feat(platform): close AI expense value loop

Add tenant-safe value, telemetry, connector, commercial, and production-readiness foundations.
This commit is contained in:
caoxiaozhu
2026-07-17 14:14:08 +08:00
parent 242d68c36f
commit 787bc3a481
507 changed files with 82072 additions and 6344 deletions

View File

@@ -9,11 +9,14 @@ from app.services.document_preview import DocumentPreviewAssets
from app.services.receipt_folder import ReceiptFolderService
def test_receipt_folder_train_ticket_uses_invoice_date_and_enriches_fields(monkeypatch, tmp_path) -> None:
def test_receipt_folder_train_ticket_uses_invoice_date_and_enriches_fields(
monkeypatch, tmp_path
) -> None:
monkeypatch.setenv("STORAGE_ROOT_DIR", str(tmp_path / "storage"))
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
@@ -77,6 +80,7 @@ def test_receipt_folder_pdf_save_eagerly_renders_image_preview(monkeypatch, tmp_
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
@@ -87,7 +91,9 @@ def test_receipt_folder_pdf_save_eagerly_renders_image_preview(monkeypatch, tmp_
preview_path.write_bytes(b"rendered-preview")
return preview_path
monkeypatch.setattr(DocumentPreviewAssets, "render_pdf_first_page", fake_render_pdf_first_page)
monkeypatch.setattr(
DocumentPreviewAssets, "render_pdf_first_page", fake_render_pdf_first_page
)
service = ReceiptFolderService()
receipt = service.save_receipt(
@@ -121,11 +127,14 @@ def test_receipt_folder_pdf_save_eagerly_renders_image_preview(monkeypatch, tmp_
get_settings.cache_clear()
def test_receipt_folder_persist_enriches_pdf_ocr_document_with_image_preview(monkeypatch, tmp_path) -> None:
def test_receipt_folder_persist_enriches_pdf_ocr_document_with_image_preview(
monkeypatch, tmp_path
) -> None:
monkeypatch.setenv("STORAGE_ROOT_DIR", str(tmp_path / "storage"))
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
@@ -136,7 +145,9 @@ def test_receipt_folder_persist_enriches_pdf_ocr_document_with_image_preview(mon
preview_path.write_bytes(b"rendered-preview")
return preview_path
monkeypatch.setattr(DocumentPreviewAssets, "render_pdf_first_page", fake_render_pdf_first_page)
monkeypatch.setattr(
DocumentPreviewAssets, "render_pdf_first_page", fake_render_pdf_first_page
)
service = ReceiptFolderService()
result = service.persist_ocr_batch(
@@ -162,7 +173,9 @@ def test_receipt_folder_persist_enriches_pdf_ocr_document_with_image_preview(mon
document = result.documents[0]
assert document.receipt_id
assert document.receipt_preview_url.endswith(f"/receipt-folder/{document.receipt_id}/preview")
assert document.receipt_preview_url.endswith(
f"/receipt-folder/{document.receipt_id}/preview"
)
assert document.preview_kind == "image"
finally:
get_settings.cache_clear()
@@ -173,13 +186,16 @@ def test_receipt_folder_pdf_preview_regenerates_stale_cached_image(monkeypatch,
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
is_admin=False,
)
stale_preview = b"stale-preview"
preview_data_url = f"data:image/png;base64,{base64.b64encode(stale_preview).decode('ascii')}"
preview_data_url = (
f"data:image/png;base64,{base64.b64encode(stale_preview).decode('ascii')}"
)
service = ReceiptFolderService()
receipt = service.save_receipt(
filename="2月20_武汉-上海.pdf",
@@ -205,7 +221,9 @@ def test_receipt_folder_pdf_preview_regenerates_stale_cached_image(monkeypatch,
preview_path.write_bytes(b"refreshed-preview")
return preview_path
monkeypatch.setattr(DocumentPreviewAssets, "render_pdf_first_page", fake_render_pdf_first_page)
monkeypatch.setattr(
DocumentPreviewAssets, "render_pdf_first_page", fake_render_pdf_first_page
)
resolved_path, media_type, file_name = service.resolve_preview(receipt.id, current_user)
@@ -227,13 +245,16 @@ def test_receipt_folder_pdf_preview_falls_back_to_source_when_render_fonts_missi
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
is_admin=False,
)
stale_preview = b"broken-preview"
preview_data_url = f"data:image/png;base64,{base64.b64encode(stale_preview).decode('ascii')}"
preview_data_url = (
f"data:image/png;base64,{base64.b64encode(stale_preview).decode('ascii')}"
)
service = ReceiptFolderService()
receipt = service.save_receipt(
filename="2月20_武汉-上海.pdf",
@@ -256,7 +277,9 @@ def test_receipt_folder_pdf_preview_falls_back_to_source_when_render_fonts_missi
def fake_render_pdf_first_page(*, pdf_path, preview_path, timeout_seconds):
raise RuntimeError("Missing language pack for 'Adobe-GB1' mapping")
monkeypatch.setattr(DocumentPreviewAssets, "render_pdf_first_page", fake_render_pdf_first_page)
monkeypatch.setattr(
DocumentPreviewAssets, "render_pdf_first_page", fake_render_pdf_first_page
)
resolved_path, media_type, file_name = service.resolve_preview(receipt.id, current_user)
@@ -280,6 +303,7 @@ def test_receipt_folder_train_ticket_extracts_passenger_from_id_line_and_purchas
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
@@ -314,9 +338,13 @@ def test_receipt_folder_train_ticket_extracts_passenger_from_id_line_and_purchas
scene_code="travel",
scene_label="差旅票据",
document_fields=[
OcrRecognizeFieldRead(key="merchant_name", label="商户", value="电子发票(铁路"),
OcrRecognizeFieldRead(
key="merchant_name", label="商户", value="电子发票(铁路"
),
OcrRecognizeFieldRead(key="amount", label="金额", value="354元"),
OcrRecognizeFieldRead(key="date", label="列车出发时间", value="2026-02-20 07:55"),
OcrRecognizeFieldRead(
key="date", label="列车出发时间", value="2026-02-20 07:55"
),
OcrRecognizeFieldRead(key="trip_no", label="车次", value="G458"),
OcrRecognizeFieldRead(key="route", label="行程", value="武汉-上海"),
],
@@ -349,6 +377,7 @@ def test_receipt_folder_train_ticket_repairs_invalid_generated_fields_from_ocr_t
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
@@ -401,12 +430,20 @@ def test_receipt_folder_train_ticket_repairs_invalid_generated_fields_from_ocr_t
scene_label="差旅票据",
document_fields=[
OcrRecognizeFieldRead(key="amount", label="金额", value="438元"),
OcrRecognizeFieldRead(key="date", label="列车出发时间", value="2026-02-21 08:30"),
OcrRecognizeFieldRead(key="invoice_number", label="票据号码", value="DEMO202602210001"),
OcrRecognizeFieldRead(
key="date", label="列车出发时间", value="2026-02-21 08:30"
),
OcrRecognizeFieldRead(
key="invoice_number", label="票据号码", value="DEMO202602210001"
),
OcrRecognizeFieldRead(key="trip_no", label="车次", value="G999"),
OcrRecognizeFieldRead(key="route", label="行程", value="上海-深圳"),
OcrRecognizeFieldRead(key="departure_station", label="出发地点", value="二等座"),
OcrRecognizeFieldRead(key="arrival_station", label="到达地点", value="扫码无效"),
OcrRecognizeFieldRead(
key="departure_station", label="出发地点", value="二等座"
),
OcrRecognizeFieldRead(
key="arrival_station", label="到达地点", value="扫码无效"
),
OcrRecognizeFieldRead(key="passenger_name", label="乘车人", value="席别二等座"),
],
),
@@ -431,6 +468,7 @@ def test_receipt_folder_delete_removes_duplicate_marker(monkeypatch, tmp_path) -
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
@@ -480,11 +518,14 @@ def test_receipt_folder_delete_removes_duplicate_marker(monkeypatch, tmp_path) -
get_settings.cache_clear()
def test_receipt_folder_duplicate_uses_newer_ocr_when_existing_meta_is_weaker(monkeypatch, tmp_path) -> None:
def test_receipt_folder_duplicate_uses_newer_ocr_when_existing_meta_is_weaker(
monkeypatch, tmp_path
) -> None:
monkeypatch.setenv("STORAGE_ROOT_DIR", str(tmp_path / "storage"))
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
@@ -538,7 +579,9 @@ def test_receipt_folder_duplicate_uses_newer_ocr_when_existing_meta_is_weaker(mo
assert document.receipt_id == stale_receipt.id
assert document.document_type == "train_ticket"
assert document.document_type_label == "火车/高铁票"
assert any(field.label == "金额" and field.value == "354元" for field in document.document_fields)
assert any(
field.label == "金额" and field.value == "354元" for field in document.document_fields
)
assert any("重复上传" in warning for warning in document.warnings)
repaired = service.get_receipt(stale_receipt.id, current_user)
@@ -549,11 +592,14 @@ def test_receipt_folder_duplicate_uses_newer_ocr_when_existing_meta_is_weaker(mo
get_settings.cache_clear()
def test_receipt_folder_recovers_train_ticket_detail_from_other_english_ocr(monkeypatch, tmp_path) -> None:
def test_receipt_folder_recovers_train_ticket_detail_from_other_english_ocr(
monkeypatch, tmp_path
) -> None:
monkeypatch.setenv("STORAGE_ROOT_DIR", str(tmp_path / "storage"))
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],
@@ -610,11 +656,14 @@ def test_receipt_folder_recovers_train_ticket_detail_from_other_english_ocr(monk
get_settings.cache_clear()
def test_receipt_folder_unlink_receipts_for_claim_marks_linked_receipts_unlinked(monkeypatch, tmp_path) -> None:
def test_receipt_folder_unlink_receipts_for_claim_marks_linked_receipts_unlinked(
monkeypatch, tmp_path
) -> None:
monkeypatch.setenv("STORAGE_ROOT_DIR", str(tmp_path / "storage"))
get_settings.cache_clear()
try:
current_user = CurrentUserContext(
tenant_id="default",
username="pytest",
name="Py Test",
role_codes=[],