BabelDOC 0.4.7__tar.gz → 0.4.9__tar.gz

{babeldoc-0.4.7 → babeldoc-0.4.9}/.github/workflows/docs.yml

@@ -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@f0ec1fc3b38f5e7cd731bb6ce540c5af426746bb # v6.1.0
+        uses: astral-sh/setup-uv@445689ea25e0de0a23313031f5fe577c74ae45a1 # v6.3.0
         with:
           python-version: "3.12"
           enable-cache: true

{babeldoc-0.4.7 → babeldoc-0.4.9}/.github/workflows/publish-to-pypi.yml

@@ -35,7 +35,7 @@ jobs:
           token: ${{ secrets.GITHUB_TOKEN }}
           
       - name: Setup uv with Python 3.12
-        uses: astral-sh/setup-uv@f0ec1fc3b38f5e7cd731bb6ce540c5af426746bb # v6.1.0
+        uses: astral-sh/setup-uv@445689ea25e0de0a23313031f5fe577c74ae45a1 # v6.3.0
         with:
           python-version: "3.12"
           enable-cache: true

{babeldoc-0.4.7 → babeldoc-0.4.9}/.github/workflows/test.yml

@@ -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@f0ec1fc3b38f5e7cd731bb6ce540c5af426746bb # v6.1.0
+        uses: astral-sh/setup-uv@445689ea25e0de0a23313031f5fe577c74ae45a1 # v6.3.0
         with:
           python-version: ${{ matrix.python-version }}
           enable-cache: true

{babeldoc-0.4.7 → babeldoc-0.4.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BabelDOC
-Version: 0.4.7
+Version: 0.4.9
 Summary: Yet Another Document Translator
 Project-URL: Homepage, https://github.com/funstory-ai/BabelDOC
 Project-URL: Issues, https://github.com/funstory-ai/BabelDOC/issues
@@ -13,6 +13,7 @@ Classifier: Operating System :: OS Independent
 Classifier: Programming Language :: Python :: 3
 Requires-Python: <3.14,>=3.10
 Requires-Dist: bitstring>=4.3.0
+Requires-Dist: chardet>=5.2.0
 Requires-Dist: charset-normalizer>=2.0.0
 Requires-Dist: configargparse>=1.7
 Requires-Dist: cryptography>=36.0.0

babeldoc-0.4.9/babeldoc/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.4.9"

{babeldoc-0.4.7 → babeldoc-0.4.9}/babeldoc/const.py

@@ -3,7 +3,7 @@ import shutil
 import subprocess
 from pathlib import Path
 
-__version__ = "0.4.7"
+__version__ = "0.4.9"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 

{babeldoc-0.4.7 → babeldoc-0.4.9}/babeldoc/format/pdf/document_il/midend/typesetting.py

@@ -6,6 +6,7 @@ import unicodedata
 from functools import cache
 
 import pymupdf
+from rtree import index
 
 from babeldoc.const import WATERMARK_VERSION
 from babeldoc.format.pdf.document_il import Box
@@ -16,6 +17,7 @@ from babeldoc.format.pdf.document_il import PdfStyle
 from babeldoc.format.pdf.document_il import il_version_1
 from babeldoc.format.pdf.document_il.utils.fontmap import FontMapper
 from babeldoc.format.pdf.document_il.utils.formular_helper import update_formula_data
+from babeldoc.format.pdf.document_il.utils.layout_helper import box_to_tuple
 from babeldoc.format.pdf.translation_config import TranslationConfig
 from babeldoc.format.pdf.translation_config import WatermarkOutputMode
 
@@ -592,6 +594,62 @@ class Typesetting:
             == WatermarkOutputMode.Watermarked
         ):
             self.add_watermark(page)
+        try:
+            para_index = index.Index()
+            para_map = {}
+            #
+            valid_paras = [
+                p
+                for p in page.pdf_paragraph
+                if p.box
+                and all(c is not None for c in [p.box.x, p.box.y, p.box.x2, p.box.y2])
+            ]
+
+            for i, para in enumerate(valid_paras):
+                para_map[i] = para
+                para_index.insert(i, box_to_tuple(para.box))
+
+            for i, p_upper in para_map.items():
+                required_gap = 6
+                if not (p_upper.box and p_upper.box.y is not None):
+                    continue
+
+                check_area = il_version_1.Box(
+                    x=p_upper.box.x,
+                    y=p_upper.box.y - required_gap,
+                    x2=p_upper.box.x2,
+                    y2=p_upper.box.y,
+                )
+
+                candidate_ids = list(para_index.intersection(box_to_tuple(check_area)))
+
+                conflicting_paras = []
+                for para_id in candidate_ids:
+                    if para_id == i:
+                        continue
+                    p_lower = para_map[para_id]
+                    if not (
+                        p_lower.box
+                        and p_upper.box
+                        and p_lower.box.x2 < p_upper.box.x
+                        or p_lower.box.x > p_upper.box.x2
+                    ):
+                        conflicting_paras.append(p_lower)
+
+                if conflicting_paras:
+                    max_y2 = max(
+                        p.box.y2
+                        for p in conflicting_paras
+                        if p.box and p.box.y2 is not None
+                    )
+
+                    new_y = max_y2 + required_gap
+                    if p_upper.box and new_y < p_upper.box.y2:
+                        p_upper.box.y = new_y
+        except Exception as e:
+            logger.warning(
+                f"Failed to adjust paragraph positions on page {page.page_number}: {e}"
+            )
         # 开始实际的渲染过程
         for paragraph in page.pdf_paragraph:
             self.render_paragraph(paragraph, page, fonts)
@@ -887,7 +945,7 @@ class Typesetting:
             if scale < 0.7 and min_line_spacing > 1.1:
                 if expand_space_flag == 0:
                     # 先尝试向下扩展
-                    min_y = self.get_max_bottom_space(box, page)
+                    min_y = self.get_max_bottom_space(box, page) + 2
                     if min_y < box.y:
                         expanded_box = Box(
                             x=box.x,
@@ -902,7 +960,7 @@ class Typesetting:
                     continue
                 elif expand_space_flag == 1:
                     # 如果向下扩展后还不够，再尝试向右扩展
-                    max_x = self.get_max_right_space(box, page)
+                    max_x = self.get_max_right_space(box, page) - 5
                     if max_x > box.x2:
                         expanded_box = Box(
                             x=box.x,

{babeldoc-0.4.7 → babeldoc-0.4.9}/babeldoc/format/pdf/document_il/utils/formular_helper.py

@@ -197,8 +197,8 @@ def is_formulas_font(font_name: str, formular_font_pattern: str | None) -> bool:
     )
     precise_formula_font_pattern = (
         r"^("
-        r"|.*CambriaMath.*"
-        r"|.*Cambria Math.*"
+        # r"|.*CambriaMath.*"
+        # r"|.*Cambria Math.*"
         r"|.*Asana.*"
         r"|.*MiriamMonoCLM-BookOblique.*"
         r"|.*Miriam Mono CLM.*"

{babeldoc-0.4.7 → babeldoc-0.4.9}/babeldoc/glossary.py

@@ -6,6 +6,7 @@ import re
 import time
 from pathlib import Path
 
+import chardet
 import hyperscan
 import regex
 
@@ -134,8 +135,11 @@ class Glossary:
         normalized_target_lang_out = target_lang_out.lower().replace("-", "_")
 
         try:
-            with file_path.open("r", encoding="utf-8") as f:
-                reader = csv.DictReader(f, doublequote=True)
+            with file_path.open("rb") as f:
+                content = f.read()
+                encoding = chardet.detect(content)["encoding"]
+                buffer = io.StringIO(content.decode(encoding))
+                reader = csv.DictReader(buffer, doublequote=True)
                 if not all(col in reader.fieldnames for col in ["source", "target"]):
                     raise ValueError(
                         f"CSV file {file_path} must contain 'source' and 'target' columns."

{babeldoc-0.4.7 → babeldoc-0.4.9}/babeldoc/main.py

@@ -21,7 +21,7 @@ from babeldoc.translator.translator import OpenAITranslator
 from babeldoc.translator.translator import set_translate_rate_limiter
 
 logger = logging.getLogger(__name__)
-__version__ = "0.4.7"
+__version__ = "0.4.9"
 
 
 def create_parser():

{babeldoc-0.4.7 → babeldoc-0.4.9}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "BabelDOC"
-version = "0.4.7"
+version = "0.4.9"
 description = "Yet Another Document Translator"
 license = "AGPL-3.0"
 readme = "README.md"
@@ -47,7 +47,8 @@ dependencies = [
     "rapidocr-onnxruntime>=1.4.4",
     "pyzstd>=0.17.0",
     "hyperscan>=0.7.13",
-    "rtree>=1.4.0"
+    "rtree>=1.4.0",
+    "chardet>=5.2.0",
 ]
 
 [project.optional-dependencies]
@@ -157,7 +158,7 @@ pythonpath = [".", "src"]
 testpaths = ["tests"]
 
 [bumpver]
-current_version = "0.4.7"
+current_version = "0.4.9"
 version_pattern = "MAJOR.MINOR.PATCH[.PYTAGNUM]"
 
 [bumpver.file_patterns]

{babeldoc-0.4.7 → babeldoc-0.4.9}/uv.lock

@@ -48,10 +48,11 @@ wheels = [
 
 [[package]]
 name = "babeldoc"
-version = "0.4.7"
+version = "0.4.9"
 source = { editable = "." }
 dependencies = [
     { name = "bitstring" },
+    { name = "chardet" },
     { name = "charset-normalizer" },
     { name = "configargparse" },
     { name = "cryptography" },
@@ -114,6 +115,7 @@ dev = [
 [package.metadata]
 requires-dist = [
     { name = "bitstring", specifier = ">=4.3.0" },
+    { name = "chardet", specifier = ">=5.2.0" },
     { name = "charset-normalizer", specifier = ">=2.0.0" },
     { name = "configargparse", specifier = ">=1.7" },
     { name = "cryptography", specifier = ">=36.0.0" },
@@ -368,6 +370,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/c5/55/51844dd50c4fc7a33b653bfaba4c2456f06955289ca770a5dbd5fd267374/cfgv-3.4.0-py2.py3-none-any.whl", hash = "sha256:b7265b1f29fd3316bfcd2b330d63d024f2bfd8bcb8b0272f8e19a504856c48f9", size = 7249, upload-time = "2023-08-12T20:38:16.269Z" },
 ]
 
+[[package]]
+name = "chardet"
+version = "5.2.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f3/0d/f7b6ab21ec75897ed80c17d79b15951a719226b9fababf1e40ea74d69079/chardet-5.2.0.tar.gz", hash = "sha256:1b3b6ff479a8c414bc3fa2c0852995695c4a026dcd6d0633b2dd092ca39c1cf7", size = 2069618, upload-time = "2023-08-01T19:23:02.662Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/38/6f/f5fbc992a329ee4e0f288c1fe0e2ad9485ed064cac731ed2fe47dcc38cbf/chardet-5.2.0-py3-none-any.whl", hash = "sha256:e1cf59446890a00105fe7b7912492ea04b6e6f06d4b742b2c788469e34c82970", size = 199385, upload-time = "2023-08-01T19:23:00.661Z" },
+]
+
 [[package]]
 name = "charset-normalizer"
 version = "3.4.2"

babeldoc-0.4.7/babeldoc/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.4.7"