BabelDOC 0.1.12__tar.gz → 0.1.14__tar.gz

{babeldoc-0.1.12 → babeldoc-0.1.14}/.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@4db96194c378173c656ce18a155ffc14a9fc4355 # v5.2.2
+        uses: astral-sh/setup-uv@1edb52594c857e2b5b13128931090f0640537287 # v5.3.0
         with:
           python-version: "3.12"
           enable-cache: true

{babeldoc-0.1.12 → babeldoc-0.1.14}/.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@4db96194c378173c656ce18a155ffc14a9fc4355 # v5.2.2
+        uses: astral-sh/setup-uv@1edb52594c857e2b5b13128931090f0640537287 # v5.3.0
         with:
           python-version: "3.12"
           enable-cache: true
@@ -68,7 +68,7 @@ jobs:
         run: "uv build"
 
       - name: Store the distribution packages
-        uses: actions/upload-artifact@v4.6.0
+        uses: actions/upload-artifact@v4.6.1
         with:
           name: python-package-distributions
           path: dist/

{babeldoc-0.1.12 → babeldoc-0.1.14}/.github/workflows/test.yml

@@ -28,7 +28,7 @@ jobs:
           path: ~/.cache/huggingface/hub/models--wybxc--DocLayout-YOLO-DocStructBench-onnx
           key: ${{runner.os}}-huggingmodel-wybxc--DocLayout-YOLO-DocStructBench-onnx
       - name: Setup uv with Python ${{ matrix.python-version }}
-        uses: astral-sh/setup-uv@4db96194c378173c656ce18a155ffc14a9fc4355 # v5.2.2
+        uses: astral-sh/setup-uv@1edb52594c857e2b5b13128931090f0640537287 # v5.3.0
         with:
           python-version: ${{ matrix.python-version }}
           enable-cache: true

{babeldoc-0.1.12 → babeldoc-0.1.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BabelDOC
-Version: 0.1.12
+Version: 0.1.14
 Summary: Yet Another Document Translator
 Project-URL: Homepage, https://github.com/funstory-ai/BabelDOC
 Project-URL: Issues, https://github.com/funstory-ai/BabelDOC/issues
@@ -23,6 +23,7 @@ Requires-Dist: opencv-python>=4.10.0.84
 Requires-Dist: orjson>=3.10.14
 Requires-Dist: pdfminer-six>=20240706
 Requires-Dist: peewee>=3.17.8
+Requires-Dist: pydantic>=2.10.6
 Requires-Dist: pymupdf>=1.25.1
 Requires-Dist: rich>=13.9.4
 Requires-Dist: toml>=0.10.2
@@ -142,8 +143,7 @@ uv run babeldoc --bing --files example.pdf --files example2.pdf
 - `--dual-translate-first`: Put translated pages first in dual PDF mode (default: original pages first)
 - `--disable-rich-text-translate`: Disable rich text translation (may help improve compatibility with some PDFs)
 - `--enhance-compatibility`: Enable all compatibility enhancement options (equivalent to --skip-clean --dual-translate-first --disable-rich-text-translate)
-- `--use-side-by-side-dual`: Use side-by-side mode for dual PDF (default). Original and translated pages are shown side by side on the same page.
-- `--use-alternating-pages-dual`: Use alternating pages mode for dual PDF. Original and translated pages are arranged in alternate order.
+- `--use-alternating-pages-dual`: Use alternating pages mode for dual PDF. When enabled, original and translated pages are arranged in alternate order. When disabled (default), original and translated pages are shown side by side on the same page.
 
 > [!TIP]
 > - Both `--skip-clean` and `--dual-translate-first` may help improve compatibility with some PDF readers

{babeldoc-0.1.12 → babeldoc-0.1.14}/README.md

@@ -110,8 +110,7 @@ uv run babeldoc --bing --files example.pdf --files example2.pdf
 - `--dual-translate-first`: Put translated pages first in dual PDF mode (default: original pages first)
 - `--disable-rich-text-translate`: Disable rich text translation (may help improve compatibility with some PDFs)
 - `--enhance-compatibility`: Enable all compatibility enhancement options (equivalent to --skip-clean --dual-translate-first --disable-rich-text-translate)
-- `--use-side-by-side-dual`: Use side-by-side mode for dual PDF (default). Original and translated pages are shown side by side on the same page.
-- `--use-alternating-pages-dual`: Use alternating pages mode for dual PDF. Original and translated pages are arranged in alternate order.
+- `--use-alternating-pages-dual`: Use alternating pages mode for dual PDF. When enabled, original and translated pages are arranged in alternate order. When disabled (default), original and translated pages are shown side by side on the same page.
 
 > [!TIP]
 > - Both `--skip-clean` and `--dual-translate-first` may help improve compatibility with some PDF readers

babeldoc-0.1.14/babeldoc/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.1.14"

{babeldoc-0.1.12 → babeldoc-0.1.14}/babeldoc/const.py

@@ -2,12 +2,12 @@ import shutil
 import subprocess
 from pathlib import Path
 
-__version__ = "0.1.12"
+__version__ = "0.1.14"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 
 
-def get_cache_file_path(filename):
+def get_cache_file_path(filename: str) -> Path:
     return CACHE_FOLDER / filename
 
 
@@ -24,5 +24,5 @@ try:
         .strip()
         .decode()
     )
-except (OSError, FileNotFoundError):
+except (OSError, FileNotFoundError, subprocess.CalledProcessError):
     WATERMARK_VERSION = f"v{__version__}"

{babeldoc-0.1.12 → babeldoc-0.1.14}/babeldoc/document_il/backend/pdf_creater.py

@@ -1,5 +1,8 @@
 import logging
+import os
 import re
+import time
+from multiprocessing import Process
 from pathlib import Path
 
 import pymupdf
@@ -16,6 +19,25 @@ SUBSET_FONT_STAGE_NAME = "Subset font"
 SAVE_PDF_STAGE_NAME = "Save PDF"
 
 
+def _subset_fonts_process(pdf_path, output_path):
+    """Function to run in subprocess for font subsetting.
+
+    Args:
+        pdf_path: Path to the PDF file to subset
+        output_path: Path where to save the result
+    """
+    try:
+        pdf = pymupdf.open(pdf_path)
+        pdf.subset_fonts(fallback=False)
+        pdf.save(output_path)
+        # 返回0表示成功
+        os._exit(0)
+    except Exception as e:
+        logger.error(f"Error in font subsetting subprocess: {e}")
+        # 返回1表示失败
+        os._exit(1)
+
+
 class PDFCreater:
     stage_name = "Generate drawing instructions"
 
@@ -203,21 +225,37 @@ class PDFCreater:
             if translation_config.dual_translate_first:
                 # Show translated page on left and original on right
                 rect_left, rect_right = rect_right, rect_left
-
-            # Show original page on left and translated on right (default)
-            dual_page.show_pdf_page(
-                rect_left,
-                original_pdf,
-                page_id,
-                keep_proportion=True,
-            )
-            dual_page.show_pdf_page(
-                rect_right,
-                translated_pdf,
-                page_id,
-                keep_proportion=True,
-            )
-
+            try:
+                # Show original page on left and translated on right (default)
+                dual_page.show_pdf_page(
+                    rect_left,
+                    original_pdf,
+                    page_id,
+                    keep_proportion=True,
+                )
+            except Exception as e:
+                logger.warning(
+                    f"Failed to show original page on left and translated on right (default). "
+                    f"Page ID: {page_id}. "
+                    f"Original PDF: {self.original_pdf_path}. "
+                    f"Translated PDF: {translation_config.input_file}. ",
+                    exc_info=e,
+                )
+            try:
+                dual_page.show_pdf_page(
+                    rect_right,
+                    translated_pdf,
+                    page_id,
+                    keep_proportion=True,
+                )
+            except Exception as e:
+                logger.warning(
+                    f"Failed to show translated page on left and original on right. "
+                    f"Page ID: {page_id}. "
+                    f"Original PDF: {self.original_pdf_path}. "
+                    f"Translated PDF: {translation_config.input_file}. ",
+                    exc_info=e,
+                )
         return dual
 
     def create_alternating_pages_dual_pdf(
@@ -334,7 +372,83 @@ class PDFCreater:
             # no additional information is needed
             pdf.update_stream(int(resource_xref_id), draw_op.tobytes())
         translation_config.raise_if_cancelled()
-        pdf.subset_fonts(fallback=False)
+
+        # 使用子进程进行字体子集化
+        if not translation_config.skip_clean:
+            pdf = self.subset_fonts_in_subprocess(pdf, translation_config)
+        return pdf
+
+    def subset_fonts_in_subprocess(
+        self, pdf: pymupdf.Document, translation_config: TranslationConfig
+    ) -> pymupdf.Document:
+        """Run font subsetting in a subprocess with timeout.
+
+        Args:
+            pdf: The PDF document object
+            translation_config: Translation configuration
+
+        Returns:
+            Path to the PDF with subsetted fonts, or original path if subsetting failed or timed out
+        """
+        original_pdf = pdf
+        # Create temporary file paths
+        temp_input = str(
+            translation_config.get_working_file_path(
+                f"temp_subset_input_{int(time.time())}.pdf"
+            )
+        )
+        temp_output = str(
+            translation_config.get_working_file_path(
+                f"temp_subset_output_{int(time.time())}.pdf"
+            )
+        )
+
+        # Save PDF to temporary file without subsetting
+        pdf.save(temp_input)
+
+        # Create and start subprocess
+        process = Process(target=_subset_fonts_process, args=(temp_input, temp_output))
+        process.start()
+
+        # Wait for subprocess with timeout (1 minute)
+        timeout = 60  # 1 minutes in seconds
+        start_time = time.time()
+
+        while process.is_alive():
+            if time.time() - start_time > timeout:
+                logger.warning(
+                    f"Font subsetting timeout after {timeout} seconds, terminating subprocess"
+                )
+                process.terminate()
+                try:
+                    process.join(5)  # Give it 5 seconds to clean up
+                    if process.is_alive():
+                        logger.warning("Subprocess did not terminate, killing it")
+                        process.kill()
+                except Exception as e:
+                    logger.error(f"Error terminating font subsetting process: {e}")
+
+                return original_pdf
+
+            time.sleep(0.5)  # Check every half second
+
+        # Process completed, check exit code
+        exit_code = process.exitcode
+        success = exit_code == 0
+
+        # Check if subsetting was successful
+        if (
+            success
+            and Path(temp_output).exists()
+            and Path(temp_output).stat().st_size > 0
+        ):
+            logger.info("Font subsetting completed successfully")
+            return pymupdf.open(temp_output)
+        else:
+            logger.warning(
+                f"Font subsetting failed with exit code {exit_code} or produced empty file"
+            )
+            return original_pdf
 
     def write(self, translation_config: TranslationConfig) -> TranslateResult:
         basename = Path(translation_config.input_file).stem
@@ -449,7 +563,8 @@ class PDFCreater:
             1,
         ) as pbar:
             if not translation_config.skip_clean:
-                pdf.subset_fonts(fallback=False)
+                pdf = self.subset_fonts_in_subprocess(pdf, translation_config)
+
             pbar.advance()
         with self.translation_config.progress_monitor.stage_start(
             SAVE_PDF_STAGE_NAME,
@@ -466,11 +581,11 @@ class PDFCreater:
                 translation_config.raise_if_cancelled()
                 pdf.save(
                     mono_out_path,
-                    garbage=3,
+                    garbage=1,
                     deflate=True,
                     clean=not translation_config.skip_clean,
                     deflate_fonts=True,
-                    linear=True,
+                    linear=False,
                 )
             pbar.advance()
             dual_out_path = None
@@ -482,34 +597,30 @@ class PDFCreater:
                 original_pdf = pymupdf.open(self.original_pdf_path)
                 translated_pdf = pdf
 
-                # Choose between side-by-side and alternating pages format
+                # Choose between alternating pages and side-by-side format
                 # Default to side-by-side if not specified
-                use_side_by_side = getattr(
-                    translation_config,
-                    "use_side_by_side_dual",
-                    True,
-                )
+                use_alternating_pages = translation_config.use_alternating_pages_dual
 
-                if use_side_by_side:
-                    # Create a dual PDF with side-by-side pages (original and translation)
-                    dual = self.create_side_by_side_dual_pdf(
-                        original_pdf,
+                if use_alternating_pages:
+                    # Create a dual PDF with alternating pages (original and translation)
+                    dual = self.create_alternating_pages_dual_pdf(
+                        self.original_pdf_path,
                         translated_pdf,
-                        dual_out_path,
                         translation_config,
                     )
                 else:
-                    # Create a dual PDF with alternating pages (original and translation)
-                    dual = self.create_alternating_pages_dual_pdf(
-                        self.original_pdf_path,
+                    # Create a dual PDF with side-by-side pages (original and translation)
+                    dual = self.create_side_by_side_dual_pdf(
+                        original_pdf,
                         translated_pdf,
+                        dual_out_path,
                         translation_config,
                     )
 
                 if translation_config.debug:
                     translation_config.raise_if_cancelled()
                     try:
-                        self.write_debug_info(dual, translation_config)
+                        dual = self.write_debug_info(dual, translation_config)
                     except Exception:
                         logger.warning(
                             "Failed to write debug info to dual PDF",
@@ -518,11 +629,11 @@ class PDFCreater:
 
                 dual.save(
                     dual_out_path,
-                    garbage=3,
+                    garbage=1,
                     deflate=True,
                     clean=not translation_config.skip_clean,
                     deflate_fonts=True,
-                    linear=True,
+                    linear=False,
                 )
                 if translation_config.debug:
                     translation_config.raise_if_cancelled()

{babeldoc-0.1.12 → babeldoc-0.1.14}/babeldoc/document_il/midend/il_translator.py

@@ -303,7 +303,7 @@ class ILTranslator:
                         composition.pdf_same_style_characters.pdf_style,
                         paragraph.pdf_style,
                     )
-                    # 字号差异在0.7-1.3之间，可能是首字母变大效果，无需占位符
+                    # 字号差异在 0.7-1.3 之间，可能是首字母变大效果，无需占位符
                     or is_same_style_except_size(
                         composition.pdf_same_style_characters.pdf_style,
                         paragraph.pdf_style,
@@ -342,7 +342,7 @@ class ILTranslator:
                 )
                 return None
 
-        # 如果占位符数量超过50，且未禁用富文本翻译，则递归调用并禁用富文本翻译
+        # 如果占位符数量超过 50，且未禁用富文本翻译，则递归调用并禁用富文本翻译
         if len(placeholders) > 50 and not disable_rich_text_translate:
             logger.warning(
                 f"Too many placeholders ({len(placeholders)}) in paragraph[{paragraph.debug_id}], "

{babeldoc-0.1.12 → babeldoc-0.1.14}/babeldoc/document_il/midend/paragraph_finder.py

@@ -254,9 +254,9 @@ class ParagraphFinder:
         self,
         char: PdfCharacter,
         page: Page,
-        xy_mode: Literal["topleft"]
-        | Literal["bottomright"]
-        | Literal["middle"] = "middle",
+        xy_mode: (
+            Literal["topleft"] | Literal["bottomright"] | Literal["middle"]
+        ) = "middle",
     ):
         tl, br, md = [
             self._get_layout(char, page, mode)
@@ -279,9 +279,9 @@ class ParagraphFinder:
         self,
         char: PdfCharacter,
         page: Page,
-        xy_mode: Literal["topleft"]
-        | Literal["bottomright"]
-        | Literal["middle"] = "middle",
+        xy_mode: (
+            Literal["topleft"] | Literal["bottomright"] | Literal["middle"]
+        ) = "middle",
     ):
         # 这几个符号，解析出来的大小经常只有实际大小的一点点。
         # if (

{babeldoc-0.1.12 → babeldoc-0.1.14}/babeldoc/document_il/midend/styles_and_formulas.py

@@ -59,7 +59,7 @@ class StylesAndFormulas:
         if not page.pdf_paragraph:
             return
 
-        # 收集该页所有的公式字体ID
+        # 收集该页所有的公式字体 ID
         formula_font_ids = set()
         for font in page.pdf_font:
             if self.is_formulas_font(font.name):
@@ -95,7 +95,7 @@ class StylesAndFormulas:
                         or char.pdf_style.font_id in formula_font_ids  # 公式字体
                         or char.vertical  # 垂直字体
                         or (
-                            #   如果是程序添加的dummy空格
+                            #   如果是程序添加的 dummy 空格
                             char.char_unicode is None and in_formula_state
                         )
                     )
@@ -248,7 +248,7 @@ class StylesAndFormulas:
             # 更新基准样式为所有样式的交集
             base_style = self._merge_styles(base_style, style)
 
-        # 如果font_id或font_size为None，则使用众数
+        # 如果 font_id 或 font_size 为 None，则使用众数
         if base_style.font_id is None:
             base_style.font_id = self._get_mode_value([s.font_id for s in styles])
         if base_style.font_size is None:
@@ -286,7 +286,7 @@ class StylesAndFormulas:
         )
 
     def _merge_graphic_states(self, state1, state2):
-        """合并两个GraphicState，返回它们的交集"""
+        """合并两个 GraphicState，返回它们的交集"""
         if state1 is None:
             return state2
         if state2 is None:
@@ -350,7 +350,7 @@ class StylesAndFormulas:
         )
 
     def process_page_offsets(self, page: Page):
-        """计算公式的x和y偏移量"""
+        """计算公式的 x 和 y 偏移量"""
         if not page.pdf_paragraph:
             return
 
@@ -358,7 +358,7 @@ class StylesAndFormulas:
             if not paragraph.pdf_paragraph_composition:
                 continue
 
-            # 计算该段落的行间距，用其80%作为容差
+            # 计算该段落的行间距，用其 80% 作为容差
             line_spacing = self.calculate_line_spacing(paragraph)
             y_tolerance = line_spacing * 0.8
 
@@ -374,7 +374,7 @@ class StylesAndFormulas:
                 for j in range(i - 1, -1, -1):
                     comp = paragraph.pdf_paragraph_composition[j]
                     if comp.pdf_line:
-                        # 检查y坐标是否接近，判断是否在同一行
+                        # 检查 y 坐标是否接近，判断是否在同一行
                         if abs(comp.pdf_line.box.y - formula.box.y) <= y_tolerance:
                             left_line = comp.pdf_line
                             break
@@ -383,22 +383,22 @@ class StylesAndFormulas:
                 for j in range(i + 1, len(paragraph.pdf_paragraph_composition)):
                     comp = paragraph.pdf_paragraph_composition[j]
                     if comp.pdf_line:
-                        # 检查y坐标是否接近，判断是否在同一行
+                        # 检查 y 坐标是否接近，判断是否在同一行
                         if abs(comp.pdf_line.box.y - formula.box.y) <= y_tolerance:
                             right_line = comp.pdf_line
                             break
 
-                # 计算x偏移量（相对于左边文本）
+                # 计算 x 偏移量（相对于左边文本）
                 if left_line:
                     formula.x_offset = formula.box.x - left_line.box.x2
                 else:
-                    formula.x_offset = 0  # 如果左边没有文字，x_offset应该为0
+                    formula.x_offset = 0  # 如果左边没有文字，x_offset 应该为 0
                 if abs(formula.x_offset) < 0.1:
                     formula.x_offset = 0
                 if formula.x_offset > 0:
                     formula.x_offset = 0
 
-                # 计算y偏移量
+                # 计算 y 偏移量
                 if left_line:
                     # 使用底部坐标计算偏移量
                     formula.y_offset = formula.box.y - left_line.box.y
@@ -415,7 +415,7 @@ class StylesAndFormulas:
         if not paragraph.pdf_paragraph_composition:
             return 0.0
 
-        # 收集所有文本行的y坐标
+        # 收集所有文本行的 y 坐标
         line_y_positions = []
         for comp in paragraph.pdf_paragraph_composition:
             if comp.pdf_line:
@@ -424,7 +424,7 @@ class StylesAndFormulas:
         if len(line_y_positions) < 2:
             return 10.0  # 如果只有一行或没有行，返回一个默认值
 
-        # 计算相邻行之间的y差值
+        # 计算相邻行之间的 y 差值
         line_spacings = []
         for i in range(len(line_y_positions) - 1):
             spacing = abs(line_y_positions[i] - line_y_positions[i + 1])
@@ -562,7 +562,7 @@ class StylesAndFormulas:
         # 必须包含逗号
         if "," not in text:
             return False
-        # 检查是否包含除了数字和[]之外的其他符号
+        # 检查是否包含除了数字和 [] 之外的其他符号
         text_without_basic = re.sub(r"[0-9\[\],\s]", "", text)
         return bool(text_without_basic)
 
@@ -570,7 +570,7 @@ class StylesAndFormulas:
         self,
         formula: PdfFormula,
     ) -> list[tuple[list[PdfCharacter], PdfCharacter]]:
-        """按逗号拆分公式字符，返回(字符组, 逗号字符)的列表，最后一组的逗号字符为None。
+        """按逗号拆分公式字符，返回 (字符组，逗号字符) 的列表，最后一组的逗号字符为 None。
         只有不在括号内的逗号才会被用作分隔符。支持的括号对包括：
         - (cid:8) 和 (cid:9)
         - ( 和 )
@@ -606,7 +606,7 @@ class StylesAndFormulas:
         """合并两个公式，保持字符的相对位置"""
         # 合并所有字符
         all_chars = formula1.pdf_character + formula2.pdf_character
-        # 按y坐标和x坐标排序，确保字符顺序正确
+        # 按 y 坐标和 x 坐标排序，确保字符顺序正确
         sorted_chars = sorted(all_chars, key=lambda c: (c.box.y, c.box.x))
 
         merged_formula = PdfFormula(pdf_character=sorted_chars)
@@ -615,7 +615,7 @@ class StylesAndFormulas:
 
     def merge_overlapping_formulas(self, page: Page):
         """
-        合并x轴重叠且y轴有交集的相邻公式
+        合并 x 轴重叠且 y 轴有交集的相邻公式
         角标可能会被识别成单独的公式，需要合并
         """
         if not page.pdf_paragraph:
@@ -638,7 +638,7 @@ class StylesAndFormulas:
                 formula1 = comp1.pdf_formula
                 formula2 = comp2.pdf_formula
 
-                # 检查x轴重叠和y轴交集
+                # 检查 x 轴重叠和 y 轴交集
                 if self.is_x_axis_contained(
                     formula1.box,
                     formula2.box,
@@ -650,18 +650,18 @@ class StylesAndFormulas:
                     )
                     # 删除第二个公式
                     del paragraph.pdf_paragraph_composition[i + 1]
-                    # 不增加i，因为合并后的公式可能还需要和下一个公式合并
+                    # 不增加 i，因为合并后的公式可能还需要和下一个公式合并
                 else:
                     i += 1
 
     def is_x_axis_contained(self, box1: Box, box2: Box) -> bool:
-        """判断box1的x轴是否完全包含在box2的x轴内，或反之"""
+        """判断 box1 的 x 轴是否完全包含在 box2 的 x 轴内，或反之"""
         return (box1.x >= box2.x and box1.x2 <= box2.x2) or (
             box2.x >= box1.x and box2.x2 <= box1.x2
         )
 
     def has_y_intersection(self, box1: Box, box2: Box) -> bool:
-        """判断两个box的y轴是否有交集"""
+        """判断两个 box 的 y 轴是否有交集"""
         return not (box1.y2 < box2.y or box2.y2 < box1.y)
 
     def process_comma_formulas(self, page: Page):

{babeldoc-0.1.12 → babeldoc-0.1.14}/babeldoc/document_il/midend/typesetting.py

@@ -53,7 +53,11 @@ class TypesettingUnit:
             )
 
             self.font = font
-            self.font_id = font.font_id
+            if font is not None:
+                self.font_id = font.font_id
+            else:
+                self.font_id = "base"
+                self.unicode = " "
             self.font_size = font_size
             self.style = style
             self.xobj_id = xobj_id
@@ -398,9 +402,9 @@ class Typesetting:
             font_size=6,
             graphic_state=il_version_1.GraphicState(),
         )
-        text = f"本文档由funstory.ai的开源PDF翻译库BabelDOC {WATERMARK_VERSION} (http://yadt.io)翻译，本仓库正在积极的建设当中，欢迎star和关注。"
+        text = f"本文档由 funstory.ai 的开源 PDF 翻译库 BabelDOC {WATERMARK_VERSION} (http://yadt.io) 翻译，本仓库正在积极的建设当中，欢迎 star 和关注。"
         if self.translation_config.debug:
-            text += "\n 当前为DEBUG模式，将显示更多辅助信息。请注意，部分框的位置对应原文，但在译文中可能不正确。"
+            text += "\n 当前为 DEBUG 模式，将显示更多辅助信息。请注意，部分框的位置对应原文，但在译文中可能不正确。"
         page.pdf_paragraph.append(
             il_version_1.PdfParagraph(
                 first_line_indent=False,
@@ -463,7 +467,7 @@ class Typesetting:
             line_spacing: 行间距
 
         Returns:
-            tuple[list[TypesettingUnit], bool]: (已布局的排版单元列表, 是否所有单元都放得下)
+            tuple[list[TypesettingUnit], bool]: (已布局的排版单元列表，是否所有单元都放得下)
         """
         # 计算字号众数
         font_sizes = []
@@ -547,7 +551,7 @@ class Typesetting:
             relocated_unit = unit.relocate(current_x, current_y, scale)
             typeset_units.append(relocated_unit)
 
-            # workaround: 超长行距暂时没找到具体原因，有待进一步修复。这里的1.2是魔法数字！
+            # workaround: 超长行距暂时没找到具体原因，有待进一步修复。这里的 1.2 是魔法数字！
             # 更新当前行的最大高度
             if line_height == 0 or line_height * 1.2 > unit_height > line_height:
                 line_height = unit_height
@@ -567,7 +571,7 @@ class Typesetting:
     ):
         box = paragraph.box
         scale = 1.0
-        line_spacing = 1.5  # 初始行距为1.7
+        line_spacing = 1.5  # 初始行距为 1.7
         min_scale = 0.1  # 最小缩放因子
         min_line_spacing = 1.4  # 最小行距
         expand_space_flag = False

{babeldoc-0.1.12 → babeldoc-0.1.14}/babeldoc/document_il/translator/translator.py

@@ -5,6 +5,7 @@ import threading
 import time
 import unicodedata
 from abc import ABC
+from abc import abstractmethod
 
 import openai
 import requests
@@ -115,6 +116,7 @@ class BaseTranslator(ABC):
             self.cache.set(text, translation)
         return translation
 
+    @abstractmethod
     def do_translate(self, text):
         """
         Actual translate text, override this method