BabelDOC 0.3.13__tar.gz → 0.3.15__tar.gz

{babeldoc-0.3.13 → babeldoc-0.3.15}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BabelDOC
-Version: 0.3.13
+Version: 0.3.15
 Summary: Yet Another Document Translator
 Project-URL: Homepage, https://github.com/funstory-ai/BabelDOC
 Project-URL: Issues, https://github.com/funstory-ai/BabelDOC/issues

babeldoc-0.3.15/babeldoc/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.3.15"

{babeldoc-0.3.13 → babeldoc-0.3.15}/babeldoc/const.py

@@ -3,7 +3,7 @@ import shutil
 import subprocess
 from pathlib import Path
 
-__version__ = "0.3.13"
+__version__ = "0.3.15"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 

{babeldoc-0.3.13 → babeldoc-0.3.15}/babeldoc/document_il/frontend/il_creater.py

@@ -426,6 +426,37 @@ def get_bbox(bbox, size, c, x, y):
     return pymupdf.Quad(ll, lr, ul, ur)
 
 
+# 常见 Unicode 空格字符的代码点
+unicode_spaces = [
+    "\u0020",  # 半角空格
+    "\u00a0",  # 不间断空格
+    "\u1680",  # Ogham 空格标记
+    "\u2000",  # En Quad
+    "\u2001",  # Em Quad
+    "\u2002",  # En Space
+    "\u2003",  # Em Space
+    "\u2004",  # 三分之一 Em 空格
+    "\u2005",  # 四分之一 Em 空格
+    "\u2006",  # 六分之一 Em 空格
+    "\u2007",  # 数样间距
+    "\u2008",  # 行首前导空格
+    "\u2009",  # 瘦弱空格
+    "\u200a",  # hair space
+    "\u202f",  # 窄不间断空格
+    "\u205f",  # 数学中等空格
+    "\u3000",  # 全角空格
+    "\u200b",  # 零宽度空格
+    "\u2060",  # 零宽度非断空格
+    "\t",  # 水平制表符
+]
+
+# 构建正则表达式
+pattern = "^[" + "".join(unicode_spaces) + "]+$"
+
+# 编译正则
+space_regex = re.compile(pattern)
+
+
 class ILCreater:
     stage_name = "Parse PDF and Create Intermediate Representation"
 
@@ -805,7 +836,7 @@ class ILCreater:
         char_unicode = char.get_text()
         if "(cid:" not in char_unicode and len(char_unicode) > 1:
             return
-        if char_unicode == "\t":
+        if space_regex.match(char_unicode):
             char_unicode = " "
         advance = char.adv
         bbox = il_version_1.Box(

{babeldoc-0.3.13 → babeldoc-0.3.15}/babeldoc/main.py

@@ -23,7 +23,7 @@ from babeldoc.translation_config import TranslationConfig
 from babeldoc.translation_config import WatermarkOutputMode
 
 logger = logging.getLogger(__name__)
-__version__ = "0.3.13"
+__version__ = "0.3.15"
 
 
 def create_parser():
@@ -266,8 +266,8 @@ async def main():
         return
 
     # 验证翻译服务选择
-    if not (args.openai or args.translate):
-        parser.error("必须选择一个翻译服务：--openai 或 --translate")
+    if not args.openai:
+        parser.error("必须选择一个翻译服务：--openai")
 
     # 验证 OpenAI 参数
     if args.openai and not args.openai_api_key:

{babeldoc-0.3.13 → babeldoc-0.3.15}/babeldoc/pdfinterp.py

@@ -393,6 +393,8 @@ class PDFPageInterpreterEx(PDFPageInterpreter):
         gs.passthrough_instruction = (
             self.il_creater.passthrough_per_char_instruction.copy()
         )
+        if isinstance(seq, int) or isinstance(seq, float):
+            seq = [seq]
         self.device.render_string(self.textstate, cast(PDFTextSeq, seq), self.ncs, gs)
         return
 

{babeldoc-0.3.13 → babeldoc-0.3.15}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "BabelDOC"
-version = "0.3.13"
+version = "0.3.15"
 description = "Yet Another Document Translator"
 license = "AGPL-3.0"
 readme = "README.md"
@@ -148,7 +148,7 @@ pythonpath = [".", "src"]
 testpaths = ["tests"]
 
 [bumpver]
-current_version = "0.3.13"
+current_version = "0.3.15"
 version_pattern = "MAJOR.MINOR.PATCH[.PYTAGNUM]"
 
 [bumpver.file_patterns]

{babeldoc-0.3.13 → babeldoc-0.3.15}/uv.lock

@@ -48,7 +48,7 @@ wheels = [
 
 [[package]]
 name = "babeldoc"
-version = "0.3.13"
+version = "0.3.15"
 source = { editable = "." }
 dependencies = [
     { name = "bitstring" },

babeldoc-0.3.13/babeldoc/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.3.13"