
     if not name.endswith((".xlsx", ".xls")):
         raise HTTPException(status_code=400, detail="expected an .xlsx/.xls file")
@@ -132,6 +153,22 @@ async def map_statement(file: UploadFile = File(...)) -> MapResponse:
     except Exception as exc:  # noqa: BLE001
         raise HTTPException(status_code=422, detail=f"could not process file: {exc}")
 
+    # file -> stream the mapped .xlsx straight back as a download (no JSON body).
+    if format == "file":
+        xlsx = await run_in_threadpool(
+            lambda: OutputResult(records=res.records, format="bytes").bytes)
+        stem = os.path.splitext(os.path.basename(file.filename or "mapped"))[0]
+        return Response(
+            content=xlsx,
+            media_type=_XLSX_MIME,
+            headers={"Content-Disposition": f'attachment; filename="{stem}_mapped.xlsx"'},
+        )
+
+    file_b64 = None
+    if format == "base64":
+        file_b64 = await run_in_threadpool(
+            lambda: OutputResult(records=res.records, format="base64").base64)
+
     return MapResponse(
         header_index=res.header_index,
         needs_review=res.needs_review,
@@ -142,6 +179,7 @@ async def map_statement(file: UploadFile = File(...)) -> MapResponse:
             "field": m.field, "confidence": m.confidence, "method": m.method,
         }) for m in res.column_maps],
         transactions=res.records,
+        file_base64=file_b64,
     )
 
 

{tabularmapper-1.0.0 → tabularmapper-1.0.1/src/tabularmapper.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tabularmapper
-Version: 1.0.0
+Version: 1.0.1
 Summary: Map any spreadsheet (.xlsx) to a schema you define — deterministic column mapping with an optional AI matcher
 Author-email: Karthikeyan Duraisamy <karthikeyanduraisamy@kultivateindia.com>
 License-Expression: MIT

{tabularmapper-1.0.0 → tabularmapper-1.0.1}/tests/test_api.py

@@ -59,6 +59,48 @@ def test_map_deterministic(client):
     assert body["schema_columns"][0] == "Date"
 
 
+def test_map_format_base64(client):
+    import base64
+    with open(os.path.join(FIX, "01_junk_split.xlsx"), "rb") as fh:
+        payload = fh.read()
+    r = client.post("/mapper/map", params={"format": "base64"},
+                    files={"file": ("stmt.xlsx", io.BytesIO(payload))})
+    assert r.status_code == 200
+    body = r.json()
+    # rows still inline...
+    assert any(t["credit"] == 45000.0 for t in body["transactions"])
+    # ...plus a base64 .xlsx that decodes to a real zip (xlsx magic = PK)
+    assert body["file_base64"]
+    assert base64.b64decode(body["file_base64"])[:2] == b"PK"
+
+
+def test_map_format_file_download(client):
+    with open(os.path.join(FIX, "01_junk_split.xlsx"), "rb") as fh:
+        payload = fh.read()
+    r = client.post("/mapper/map", params={"format": "file"},
+                    files={"file": ("statement.xlsx", io.BytesIO(payload))})
+    assert r.status_code == 200
+    assert r.headers["content-type"].startswith(
+        "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet")
+    assert 'attachment; filename="statement_mapped.xlsx"' in r.headers["content-disposition"]
+    assert r.content[:2] == b"PK"          # real .xlsx bytes
+
+
+def test_map_default_format_has_no_file_base64(client):
+    with open(os.path.join(FIX, "01_junk_split.xlsx"), "rb") as fh:
+        payload = fh.read()
+    r = client.post("/mapper/map", files={"file": ("stmt.xlsx", io.BytesIO(payload))})
+    assert r.json()["file_base64"] is None
+
+
+def test_map_rejects_bad_format(client):
+    with open(os.path.join(FIX, "01_junk_split.xlsx"), "rb") as fh:
+        payload = fh.read()
+    r = client.post("/mapper/map", params={"format": "pdf"},
+                    files={"file": ("stmt.xlsx", io.BytesIO(payload))})
+    assert r.status_code == 422       # fails the regex pattern
+
+
 def test_map_rejects_non_xlsx(client):
     r = client.post("/mapper/map",
                     files={"file": ("notes.txt", io.BytesIO(b"hello"), "text/plain")})