paddleocr-haystack 0.1.0__tar.gz → 1.1.0__tar.gz

paddleocr_haystack-1.1.0/CHANGELOG.md

@@ -0,0 +1,20 @@
+# Changelog
+
+## [integrations/paddleocr-v1.0.0] - 2026-01-12
+
+### 🧹 Chores
+
+- Make fmt command more forgiving (#2671)
+- [**breaking**] Paddleocr - drop Python 3.9 and use X|Y typing (#2714)
+
+### 🌀 Miscellaneous
+
+- Feat: Add 'Client-Platform' header for server processing for PaddleOCR (#2657)
+
+## [integrations/paddleocr-v0.1.0] - 2025-12-10
+
+### 🚀 Features
+
+- Add PaddleOCR-VL document converter (#2567)
+
+<!-- generated by git-cliff -->

{paddleocr_haystack-0.1.0 → paddleocr_haystack-1.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: paddleocr-haystack
-Version: 0.1.0
+Version: 1.1.0
 Summary: An integration of PaddleOCR with Haystack
 Project-URL: Documentation, https://github.com/haystack-core-integrations/tree/main/integrations/paddleocr#readme
 Project-URL: Issues, https://github.com/haystack-core-integrations/issues
@@ -10,16 +10,15 @@ License-Expression: Apache-2.0
 License-File: LICENSE.txt
 Classifier: Development Status :: 4 - Beta
 Classifier: Programming Language :: Python
-Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: Implementation :: CPython
 Classifier: Programming Language :: Python :: Implementation :: PyPy
-Requires-Python: >=3.9
-Requires-Dist: haystack-ai>=2.19.0
-Requires-Dist: paddleocr>=3.3.2
-Requires-Dist: paddlex[serving]>=3.3.10
+Requires-Python: >=3.10
+Requires-Dist: haystack-ai>=2.22.0
+Requires-Dist: paddleocr>=3.4.0
+Requires-Dist: paddlex[serving]>=3.4.0
 Requires-Dist: requests>=2.25.0
 Description-Content-Type: text/markdown
 

{paddleocr_haystack-0.1.0 → paddleocr_haystack-1.1.0}/pyproject.toml

@@ -7,7 +7,7 @@ name = "paddleocr-haystack"
 dynamic = ["version"]
 description = 'An integration of PaddleOCR with Haystack'
 readme = "README.md"
-requires-python = ">=3.9"
+requires-python = ">=3.10"
 license = "Apache-2.0"
 keywords = []
 authors = [
@@ -16,7 +16,6 @@ authors = [
 classifiers = [
   "Development Status :: 4 - Beta",
   "Programming Language :: Python",
-  "Programming Language :: Python :: 3.9",
   "Programming Language :: Python :: 3.10",
   "Programming Language :: Python :: 3.11",
   "Programming Language :: Python :: 3.12",
@@ -24,9 +23,9 @@ classifiers = [
   "Programming Language :: Python :: Implementation :: PyPy",
 ]
 dependencies = [
-  "haystack-ai>=2.19.0",
-  "paddleocr>=3.3.2",
-  "paddlex[serving]>=3.3.10",
+  "haystack-ai>=2.22.0",
+  "paddleocr>=3.4.0",
+  "paddlex[serving]>=3.4.0",
   "requests>=2.25.0",
 ]
 
@@ -52,7 +51,7 @@ dependencies = ["haystack-pydoc-tools", "ruff"]
 
 [tool.hatch.envs.default.scripts]
 docs = ["pydoc-markdown pydoc/config_docusaurus.yml"]
-fmt = "ruff check --fix {args} && ruff format {args}"
+fmt = "ruff check --fix {args}; ruff format {args}"
 fmt-check = "ruff check {args} && ruff format --check {args}"
 
 [tool.hatch.envs.test]
@@ -80,7 +79,6 @@ check_untyped_defs = true
 disallow_incomplete_defs = true
 
 [tool.ruff]
-target-version = "py39"
 line-length = 120
 
 [tool.ruff.lint]
@@ -127,10 +125,6 @@ ignore = [
   "B008",
   "S101",
 ]
-unfixable = [
-  # Don't touch unused imports
-  "F401",
-]
 
 [tool.ruff.lint.isort]
 known-first-party = ["haystack_integrations"]

{paddleocr_haystack-0.1.0 → paddleocr_haystack-1.1.0}/src/haystack_integrations/components/converters/paddleocr/paddleocr_vl_document_converter.py

@@ -3,7 +3,7 @@
 # SPDX-License-Identifier: Apache-2.0
 import base64
 from pathlib import Path
-from typing import Any, Literal, Optional, Union
+from typing import Any, Literal
 
 import requests
 from haystack import Document, component, default_from_dict, default_to_dict, logging
@@ -24,7 +24,7 @@ from paddlex.inference.serving.schemas.shared.ocr import FileType  # type: ignor
 logger = logging.getLogger(__name__)
 
 
-FileTypeInput = Union[Literal["pdf", "image"], None]
+FileTypeInput = Literal["pdf", "image"] | None
 
 # Supported image file extensions
 _IMAGE_EXTENSIONS = {
@@ -41,9 +41,9 @@ _PDF_EXTENSIONS = {".pdf"}
 
 
 def _infer_file_type_from_source(
-    source: Union[str, Path, ByteStream],
-    mime_type: Optional[str] = None,
-) -> Optional[FileType]:
+    source: str | Path | ByteStream,
+    mime_type: str | None = None,
+) -> FileType | None:
     """
     Infer file type from file extension or MIME type.
 
@@ -56,7 +56,7 @@ def _infer_file_type_from_source(
         determined.
     """
     # Try to get extension from file path
-    file_path: Optional[str] = None
+    file_path: str | None = None
 
     # Check if source is a file path
     if isinstance(source, (str, Path)):
@@ -86,7 +86,7 @@ def _infer_file_type_from_source(
     return None
 
 
-def _normalize_file_type(file_type: Optional[FileTypeInput]) -> Optional[FileType]:
+def _normalize_file_type(file_type: FileTypeInput) -> FileType | None:
     """
     Normalize file type input to the numeric format expected by the API.
 
@@ -145,36 +145,45 @@ class PaddleOCRVLDocumentConverter:
         *,
         api_url: str,
         access_token: Secret = Secret.from_env_var("AISTUDIO_ACCESS_TOKEN"),
-        file_type: Optional[FileTypeInput] = None,
-        use_doc_orientation_classify: Optional[bool] = None,
-        use_doc_unwarping: Optional[bool] = None,
-        use_layout_detection: Optional[bool] = None,
-        use_chart_recognition: Optional[bool] = None,
-        layout_threshold: Optional[Union[float, dict]] = None,
-        layout_nms: Optional[bool] = None,
-        layout_unclip_ratio: Optional[Union[float, tuple[float, float], dict]] = None,
-        layout_merge_bboxes_mode: Optional[Union[str, dict]] = None,
-        prompt_label: Optional[str] = None,
-        format_block_content: Optional[bool] = None,
-        repetition_penalty: Optional[float] = None,
-        temperature: Optional[float] = None,
-        top_p: Optional[float] = None,
-        min_pixels: Optional[int] = None,
-        max_pixels: Optional[int] = None,
-        prettify_markdown: Optional[bool] = None,
-        show_formula_number: Optional[bool] = None,
-        visualize: Optional[bool] = None,
-        additional_params: Optional[dict[str, Any]] = None,
+        file_type: FileTypeInput = None,
+        use_doc_orientation_classify: bool | None = False,
+        use_doc_unwarping: bool | None = False,
+        use_layout_detection: bool | None = None,
+        use_chart_recognition: bool | None = None,
+        use_seal_recognition: bool | None = None,
+        use_ocr_for_image_block: bool | None = None,
+        layout_threshold: float | dict | None = None,
+        layout_nms: bool | None = None,
+        layout_unclip_ratio: float | tuple[float, float] | dict | None = None,
+        layout_merge_bboxes_mode: str | dict | None = None,
+        layout_shape_mode: str | None = None,
+        prompt_label: str | None = None,
+        format_block_content: bool | None = None,
+        repetition_penalty: float | None = None,
+        temperature: float | None = None,
+        top_p: float | None = None,
+        min_pixels: int | None = None,
+        max_pixels: int | None = None,
+        max_new_tokens: int | None = None,
+        merge_layout_blocks: bool | None = None,
+        markdown_ignore_labels: list[str] | None = None,
+        vlm_extra_args: dict | None = None,
+        prettify_markdown: bool | None = None,
+        show_formula_number: bool | None = None,
+        restructure_pages: bool | None = None,
+        merge_tables: bool | None = None,
+        relevel_titles: bool | None = None,
+        visualize: bool | None = None,
+        additional_params: dict[str, Any] | None = None,
     ):
         """
         Create a `PaddleOCRVLDocumentConverter` component.
 
         :param api_url:
             API URL. To obtain the API URL, visit the [PaddleOCR official
-            website](https://aistudio.baidu.com/paddleocr/task), click the
-            **API** button in the upper-left corner, choose the example code
-            for **Large Model document parsing(PaddleOCR-VL)**, and copy the
-            `API_URL`.
+            website](https://aistudio.baidu.com/paddleocr), click the
+            **API** button, choose the example code for PaddleOCR-VL, and copy
+            the `API_URL`.
         :param access_token:
             AI Studio access token. You can obtain it from [this
             page](https://aistudio.baidu.com/account/accessToken).
@@ -193,6 +202,10 @@ class PaddleOCRVLDocumentConverter:
             Whether to enable the layout detection function.
         :param use_chart_recognition:
             Whether to enable the chart recognition function.
+        :param use_seal_recognition:
+            Whether to enable the seal recognition function.
+        :param use_ocr_for_image_block:
+            Whether to recognize text in image blocks.
         :param layout_threshold:
             Layout detection threshold. Can be a float or a dict with
             page-specific thresholds.
@@ -204,9 +217,11 @@ class PaddleOCRVLDocumentConverter:
             dict with page-specific values.
         :param layout_merge_bboxes_mode:
             Layout merge bounding boxes mode. Can be a string or a dict.
+        :param layout_shape_mode:
+            Layout shape mode.
         :param prompt_label:
             Prompt type for the VLM. Possible values are "ocr", "formula",
-            "table", and "chart".
+            "table", "chart", "seal", and "spotting".
         :param format_block_content:
             Whether to format block content.
         :param repetition_penalty:
@@ -219,10 +234,25 @@ class PaddleOCRVLDocumentConverter:
             Minimum number of pixels allowed during VLM preprocessing.
         :param max_pixels:
             Maximum number of pixels allowed during VLM preprocessing.
+        :param max_new_tokens:
+            Maximum number of tokens generated by the VLM.
+        :param merge_layout_blocks:
+            Whether to merge the layout detection boxes for cross-column or
+            staggered top and bottom columns.
+        :param markdown_ignore_labels:
+            Layout labels that need to be ignored in Markdown.
+        :param vlm_extra_args:
+            Additional configuration parameters for the VLM.
         :param prettify_markdown:
             Whether to prettify the output Markdown text.
         :param show_formula_number:
             Whether to include formula numbers in the output markdown text.
+        :param restructure_pages:
+            Whether to restructure results across multiple pages.
+        :param merge_tables:
+            Whether to merge tables across pages.
+        :param relevel_titles:
+            Whether to relevel titles.
         :param visualize:
             Whether to return visualization results.
         :param additional_params:
@@ -235,10 +265,13 @@ class PaddleOCRVLDocumentConverter:
         self.use_doc_unwarping = use_doc_unwarping
         self.use_layout_detection = use_layout_detection
         self.use_chart_recognition = use_chart_recognition
+        self.use_seal_recognition = use_seal_recognition
+        self.use_ocr_for_image_block = use_ocr_for_image_block
         self.layout_threshold = layout_threshold
         self.layout_nms = layout_nms
         self.layout_unclip_ratio = layout_unclip_ratio
         self.layout_merge_bboxes_mode = layout_merge_bboxes_mode
+        self.layout_shape_mode = layout_shape_mode
         self.prompt_label = prompt_label
         self.format_block_content = format_block_content
         self.repetition_penalty = repetition_penalty
@@ -246,8 +279,15 @@ class PaddleOCRVLDocumentConverter:
         self.top_p = top_p
         self.min_pixels = min_pixels
         self.max_pixels = max_pixels
+        self.max_new_tokens = max_new_tokens
+        self.merge_layout_blocks = merge_layout_blocks
+        self.markdown_ignore_labels = markdown_ignore_labels
+        self.vlm_extra_args = vlm_extra_args
         self.prettify_markdown = prettify_markdown
         self.show_formula_number = show_formula_number
+        self.restructure_pages = restructure_pages
+        self.merge_tables = merge_tables
+        self.relevel_titles = relevel_titles
         self.visualize = visualize
         self.additional_params = additional_params
 
@@ -267,10 +307,13 @@ class PaddleOCRVLDocumentConverter:
             use_doc_unwarping=self.use_doc_unwarping,
             use_layout_detection=self.use_layout_detection,
             use_chart_recognition=self.use_chart_recognition,
+            use_seal_recognition=self.use_seal_recognition,
+            use_ocr_for_image_block=self.use_ocr_for_image_block,
             layout_threshold=self.layout_threshold,
             layout_nms=self.layout_nms,
             layout_unclip_ratio=self.layout_unclip_ratio,
             layout_merge_bboxes_mode=self.layout_merge_bboxes_mode,
+            layout_shape_mode=self.layout_shape_mode,
             prompt_label=self.prompt_label,
             format_block_content=self.format_block_content,
             repetition_penalty=self.repetition_penalty,
@@ -278,8 +321,15 @@ class PaddleOCRVLDocumentConverter:
             top_p=self.top_p,
             min_pixels=self.min_pixels,
             max_pixels=self.max_pixels,
+            max_new_tokens=self.max_new_tokens,
+            merge_layout_blocks=self.merge_layout_blocks,
+            markdown_ignore_labels=self.markdown_ignore_labels,
+            vlm_extra_args=self.vlm_extra_args,
             prettify_markdown=self.prettify_markdown,
             show_formula_number=self.show_formula_number,
+            restructure_pages=self.restructure_pages,
+            merge_tables=self.merge_tables,
+            relevel_titles=self.relevel_titles,
             visualize=self.visualize,
             additional_params=self.additional_params,
         )
@@ -331,6 +381,10 @@ class PaddleOCRVLDocumentConverter:
             request_data["useLayoutDetection"] = self.use_layout_detection
         if self.use_chart_recognition is not None:
             request_data["useChartRecognition"] = self.use_chart_recognition
+        if self.use_seal_recognition is not None:
+            request_data["useSealRecognition"] = self.use_seal_recognition
+        if self.use_ocr_for_image_block is not None:
+            request_data["useOcrForImageBlock"] = self.use_ocr_for_image_block
         if self.layout_threshold is not None:
             request_data["layoutThreshold"] = self.layout_threshold
         if self.layout_nms is not None:
@@ -339,6 +393,8 @@ class PaddleOCRVLDocumentConverter:
             request_data["layoutUnclipRatio"] = self.layout_unclip_ratio
         if self.layout_merge_bboxes_mode is not None:
             request_data["layoutMergeBboxesMode"] = self.layout_merge_bboxes_mode
+        if self.layout_shape_mode is not None:
+            request_data["layoutShapeMode"] = self.layout_shape_mode
         if self.prompt_label is not None:
             request_data["promptLabel"] = self.prompt_label
         if self.format_block_content is not None:
@@ -353,10 +409,24 @@ class PaddleOCRVLDocumentConverter:
             request_data["minPixels"] = self.min_pixels
         if self.max_pixels is not None:
             request_data["maxPixels"] = self.max_pixels
+        if self.max_new_tokens is not None:
+            request_data["maxNewTokens"] = self.max_new_tokens
+        if self.merge_layout_blocks is not None:
+            request_data["mergeLayoutBlocks"] = self.merge_layout_blocks
+        if self.markdown_ignore_labels is not None:
+            request_data["markdownIgnoreLabels"] = self.markdown_ignore_labels
+        if self.vlm_extra_args is not None:
+            request_data["vlmExtraArgs"] = self.vlm_extra_args
         if self.prettify_markdown is not None:
             request_data["prettifyMarkdown"] = self.prettify_markdown
         if self.show_formula_number is not None:
             request_data["showFormulaNumber"] = self.show_formula_number
+        if self.restructure_pages is not None:
+            request_data["restructurePages"] = self.restructure_pages
+        if self.merge_tables is not None:
+            request_data["mergeTables"] = self.merge_tables
+        if self.relevel_titles is not None:
+            request_data["relevelTitles"] = self.relevel_titles
         if self.visualize is not None:
             request_data["visualize"] = self.visualize
         if self.additional_params is not None:
@@ -372,7 +442,7 @@ class PaddleOCRVLDocumentConverter:
 
         # Prepare headers with authentication
         access_token_value = self.access_token.resolve_value() if self.access_token else None
-        headers = {"Content-Type": "application/json"}
+        headers = {"Content-Type": "application/json", "Client-Platform": "haystack"}
         if access_token_value:
             headers["Authorization"] = f"token {access_token_value}"
 
@@ -421,8 +491,8 @@ class PaddleOCRVLDocumentConverter:
     @component.output_types(documents=list[Document], raw_paddleocr_responses=list[dict[str, Any]])
     def run(
         self,
-        sources: list[Union[str, Path, ByteStream]],
-        meta: Optional[Union[dict[str, Any], list[dict[str, Any]]]] = None,
+        sources: list[str | Path | ByteStream],
+        meta: dict[str, Any] | list[dict[str, Any]] | None = None,
     ) -> dict[str, Any]:
         """
         Convert image or PDF files to Documents.
@@ -448,7 +518,7 @@ class PaddleOCRVLDocumentConverter:
 
         meta_list = normalize_metadata(meta, sources_count=len(sources))
 
-        for source, metadata in zip(sources, meta_list):
+        for source, metadata in zip(sources, meta_list, strict=True):
             try:
                 bytestream = get_bytestream_from_source(source)
             except Exception as e:

{paddleocr_haystack-0.1.0 → paddleocr_haystack-1.1.0}/tests/test_paddleocr_vl_document_converter.py

@@ -42,8 +42,8 @@ class TestPaddleOCRVLDocumentConverter:
         assert converter.access_token == Secret.from_env_var("AISTUDIO_ACCESS_TOKEN")
         assert converter.api_url == "http://test-api-url.com"
         assert converter.file_type is None
-        assert converter.use_doc_orientation_classify is None
-        assert converter.use_doc_unwarping is None
+        assert converter.use_doc_orientation_classify is False
+        assert converter.use_doc_unwarping is False
         assert converter.use_layout_detection is None
         assert converter.use_chart_recognition is None
         assert converter.layout_threshold is None
@@ -126,14 +126,17 @@ class TestPaddleOCRVLDocumentConverter:
                     "type": "env_var",
                 },
                 "file_type": None,
-                "use_doc_orientation_classify": None,
-                "use_doc_unwarping": None,
+                "use_doc_orientation_classify": False,
+                "use_doc_unwarping": False,
                 "use_layout_detection": None,
                 "use_chart_recognition": None,
+                "use_seal_recognition": None,
+                "use_ocr_for_image_block": None,
                 "layout_threshold": None,
                 "layout_nms": None,
                 "layout_unclip_ratio": None,
                 "layout_merge_bboxes_mode": None,
+                "layout_shape_mode": None,
                 "prompt_label": None,
                 "format_block_content": None,
                 "repetition_penalty": None,
@@ -141,8 +144,15 @@ class TestPaddleOCRVLDocumentConverter:
                 "top_p": None,
                 "min_pixels": None,
                 "max_pixels": None,
+                "max_new_tokens": None,
+                "merge_layout_blocks": None,
+                "markdown_ignore_labels": None,
+                "vlm_extra_args": None,
                 "prettify_markdown": None,
                 "show_formula_number": None,
+                "restructure_pages": None,
+                "merge_tables": None,
+                "relevel_titles": None,
                 "visualize": None,
                 "additional_params": None,
             },
@@ -158,10 +168,13 @@ class TestPaddleOCRVLDocumentConverter:
             use_doc_unwarping=False,
             use_layout_detection=True,
             use_chart_recognition=False,
+            use_seal_recognition=None,
+            use_ocr_for_image_block=None,
             layout_threshold=0.7,
             layout_nms=False,
             layout_unclip_ratio=2.0,
             layout_merge_bboxes_mode="separate",
+            layout_shape_mode=None,
             prompt_label="formula",
             format_block_content=False,
             repetition_penalty=1.2,
@@ -169,8 +182,15 @@ class TestPaddleOCRVLDocumentConverter:
             top_p=0.95,
             min_pixels=200,
             max_pixels=2000,
+            max_new_tokens=None,
+            merge_layout_blocks=None,
+            markdown_ignore_labels=None,
+            vlm_extra_args=None,
             prettify_markdown=True,
             show_formula_number=True,
+            restructure_pages=None,
+            merge_tables=None,
+            relevel_titles=None,
             visualize=False,
             additional_params={},
         )
@@ -190,10 +210,13 @@ class TestPaddleOCRVLDocumentConverter:
                 "use_doc_unwarping": False,
                 "use_layout_detection": True,
                 "use_chart_recognition": False,
+                "use_seal_recognition": None,
+                "use_ocr_for_image_block": None,
                 "layout_threshold": 0.7,
                 "layout_nms": False,
                 "layout_unclip_ratio": 2.0,
                 "layout_merge_bboxes_mode": "separate",
+                "layout_shape_mode": None,
                 "prompt_label": "formula",
                 "format_block_content": False,
                 "repetition_penalty": 1.2,
@@ -201,8 +224,15 @@ class TestPaddleOCRVLDocumentConverter:
                 "top_p": 0.95,
                 "min_pixels": 200,
                 "max_pixels": 2000,
+                "max_new_tokens": None,
+                "merge_layout_blocks": None,
+                "markdown_ignore_labels": None,
+                "vlm_extra_args": None,
                 "prettify_markdown": True,
                 "show_formula_number": True,
+                "restructure_pages": None,
+                "merge_tables": None,
+                "relevel_titles": None,
                 "visualize": False,
                 "additional_params": {},
             },