BabelDOC 0.4.22__tar.gz → 0.5.1__tar.gz

{babeldoc-0.4.22 → babeldoc-0.5.1}/.github/workflows/codeql.yml

@@ -55,7 +55,7 @@ jobs:
         # your codebase is analyzed, see https://docs.github.com/en/code-security/code-scanning/creating-an-advanced-setup-for-code-scanning/codeql-code-scanning-for-compiled-languages
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v4
+      uses: actions/checkout@v5
 
     # Initializes the CodeQL tools for scanning.
     - name: Initialize CodeQL

{babeldoc-0.4.22 → babeldoc-0.5.1}/.github/workflows/docs.yml

@@ -9,7 +9,7 @@ jobs:
   deploy:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           fetch-depth: 0
       - name: Configure Git Credentials
@@ -17,7 +17,7 @@ jobs:
           git config user.name github-actions[bot]
           git config user.email 41898282+github-actions[bot]@users.noreply.github.com
       - name: Setup uv with Python 3.12
-        uses: astral-sh/setup-uv@e92bafb6253dcd438e0484186d7669ea7a8ca1cc # v6.4.3
+        uses: astral-sh/setup-uv@d9e0f98d3fc6adb07d1e3d37f3043649ddad06a1 # v6.5.0
         with:
           python-version: "3.12"
           enable-cache: true

{babeldoc-0.4.22 → babeldoc-0.5.1}/.github/workflows/labeler.yml

@@ -22,7 +22,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Check out the repository
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
 
       - name: Run Labeler
         uses: crazy-max/ghaction-github-labeler@24d110aa46a59976b8a7f35518cb7f14f434c916 # v5.3.0

{babeldoc-0.4.22 → babeldoc-0.5.1}/.github/workflows/lint.yml

@@ -10,7 +10,7 @@ jobs:
       fail-fast: false
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
       - name: Ruff
         uses: astral-sh/ruff-action@v3
       - name: AutoCorrect

{babeldoc-0.4.22 → babeldoc-0.5.1}/.github/workflows/pr-lint.yml

@@ -9,7 +9,7 @@ jobs:
     name: runner / ruff
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
       
       - name: Install Python
         uses: actions/setup-python@v5
@@ -34,7 +34,7 @@ jobs:
     name: runner / autocorrect
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
       - name: AutoCorrect
         uses: huacnlee/autocorrect-action@bf91ab3904c2908dd8e71312a8a83ed1eb632997 # v2.13.3
       - name: Report ReviewDog

{babeldoc-0.4.22 → babeldoc-0.5.1}/.github/workflows/publish-to-pypi.yml

@@ -28,14 +28,14 @@ jobs:
     outputs:
       is_release: ${{ steps.check-version.outputs.tag }}
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           persist-credentials: true
           fetch-depth: 2
           token: ${{ secrets.GITHUB_TOKEN }}
           
       - name: Setup uv with Python 3.12
-        uses: astral-sh/setup-uv@e92bafb6253dcd438e0484186d7669ea7a8ca1cc # v6.4.3
+        uses: astral-sh/setup-uv@d9e0f98d3fc6adb07d1e3d37f3043649ddad06a1 # v6.5.0
         with:
           python-version: "3.12"
           enable-cache: true
@@ -90,7 +90,7 @@ jobs:
 
     steps:
       - name: Download all the dists
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        uses: actions/download-artifact@634f93cb2916e3fdff6788551b99b062d0335ce0 # v5.0.0
         with:
           name: python-package-distributions
           path: dist/
@@ -114,7 +114,7 @@ jobs:
 
     steps:
       - name: Download all the dists
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+        uses: actions/download-artifact@634f93cb2916e3fdff6788551b99b062d0335ce0 # v5.0.0
         with:
           name: python-package-distributions
           path: dist/
@@ -139,7 +139,7 @@ jobs:
       contents: write
       pull-requests: write
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           persist-credentials: true
           fetch-depth: 2

{babeldoc-0.4.22 → babeldoc-0.5.1}/.github/workflows/test.yml

@@ -18,7 +18,7 @@ jobs:
         python-version: ["3.10", "3.11", "3.12", "3.13"]
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v5
         with:
           persist-credentials: false
       - name: Cached Assets
@@ -28,7 +28,7 @@ jobs:
           path: ~/.cache/babeldoc
           key: babeldoc-assets-${{ hashFiles('babeldoc/assets/embedding_assets_metadata.py') }}
       - name: Setup uv with Python ${{ matrix.python-version }}
-        uses: astral-sh/setup-uv@e92bafb6253dcd438e0484186d7669ea7a8ca1cc # v6.4.3
+        uses: astral-sh/setup-uv@d9e0f98d3fc6adb07d1e3d37f3043649ddad06a1 # v6.5.0
         with:
           python-version: ${{ matrix.python-version }}
           enable-cache: true

{babeldoc-0.4.22 → babeldoc-0.5.1}/.gitignore

@@ -40,3 +40,7 @@ babeldoc/format/txt
 
 # uv
 uv.lock
+
+# Claude Code memory file
+CLAUDE.md
+/.claude

{babeldoc-0.4.22 → babeldoc-0.5.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BabelDOC
-Version: 0.4.22
+Version: 0.5.1
 Summary: Yet Another Document Translator
 Project-URL: Homepage, https://github.com/funstory-ai/BabelDOC
 Project-URL: Issues, https://github.com/funstory-ai/BabelDOC/issues
@@ -218,6 +218,13 @@ uv run babeldoc --files example.pdf --files example2.pdf --openai --openai-model
 - `--auto-enable-ocr-workaround`: Enable automatic OCR workaround (default: False). If a document is detected as heavily scanned, this will attempt to enable OCR processing and skip further scan detection. See "Important Interaction Note" below for crucial details on how this interacts with `--ocr-workaround` and `--skip-scanned-detection`.
 - `--primary-font-family`: Override primary font family for translated text. Choices: 'serif' for serif fonts, 'sans-serif' for sans-serif fonts, 'script' for script/italic fonts. If not specified, uses automatic font selection based on original text properties.
 - `--only-include-translated-page`: Only include translated pages in the output PDF. This option is only effective when `--pages` is used. (default: False)
+- `--merge-alternating-line-numbers`: Enable post-processing to merge alternating line-number layouts (keep the number paragraph as an independent paragraph b; merge adjacent text paragraphs a and c across it when `layout_id` and `xobj_id` match, digits are ASCII and spaces only). Default: off.
+- `--skip-form-render`: Skip form rendering (default: False). When enabled, PDF forms will not be rendered in the output.
+- `--skip-curve-render`: Skip curve rendering (default: False). When enabled, PDF curves will not be rendered in the output.
+- `--only-parse-generate-pdf`: Only parse PDF and generate output PDF without translation (default: False). This skips all translation-related processing including layout analysis, paragraph finding, style processing, and translation itself. Useful for testing PDF parsing and reconstruction functionality.
+- `--remove-non-formula-lines`: Remove non-formula lines from paragraph areas (default: False). This removes decorative lines that are not part of formulas, while protecting lines in figure/table areas. Useful for cleaning up documents with decorative elements that interfere with text flow.
+- `--non-formula-line-iou-threshold`: IoU threshold for detecting paragraph overlap when removing non-formula lines (default: 0.9). Higher values are more conservative and will remove fewer lines.
+- `--figure-table-protection-threshold`: IoU threshold for protecting lines in figure/table areas when removing non-formula lines (default: 0.9). Higher values provide more protection for structural elements in figures and tables.
 
 - `--rpc-doclayout`: RPC service host address for document layout analysis (default: None)
 - `--working-dir`: Working directory for translation. If not set, use temp directory.
@@ -338,6 +345,12 @@ ocr_workaround = false # Use OCR workaround for scanned PDFs
 rpc_doclayout = "" # RPC service host for document layout analysis
 working_dir = "" # Working directory for translation
 auto_enable_ocr_workaround = false # Enable automatic OCR workaround for scanned PDFs. See docs for interaction with ocr_workaround and skip_scanned_detection.
+skip_form_render = false # Skip form rendering (default: False)
+skip_curve_render = false # Skip curve rendering (default: False)
+only_parse_generate_pdf = false # Only parse PDF and generate output PDF without translation (default: False)
+remove_non_formula_lines = false # Remove non-formula lines from paragraph areas (default: False)
+non_formula_line_iou_threshold = 0.2 # IoU threshold for paragraph overlap detection (default: 0.2)
+figure_table_protection_threshold = 0.3 # IoU threshold for figure/table protection (default: 0.3)
 
 # Translation service
 openai = true
@@ -363,48 +376,10 @@ report-interval = 0.5
 
 ## Python API
 
-> [!TIP]
->
-> 1. Before pdf2zh 2.0 is released, you can temporarily use BabelDOC's Python API. However, after pdf2zh 2.0 is released, please directly use pdf2zh's Python API.
->
-> 2. This project's Python API does not guarantee any compatibility. However, the Python API from pdf2zh will guarantee a certain level of compatibility.
->
-> 3. We do not provide any technical support for the BabelDOC API.
->
-> 4. When performing secondary development, please refer to [pdf2zh 2.0 high level](https://github.com/PDFMathTranslate/PDFMathTranslate-next/blob/main/pdf2zh_next/high_level.py) and ensure that BabelDOC runs in a subprocess.
-
-You can refer to the example in [main.py](https://github.com/funstory-ai/yadt/blob/main/babeldoc/main.py) to use BabelDOC's Python API.
-
-Please note:
-
-1. Make sure call `babeldoc.format.pdf.high_level.init()` before using the API
-
-2. The current `TranslationConfig` does not fully validate input parameters, so you need to ensure the validity of input parameters
-
-3. For offline assets management, you can use the following functions:
-   ```python
-   # Generate an offline assets package
-   from pathlib import Path
-   import babeldoc.assets.assets
-   
-   # Generate package to a specific directory
-   # path is optional, default is ~/.cache/babeldoc/assets/offline_assets_{hash}.zip
-   babeldoc.assets.assets.generate_offline_assets_package(Path("/path/to/output/dir"))
-   
-   # Restore from a package file
-   # path is optional, default is ~/.cache/babeldoc/assets/offline_assets_{hash}.zip
-   babeldoc.assets.assets.restore_offline_assets_package(Path("/path/to/offline_assets_package.zip"))
-   
-   # You can also restore from a directory containing the offline assets package
-   # The tool will automatically find the correct package file based on the hash
-   babeldoc.assets.assets.restore_offline_assets_package(Path("/path/to/directory"))
-   ```
+The current recommended way to call BabelDOC in Python is to call the `high_level.do_translate_async_stream` function of [pdf2zh next](https://github.com/PDFMathTranslate/PDFMathTranslate-next).
 
-> [!TIP]
-> 
-> 1. The offline assets package name cannot be modified because the file list hash is encoded in the name.
-> 2. When using in production environments, it's recommended to pre-generate the assets package and include it with your application distribution.
-> 3. The package verification ensures that all required assets are intact and match their expected checksums.
+> [!WARNING]
+> **All APIs of BabelDOC should be considered as internal APIs, and any direct use of BabelDOC is not supported.**
 
 ## Background
 
@@ -459,6 +434,11 @@ And meet the following requirements:
 
 This project uses a combination of [Semantic Versioning](https://semver.org/) and [Pride Versioning](https://pridever.org/). The version number format is: "0.MAJOR.MINOR".
 
+> [!NOTE]
+>
+> The API compatibility here mainly refers to the compatibility with [pdf2zh_next](https://github.com/PDFMathTranslate/PDFMathTranslate-next).
+
+
 - MAJOR: Incremented by 1 when API incompatible changes are made or when proud improvements are implemented.
 
 - MINOR: Incremented by 1 when any API compatible changes are made.

{babeldoc-0.4.22 → babeldoc-0.5.1}/README.md

@@ -162,6 +162,13 @@ uv run babeldoc --files example.pdf --files example2.pdf --openai --openai-model
 - `--auto-enable-ocr-workaround`: Enable automatic OCR workaround (default: False). If a document is detected as heavily scanned, this will attempt to enable OCR processing and skip further scan detection. See "Important Interaction Note" below for crucial details on how this interacts with `--ocr-workaround` and `--skip-scanned-detection`.
 - `--primary-font-family`: Override primary font family for translated text. Choices: 'serif' for serif fonts, 'sans-serif' for sans-serif fonts, 'script' for script/italic fonts. If not specified, uses automatic font selection based on original text properties.
 - `--only-include-translated-page`: Only include translated pages in the output PDF. This option is only effective when `--pages` is used. (default: False)
+- `--merge-alternating-line-numbers`: Enable post-processing to merge alternating line-number layouts (keep the number paragraph as an independent paragraph b; merge adjacent text paragraphs a and c across it when `layout_id` and `xobj_id` match, digits are ASCII and spaces only). Default: off.
+- `--skip-form-render`: Skip form rendering (default: False). When enabled, PDF forms will not be rendered in the output.
+- `--skip-curve-render`: Skip curve rendering (default: False). When enabled, PDF curves will not be rendered in the output.
+- `--only-parse-generate-pdf`: Only parse PDF and generate output PDF without translation (default: False). This skips all translation-related processing including layout analysis, paragraph finding, style processing, and translation itself. Useful for testing PDF parsing and reconstruction functionality.
+- `--remove-non-formula-lines`: Remove non-formula lines from paragraph areas (default: False). This removes decorative lines that are not part of formulas, while protecting lines in figure/table areas. Useful for cleaning up documents with decorative elements that interfere with text flow.
+- `--non-formula-line-iou-threshold`: IoU threshold for detecting paragraph overlap when removing non-formula lines (default: 0.9). Higher values are more conservative and will remove fewer lines.
+- `--figure-table-protection-threshold`: IoU threshold for protecting lines in figure/table areas when removing non-formula lines (default: 0.9). Higher values provide more protection for structural elements in figures and tables.
 
 - `--rpc-doclayout`: RPC service host address for document layout analysis (default: None)
 - `--working-dir`: Working directory for translation. If not set, use temp directory.
@@ -282,6 +289,12 @@ ocr_workaround = false # Use OCR workaround for scanned PDFs
 rpc_doclayout = "" # RPC service host for document layout analysis
 working_dir = "" # Working directory for translation
 auto_enable_ocr_workaround = false # Enable automatic OCR workaround for scanned PDFs. See docs for interaction with ocr_workaround and skip_scanned_detection.
+skip_form_render = false # Skip form rendering (default: False)
+skip_curve_render = false # Skip curve rendering (default: False)
+only_parse_generate_pdf = false # Only parse PDF and generate output PDF without translation (default: False)
+remove_non_formula_lines = false # Remove non-formula lines from paragraph areas (default: False)
+non_formula_line_iou_threshold = 0.2 # IoU threshold for paragraph overlap detection (default: 0.2)
+figure_table_protection_threshold = 0.3 # IoU threshold for figure/table protection (default: 0.3)
 
 # Translation service
 openai = true
@@ -307,48 +320,10 @@ report-interval = 0.5
 
 ## Python API
 
-> [!TIP]
->
-> 1. Before pdf2zh 2.0 is released, you can temporarily use BabelDOC's Python API. However, after pdf2zh 2.0 is released, please directly use pdf2zh's Python API.
->
-> 2. This project's Python API does not guarantee any compatibility. However, the Python API from pdf2zh will guarantee a certain level of compatibility.
->
-> 3. We do not provide any technical support for the BabelDOC API.
->
-> 4. When performing secondary development, please refer to [pdf2zh 2.0 high level](https://github.com/PDFMathTranslate/PDFMathTranslate-next/blob/main/pdf2zh_next/high_level.py) and ensure that BabelDOC runs in a subprocess.
-
-You can refer to the example in [main.py](https://github.com/funstory-ai/yadt/blob/main/babeldoc/main.py) to use BabelDOC's Python API.
-
-Please note:
-
-1. Make sure call `babeldoc.format.pdf.high_level.init()` before using the API
-
-2. The current `TranslationConfig` does not fully validate input parameters, so you need to ensure the validity of input parameters
-
-3. For offline assets management, you can use the following functions:
-   ```python
-   # Generate an offline assets package
-   from pathlib import Path
-   import babeldoc.assets.assets
-   
-   # Generate package to a specific directory
-   # path is optional, default is ~/.cache/babeldoc/assets/offline_assets_{hash}.zip
-   babeldoc.assets.assets.generate_offline_assets_package(Path("/path/to/output/dir"))
-   
-   # Restore from a package file
-   # path is optional, default is ~/.cache/babeldoc/assets/offline_assets_{hash}.zip
-   babeldoc.assets.assets.restore_offline_assets_package(Path("/path/to/offline_assets_package.zip"))
-   
-   # You can also restore from a directory containing the offline assets package
-   # The tool will automatically find the correct package file based on the hash
-   babeldoc.assets.assets.restore_offline_assets_package(Path("/path/to/directory"))
-   ```
+The current recommended way to call BabelDOC in Python is to call the `high_level.do_translate_async_stream` function of [pdf2zh next](https://github.com/PDFMathTranslate/PDFMathTranslate-next).
 
-> [!TIP]
-> 
-> 1. The offline assets package name cannot be modified because the file list hash is encoded in the name.
-> 2. When using in production environments, it's recommended to pre-generate the assets package and include it with your application distribution.
-> 3. The package verification ensures that all required assets are intact and match their expected checksums.
+> [!WARNING]
+> **All APIs of BabelDOC should be considered as internal APIs, and any direct use of BabelDOC is not supported.**
 
 ## Background
 
@@ -403,6 +378,11 @@ And meet the following requirements:
 
 This project uses a combination of [Semantic Versioning](https://semver.org/) and [Pride Versioning](https://pridever.org/). The version number format is: "0.MAJOR.MINOR".
 
+> [!NOTE]
+>
+> The API compatibility here mainly refers to the compatibility with [pdf2zh_next](https://github.com/PDFMathTranslate/PDFMathTranslate-next).
+
+
 - MAJOR: Incremented by 1 when API incompatible changes are made or when proud improvements are implemented.
 
 - MINOR: Incremented by 1 when any API compatible changes are made.

babeldoc-0.5.1/babeldoc/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.5.1"

{babeldoc-0.4.22 → babeldoc-0.5.1}/babeldoc/const.py

@@ -3,7 +3,7 @@ import shutil
 import subprocess
 from pathlib import Path
 
-__version__ = "0.4.22"
+__version__ = "0.5.1"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 

{babeldoc-0.4.22 → babeldoc-0.5.1}/babeldoc/docvision/rpc_doclayout6.py

@@ -2,6 +2,7 @@ import base64
 import json
 import logging
 import threading
+import unicodedata
 from concurrent.futures import ThreadPoolExecutor
 from pathlib import Path
 
@@ -25,6 +26,8 @@ from babeldoc.format.pdf.document_il.utils.extract_char import (
 from babeldoc.format.pdf.document_il.utils.extract_char import (
     process_page_chars_to_lines,
 )
+from babeldoc.format.pdf.document_il.utils.fontmap import FontMapper
+from babeldoc.format.pdf.document_il.utils.layout_helper import SPACE_REGEX
 from babeldoc.format.pdf.document_il.utils.mupdf_helper import get_no_rotation_img
 
 logger = logging.getLogger(__name__)
@@ -64,11 +67,21 @@ def encode_image(image) -> bytes:
         f"(Attempt {retry_state.attempt_number}/3)"
     ),
 )
+def clip_num(num: float, min_value: float, max_value: float) -> float:
+    """Clip a number to a specified range."""
+    if num < min_value:
+        return min_value
+    elif num > max_value:
+        return max_value
+    return num
+
+
 def predict_layout(
     image,
     host: str = "http://localhost:8000",
     _imgsz: int = 1024,
     lines=None,
+    font_mapper: FontMapper | None = None,
 ):
     """Predict document layout using OCR line information (RPC service)."""
 
@@ -86,20 +99,30 @@ def predict_layout(
         min_y = min(b.y for b in boxes)
         max_y = max(b.y2 for b in boxes)
 
+        image_height, image_width = image.shape[:2]
+
         # Transform to image pixel coordinates
         min_x = min_x / 72 * DPI
         max_x = max_x / 72 * DPI
         min_y = min_y / 72 * DPI
         max_y = max_y / 72 * DPI
 
-        image_height = image.shape[0]
         min_y, max_y = image_height - max_y, image_height - min_y
 
         box_volume = (max_x - min_x) * (max_y - min_y)
         if box_volume < 1:
             return None
 
-        return {"box": [min_x, min_y, max_x, max_y], "text": line.text}
+        min_x = clip_num(min_x, 0, image_width - 1)
+        max_x = clip_num(max_x, 0, image_width - 1)
+        min_y = clip_num(min_y, 0, image_height - 1)
+        max_y = clip_num(max_y, 0, image_height - 1)
+
+        filtered_text = filter_text(line.text, font_mapper)
+        if not filtered_text:
+            return None
+
+        return {"box": [min_x, min_y, max_x, max_y], "text": filtered_text}
 
     formatted_results = [convert_line(l) for l in lines]
     formatted_results = [r for r in formatted_results if r is not None]
@@ -261,6 +284,17 @@ class ResultContainer:
         self.result = YoloResult(boxes_data=np.array([]), names=[])
 
 
+def filter_text(txt: str, font_mapper: FontMapper):
+    normalize = unicodedata.normalize("NFKC", txt)
+    unicodes = []
+    for c in normalize:
+        if font_mapper.has_char(c):
+            unicodes.append(c)
+    normalize = "".join(unicodes)
+    result = SPACE_REGEX.sub(" ", normalize).strip()
+    return result
+
+
 class RpcDocLayoutModel(DocLayoutModel):
     """DocLayoutModel implementation that uses RPC service."""
 
@@ -283,6 +317,10 @@ class RpcDocLayoutModel(DocLayoutModel):
         self._stride = 32  # Default stride value
         self._names = ["text", "title", "list", "table", "figure"]
         self.lock = threading.Lock()
+        self.font_mapper = None
+
+    def init_font_mapper(self, translation_config):
+        self.font_mapper = FontMapper(translation_config)
 
     @property
     def stride(self) -> int:
@@ -356,6 +394,84 @@ class RpcDocLayoutModel(DocLayoutModel):
         boxes = (boxes - [pad_x, pad_y, pad_x, pad_y]) / gain
         return boxes
 
+    def calculate_iou(self, box1, box2):
+        """Calculate IoU between two boxes in xyxy format."""
+        x1_1, y1_1, x2_1, y2_1 = box1
+        x1_2, y1_2, x2_2, y2_2 = box2
+
+        # Calculate intersection area
+        x1_inter = max(x1_1, x1_2)
+        y1_inter = max(y1_1, y1_2)
+        x2_inter = min(x2_1, x2_2)
+        y2_inter = min(y2_1, y2_2)
+
+        if x2_inter <= x1_inter or y2_inter <= y1_inter:
+            return 0.0
+
+        intersection = (x2_inter - x1_inter) * (y2_inter - y1_inter)
+
+        # Calculate union area
+        area1 = (x2_1 - x1_1) * (y2_1 - y1_1)
+        area2 = (x2_2 - x1_2) * (y2_2 - y1_2)
+        union = area1 + area2 - intersection
+
+        return intersection / union if union > 0 else 0.0
+
+    def is_subset(self, inner_box, outer_box):
+        """Check if inner_box is a subset of outer_box."""
+        x1_inner, y1_inner, x2_inner, y2_inner = inner_box
+        x1_outer, y1_outer, x2_outer, y2_outer = outer_box
+
+        return (
+            x1_inner >= x1_outer
+            and y1_inner >= y1_outer
+            and x2_inner <= x2_outer
+            and y2_inner <= y2_outer
+        )
+
+    def expand_box_to_contain(self, box_to_expand, box_to_contain):
+        """Expand box_to_expand to fully contain box_to_contain."""
+        x1_expand, y1_expand, x2_expand, y2_expand = box_to_expand
+        x1_contain, y1_contain, x2_contain, y2_contain = box_to_contain
+
+        return [
+            min(x1_expand, x1_contain),
+            min(y1_expand, y1_contain),
+            max(x2_expand, x2_contain),
+            max(y2_expand, y2_contain),
+        ]
+
+    def post_process_boxes(self, merged_boxes: list[YoloBox], names: dict[int, str]):
+        """Post-process merged boxes to handle text and paragraph_hybrid overlaps."""
+        for i, text_box in enumerate(merged_boxes):
+            text_label = names.get(text_box.cls, "")
+            if "text" not in text_label:
+                continue
+
+            for j, para_box in enumerate(merged_boxes):
+                if i == j:
+                    continue
+
+                para_label = names.get(para_box.cls, "")
+                if "paragraph_hybrid" not in para_label:
+                    continue
+
+                # Calculate IoU
+                iou = self.calculate_iou(text_box.xyxy, para_box.xyxy)
+
+                # Check if IoU > 0.95 and paragraph is not subset of text
+                if iou > 0.95 and not self.is_subset(para_box.xyxy, text_box.xyxy):
+                    # Expand text box to contain paragraph_hybrid
+                    expanded_box = self.expand_box_to_contain(
+                        text_box.xyxy, para_box.xyxy
+                    )
+                    merged_boxes[i] = YoloBox(
+                        None,
+                        np.array(expanded_box),
+                        text_box.conf,
+                        text_box.cls,
+                    )
+
     def predict_image(
         self,
         image,
@@ -376,7 +492,12 @@ class RpcDocLayoutModel(DocLayoutModel):
         with ThreadPoolExecutor(max_workers=2) as ex:
             if lines:
                 future1 = ex.submit(
-                    predict_layout, image_proc, self.host1, imgsz, lines
+                    predict_layout,
+                    image_proc,
+                    self.host1,
+                    imgsz,
+                    lines,
+                    self.font_mapper,
                 )
             future2 = ex.submit(predict_layout2, image_proc, self.host2, imgsz)
 
@@ -429,6 +550,9 @@ class RpcDocLayoutModel(DocLayoutModel):
         # Sort boxes by confidence desc (YoloResult expects sorted list)
         merged_boxes.sort(key=lambda b: b.conf, reverse=True)
 
+        # Post-process boxes to handle text and paragraph_hybrid overlaps
+        self.post_process_boxes(merged_boxes, names)
+
         return YoloResult(boxes=merged_boxes, names=names)
 
     def predict(self, image, imgsz=1024, **kwargs) -> list[YoloResult]:  # type: ignore[override]

{babeldoc-0.4.22 → babeldoc-0.5.1}/babeldoc/format/pdf/babelpdf/base14.py

@@ -1,3 +1,5 @@
+from .encoding import win_ansi
+
 base14_bbox = {
     "Courier-BoldOblique": [
         (0, 0, 0, 0),
@@ -1965,6 +1967,15 @@ base14_name = {
 }
 
 
-def get_base14_bbox(family, name):
-    if family in base14_bbox and name in base14_name:
-        return base14_bbox[family][base14_name[name]]
+def get_base14_bbox(family):
+    bbox = []
+    if family in base14_bbox:
+        base = base14_bbox[family]
+        for name in win_ansi:
+            cur_bbox = (0, 0, 0, 0)
+            if name:
+                index = base14_name.get(name)
+                if index:
+                    cur_bbox = base[index]
+            bbox.append(cur_bbox)
+    return bbox

{babeldoc-0.4.22 → babeldoc-0.5.1}/babeldoc/format/pdf/converter.py

@@ -9,6 +9,7 @@ from babeldoc.format.pdf.document_il.frontend.il_creater import ILCreater
 from babeldoc.pdfminer.converter import PDFConverter
 from babeldoc.pdfminer.layout import LTChar
 from babeldoc.pdfminer.layout import LTComponent
+from babeldoc.pdfminer.layout import LTCurve
 from babeldoc.pdfminer.layout import LTFigure
 from babeldoc.pdfminer.layout import LTLine
 from babeldoc.pdfminer.layout import LTPage
@@ -115,6 +116,7 @@ class PDFConverterEx(PDFConverter):
             graphicstate,
             self.il_creater.xobj_id,
             font_id,
+            self.il_creater.get_render_order_and_increase(),
         )
         self.cur_item.add(item)
         item.cid = cid  # hack 插入原字符编码
@@ -139,6 +141,7 @@ class AWLTChar(LTChar):
         graphicstate: PDFGraphicState,
         xobj_id: int,
         font_id: str,
+        render_order: int,
     ) -> None:
         LTText.__init__(self)
         self._text = text
@@ -149,6 +152,7 @@ class AWLTChar(LTChar):
         self.xobj_id = xobj_id
         self.adv = textwidth * fontsize * scaling
         self.aw_font_id = font_id
+        self.render_order = render_order
         # compute the boundary rectangle.
         if font.is_vertical():
             # vertical
@@ -378,18 +382,21 @@ class TranslateConverter(PDFConverterEx):
                 # 图表
                 self.il_creater.on_pdf_figure(child)
                 pass
-            elif isinstance(child, LTLine):     # 线条
-                continue
-                layout = self.layout[ltpage.pageid]
-                # ltpage.height 可能是 fig 里面的高度，这里统一用 layout.shape
-                h, w = layout.shape
-                # 读取当前线条在 layout 中的类别
-                cx, cy = np.clip(int(child.x0), 0, w - 1), np.clip(int(child.y0), 0, h - 1)
-                cls = layout[cy, cx]
-                if vstk and cls == xt_cls:      # 公式线条
-                    vlstk.append(child)
-                else:                           # 全局线条
-                    lstk.append(child)
+            # elif isinstance(child, LTLine):     # 线条
+            #     continue
+            #     layout = self.layout[ltpage.pageid]
+            #     # ltpage.height 可能是 fig 里面的高度，这里统一用 layout.shape
+            #     h, w = layout.shape
+            #     # 读取当前线条在 layout 中的类别
+            #     cx, cy = np.clip(int(child.x0), 0, w - 1), np.clip(int(child.y0), 0, h - 1)
+            #     cls = layout[cy, cx]
+            #     if vstk and cls == xt_cls:      # 公式线条
+            #         vlstk.append(child)
+            #     else:                           # 全局线条
+            #         lstk.append(child)
+            elif isinstance(child, LTCurve):
+                self.il_creater.on_lt_curve(child)
+                pass
             else:
                 pass
         return