Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
Python ai4sciencedocument-analysisdocxextract-datalayout-analysisocrparserpdfpdf-converterpdf-extractor-llmpdf-extractor-pretrainpdf-extractor-ragpdf-parserpptxpythonxlsx
feat: accept legacy 3.x middle JSON in remote API results
jinzhenj79307fbb61d5
diff --git a/mineru/parser/api_client.py b/mineru/parser/api_client.pyindex e6caf487..f9f699d8 100644--- a/mineru/parser/api_client.py+++ b/mineru/parser/api_client.py@@ -1159,13 +1159,20 @@ def _pages_from_middle_json(mid_json: dict[str, Any] | None) -> list[PageInfo]:def _parse_result_from_middle_json(mid_json: dict[str, Any]) -> ParseResult:- """把远端 middle_json 恢复为 ParseResult;只接受当前共享文档协议。"""+ """把远端 middle_json 恢复为 ParseResult;优先当前协议,临时兼容 3.x 历史协议。"""if not isinstance(mid_json, dict):raise _V1APIError("invalid_middle_json_output", "middle_json output must be a JSON object")try:return ParseResult.from_dict(mid_json)except ValueError as exc:- raise _V1APIError("invalid_middle_json_output", str(exc)) from exc+ strict_error = exc+ # 临时:兼容 3.x remote server 的历史 middle_json(pdf_info / 旧 1.0 / 旧 2.0),待其升级到 4.0 后移除。+ from ..backend.postprocess.legacy_middle_json import read_legacy_middle_json++ try:+ return ParseResult(middle_json=read_legacy_middle_json(mid_json))+ except ValueError:+ raise _V1APIError("invalid_middle_json_output", str(strict_error)) from strict_errordef _raise_for_terminal_job_error(job: dict[str, Any]) -> None:diff --git a/tests/unittest/test_parser_api_contract.py b/tests/unittest/test_parser_api_contract.pyindex eee51e3a..05ac8357 100644--- a/tests/unittest/test_parser_api_contract.py+++ b/tests/unittest/test_parser_api_contract.py@@ -811,8 +811,8 @@ def test_api_client_downloads_model_output_from_zip(monkeypatch: pytest.MonkeyPaassert result._model_output.pages == [[{"raw": "model"}]]-def test_api_client_rejects_legacy_official_layout_json(monkeypatch: pytest.MonkeyPatch) -> None:- """官方 API ZIP 内的历史 pdf_info 文档必须明确拒绝,不再迁移正文。"""+def test_api_client_accepts_legacy_official_layout_json(monkeypatch: pytest.MonkeyPatch) -> None:+ """官方 API ZIP 内的历史 pdf_info 文档迁移为当前 MiddleJson,不再拒绝。"""parser = MinerUApiParser(api_url="https://mineru.net/api",tier="standard",@@ -850,16 +850,18 @@ def test_api_client_rejects_legacy_official_layout_json(monkeypatch: pytest.Monkmonkeypatch.setattr(api_client, "_download_bytes", lambda _parser, ref: zip_buffer.getvalue() if ref is zip_ref else b"")- with pytest.raises(api_client._V1APIError, match="reparse"):- _parse_result_from_job(- {- "job_id": "job_1",- "status": "completed",- "files": [{"output_files": {"zip": zip_ref}}],- },- "demo.pdf",- parser,- )+ result = _parse_result_from_job(+ {+ "job_id": "job_1",+ "status": "completed",+ "files": [{"output_files": {"zip": zip_ref}}],+ },+ "demo.pdf",+ parser,+ )++ assert isinstance(result, ParseResult)+ assert isinstance(result.pages, list)def test_api_client_async_downloads_model_output_from_zip(monkeypatch: pytest.MonkeyPatch) -> None:@@ -1052,8 +1054,8 @@ def test_api_client_include_images_rejects_unsafe_image_entries(monkeypatch: pyt)-def test_api_client_rejects_remote_pdf_info_middle_json(monkeypatch: pytest.MonkeyPatch) -> None:- """远程返回历史 pdf_info 协议时必须提示重新解析,不猜测其版本。"""+def test_api_client_accepts_remote_pdf_info_middle_json(monkeypatch: pytest.MonkeyPatch) -> None:+ """远程返回历史 pdf_info 协议时迁移为当前 MiddleJson,不猜测但也不拒绝。"""parser = MinerUApiParser(api_url="https://mineru.net/api", tier="standard")middle_json = {"_backend": "hybrid",@@ -1063,20 +1065,22 @@ def test_api_client_rejects_remote_pdf_info_middle_json(monkeypatch: pytest.Monkmonkeypatch.setattr(api_client, "_download_json", lambda _parser, _outputs: middle_json)- with pytest.raises(api_client._V1APIError, match="reparse"):- _parse_result_from_job(- {- "job_id": "job_1",- "status": "completed",- "files": [{"output_files": {"middle_json": {"file_id": "file-middle-json", "bytes": 10}}}],- },- "demo.pdf",- parser,- )+ result = _parse_result_from_job(+ {+ "job_id": "job_1",+ "status": "completed",+ "files": [{"output_files": {"middle_json": {"file_id": "file-middle-json", "bytes": 10}}}],+ },+ "demo.pdf",+ parser,+ )+ assert isinstance(result, ParseResult)+ assert isinstance(result.pages, list)-def test_async_api_client_rejects_remote_pdf_info_middle_json(monkeypatch: pytest.MonkeyPatch) -> None:- """异步客户端同样拒绝历史 pdf_info 协议,不把空页当成有效结果。"""++def test_async_api_client_accepts_remote_pdf_info_middle_json(monkeypatch: pytest.MonkeyPatch) -> None:+ """异步客户端同样迁移历史 pdf_info 协议。"""parser = MinerUApiParser(api_url="https://mineru.net/api", tier="standard")middle_json = {"_backend": "hybrid",@@ -1088,18 +1092,20 @@ def test_async_api_client_rejects_remote_pdf_info_middle_json(monkeypatch: pytesmonkeypatch.setattr(api_client, "_async_download_json", _download_json)- with pytest.raises(api_client._V1APIError, match="reparse"):- asyncio.run(- api_client._async_parse_result_from_job(- {- "job_id": "job_1",- "status": "completed",- "files": [{"output_files": {"middle_json": {"file_id": "file-middle-json", "bytes": 10}}}],- },- "demo.pdf",- parser,- )+ result = asyncio.run(+ api_client._async_parse_result_from_job(+ {+ "job_id": "job_1",+ "status": "completed",+ "files": [{"output_files": {"middle_json": {"file_id": "file-middle-json", "bytes": 10}}}],+ },+ "demo.pdf",+ parser,)+ )++ assert isinstance(result, ParseResult)+ assert isinstance(result.pages, list)def test_api_client_rejects_legacy_json_output_file_key() -> None: