BabelDOC 0.3.28__tar.gz → 0.3.30__tar.gz

{babeldoc-0.3.28 → babeldoc-0.3.30}/.github/ISSUE_TEMPLATE/bug_report.yaml

@@ -13,6 +13,10 @@ body:
           required: true
         - label: I confirmed this is not caused by a network issue
           required: true
+        - label: I have fully read and understood the [README](https://github.com/funstory-ai/BabelDOC/blob/main/README.md)
+          required: true
+        - label: I am certain that this issue is with BabelDOC itself and can be reproduced through the BabelDOC cli
+          required: true
 
   - type: markdown
     attributes:

{babeldoc-0.3.28 → babeldoc-0.3.30}/.github/PULL_REQUEST_TEMPLATE/pr_form.yml

@@ -52,7 +52,7 @@ body:
     attributes:
       label: Contributor Checklist
       options:
-        - label: I’ve read the **CONTRIBUTING.md** guide
+        - label: I’ve fully read and understood the **[CONTRIBUTING.md](https://funstory-ai.github.io/BabelDOC/CONTRIBUTING/)** guide
           required: true
         - label: My changes follow the project’s code style and guidelines
           required: true

{babeldoc-0.3.28 → babeldoc-0.3.30}/.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@d4b2f3b6ecc6e67c4457f6d3e41ec42d3d0fcb86 # v5.4.2
+        uses: astral-sh/setup-uv@c7f87aa956e4c323abf06d5dec078e358f6b4d04 # v6.0.0
         with:
           python-version: "3.12"
           enable-cache: true

{babeldoc-0.3.28 → babeldoc-0.3.30}/.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@d4b2f3b6ecc6e67c4457f6d3e41ec42d3d0fcb86 # v5.4.2
+        uses: astral-sh/setup-uv@c7f87aa956e4c323abf06d5dec078e358f6b4d04 # v6.0.0
         with:
           python-version: "3.12"
           enable-cache: true
@@ -89,7 +89,7 @@ jobs:
 
     steps:
       - name: Download all the dists
-        uses: actions/download-artifact@95815c38cf2ff2164869cbab79da8d1f422bc89e # v4.2.1
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
         with:
           name: python-package-distributions
           path: dist/
@@ -113,7 +113,7 @@ jobs:
 
     steps:
       - name: Download all the dists
-        uses: actions/download-artifact@95815c38cf2ff2164869cbab79da8d1f422bc89e # v4.2.1
+        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
         with:
           name: python-package-distributions
           path: dist/

{babeldoc-0.3.28 → babeldoc-0.3.30}/.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@d4b2f3b6ecc6e67c4457f6d3e41ec42d3d0fcb86 # v5.4.2
+        uses: astral-sh/setup-uv@c7f87aa956e4c323abf06d5dec078e358f6b4d04 # v6.0.0
         with:
           python-version: ${{ matrix.python-version }}
           enable-cache: true

{babeldoc-0.3.28 → babeldoc-0.3.30}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BabelDOC
-Version: 0.3.28
+Version: 0.3.30
 Summary: Yet Another Document Translator
 Project-URL: Homepage, https://github.com/funstory-ai/BabelDOC
 Project-URL: Issues, https://github.com/funstory-ai/BabelDOC/issues
@@ -306,6 +306,10 @@ report-interval = 0.5
 > 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/awwaawwa/PDFMathTranslate/blob/v2-rc/pdf2zh/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.
 

{babeldoc-0.3.28 → babeldoc-0.3.30}/README.md

@@ -257,6 +257,10 @@ report-interval = 0.5
 > 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/awwaawwa/PDFMathTranslate/blob/v2-rc/pdf2zh/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.
 

babeldoc-0.3.30/babeldoc/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.3.30"

{babeldoc-0.3.28 → babeldoc-0.3.30}/babeldoc/const.py

@@ -3,7 +3,7 @@ import shutil
 import subprocess
 from pathlib import Path
 
-__version__ = "0.3.28"
+__version__ = "0.3.30"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 

{babeldoc-0.3.28 → babeldoc-0.3.30}/babeldoc/document_il/midend/typesetting.py

@@ -115,7 +115,7 @@ class TypesettingUnit:
             r"\u0500-\u052F"  # Cyrillic Supplement
             r"\u0370-\u03FF"  # Greek and Coptic
             r"\u2DE0-\u2DFF"  # Cyrillic Extended-A
-            r"\uA650-\uA69F"  # Cyrillic Extended-B   
+            r"\uA650-\uA69F"  # Cyrillic Extended-B
             r"\u1200-\u137F"  # Ethiopic
             r"\u1380-\u139F"  # Ethiopic Supplement
             r"\u2D80-\u2DDF"  # Ethiopic Extended
@@ -138,9 +138,18 @@ class TypesettingUnit:
             r"\u2D00-\u2D2F"  # Georgian Supplement
             r"\u1780-\u17FF"  # Khmer
             r"\u19E0-\u19FF"  # Khmer Symbols
+            r"\U00010B00-\U00010B3F"  # Avestan
+            r"\u1D00-\u1D7F"  # Phonetic Extensions
+            r"\u1400-\u167F"  # Unified Canadian Aboriginal Syllabics
+            r"\u0B00-\u0B7F"  # Oriya
+            r"\u0780-\u07BF"  # Thaana
+            r"\U0001E900-\U0001E95F"  # Adlam
+            r"\u1C80-\u1C8F"  # Cyrillic Extended-C
+            r"\U0001E030-\U0001E08F"  # Cyrillic Extended-D
             r"'"
             r"-"  # Hyphen
             r"·"  # Middle Dot (U+00B7) For Català
+            r"ʻ"  # Spacing Modifier Letters U+02BB
             r"]+$",
             unicode,
         ):
@@ -148,7 +157,7 @@ class TypesettingUnit:
         return True
 
     @property
-    def is_chinese_char(self):
+    def is_cjk_char(self):
         if self.formular:
             return False
         unicode = self.try_get_unicode()
@@ -185,6 +194,30 @@ class TypesettingUnit:
         ]:
             return True
         if unicode:
+            if re.match(
+                r"^["
+                r"\u3000-\u303f"  # CJK Symbols and Punctuation
+                r"\u3040-\u309f"  # Hiragana
+                r"\u30a0-\u30ff"  # Katakana
+                r"\u3100-\u312f"  # Bopomofo
+                r"\uac00-\ud7af"  # Hangul Syllables
+                r"\u1100-\u11ff"  # Hangul Jamo
+                r"\u3130-\u318f"  # Hangul Compatibility Jamo
+                r"\ua960-\ua97f"  # Hangul Jamo Extended-A
+                r"\ud7b0-\ud7ff"  # Hangul Jamo Extended-B
+                r"\u3190-\u319f"  # Kanbun
+                r"\u3200-\u32ff"  # Enclosed CJK Letters and Months
+                r"\u3300-\u33ff"  # CJK Compatibility
+                r"\ufe30-\ufe4f"  # CJK Compatibility Forms
+                r"\u4e00-\u9fff"  # CJK Unified Ideographs
+                r"\u2e80-\u2eff"  # CJK Radicals Supplement
+                r"\u31c0-\u31ef"  # CJK Strokes
+                r"\u2f00-\u2fdf"  # Kangxi Radicals
+                r"\ufe10-\ufe1f"  # Vertical Forms
+                r"]+$",
+                unicode,
+            ):
+                return True
             try:
                 unicodedata_name = unicodedata.name(unicode)
                 return (
@@ -696,7 +729,7 @@ class Typesetting:
 
             if (
                 last_unit  # 有上一个单元
-                and last_unit.is_chinese_char ^ unit.is_chinese_char  # 中英文交界处
+                and last_unit.is_cjk_char ^ unit.is_cjk_char  # 中英文交界处
                 and (
                     last_unit.box
                     and last_unit.box.y

{babeldoc-0.3.28 → babeldoc-0.3.30}/babeldoc/main.py

@@ -23,7 +23,7 @@ from babeldoc.translation_config import TranslationConfig
 from babeldoc.translation_config import WatermarkOutputMode
 
 logger = logging.getLogger(__name__)
-__version__ = "0.3.28"
+__version__ = "0.3.30"
 
 
 def create_parser():

{babeldoc-0.3.28 → babeldoc-0.3.30}/docs/CONTRIBUTING.md

@@ -39,6 +39,8 @@ Please pay special attention to:
 > You can contact the maintainers in the pdf2zh discussion group.
 > 
 > Due to the current high rate of code changes, this project only accepts small PRs. If you would like to suggest a change and you include a patch as a proof-of-concept, that would be great. However, please do not be offended if we rewrite your patch from scratch.
+>
+> Please discuss with the maintainers in the [Issue](https://github.com/funstory-ai/BabelDOC/issues) before sending a PR.
 
 [//]: # (> We welcome pull requests and will review your contributions.)
 

{babeldoc-0.3.28 → babeldoc-0.3.30}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "BabelDOC"
-version = "0.3.28"
+version = "0.3.30"
 description = "Yet Another Document Translator"
 license = "AGPL-3.0"
 readme = "README.md"
@@ -150,7 +150,7 @@ pythonpath = [".", "src"]
 testpaths = ["tests"]
 
 [bumpver]
-current_version = "0.3.28"
+current_version = "0.3.30"
 version_pattern = "MAJOR.MINOR.PATCH[.PYTAGNUM]"
 
 [bumpver.file_patterns]