BabelDOC 0.2.2__tar.gz → 0.2.4__tar.gz

{babeldoc-0.2.2 → babeldoc-0.2.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BabelDOC
-Version: 0.2.2
+Version: 0.2.4
 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-0.2.4/babeldoc/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.2.4"

{babeldoc-0.2.2 → babeldoc-0.2.4}/babeldoc/const.py

@@ -3,7 +3,7 @@ import shutil
 import subprocess
 from pathlib import Path
 
-__version__ = "0.2.2"
+__version__ = "0.2.4"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 

{babeldoc-0.2.2 → babeldoc-0.2.4}/babeldoc/document_il/midend/il_translator_llm_only.py

@@ -79,7 +79,13 @@ class ILTranslatorLLMOnly:
         # count total paragraph
         total = sum(
             [
-                len([p for p in page.pdf_paragraph if p.debug_id is not None])
+                len(
+                    [
+                        p
+                        for p in page.pdf_paragraph
+                        if p.debug_id is not None and p.unicode is not None
+                    ]
+                )
                 for page in docs.page
             ]
         )
@@ -134,7 +140,7 @@ class ILTranslatorLLMOnly:
 
         total_unicode_counts = 0
         for paragraph in page.pdf_paragraph:
-            if paragraph.debug_id is None:
+            if paragraph.debug_id is None or paragraph.unicode is None:
                 continue
             # self.translate_paragraph(paragraph, pbar,tracker.new_paragraph(), page_font_map, page_xobj_font_map)
             total_unicode_counts += len(paragraph.unicode)

{babeldoc-0.2.2 → babeldoc-0.2.4}/babeldoc/document_il/midend/typesetting.py

@@ -854,9 +854,15 @@ class Typesetting:
                     ],
                 )
             elif composition.pdf_same_style_unicode_characters:
-                font_id = (
-                    composition.pdf_same_style_unicode_characters.pdf_style.font_id
-                )
+                style = composition.pdf_same_style_unicode_characters.pdf_style
+                if style is None:
+                    logger.warning(
+                        f"Style is None. "
+                        f"Composition: {composition}. "
+                        f"Paragraph: {paragraph}. ",
+                    )
+                    continue
+                font_id = style.font_id
                 font = get_font(font_id, paragraph.xobj_id)
                 result.extend(
                     [
@@ -867,8 +873,8 @@ class Typesetting:
                                 char_unicode,
                             ),
                             original_font=font,
-                            font_size=composition.pdf_same_style_unicode_characters.pdf_style.font_size,
-                            style=composition.pdf_same_style_unicode_characters.pdf_style,
+                            font_size=style.font_size,
+                            style=style,
                             xobj_id=paragraph.xobj_id,
                             debug_info=composition.pdf_same_style_unicode_characters.debug_info,
                         )

{babeldoc-0.2.2 → babeldoc-0.2.4}/babeldoc/high_level.py

@@ -304,8 +304,11 @@ def do_translate(pm, translation_config):
 
         # 修复 pdf 文件中 xref 为 null 的情况
         for i in range(1, doc_pdf2zh.xref_length()):
-            if doc_pdf2zh.xref_object(i) == "null":
-                doc_pdf2zh.update_object(i, "null")
+            try:
+                if doc_pdf2zh.xref_object(i) == "null":
+                    doc_pdf2zh.update_object(i, "null")
+            except Exception:
+                logger.warning(f"try fix xref {i} fail, continue")
 
         for page in doc_pdf2zh:
             page.insert_font(resfont, None)

{babeldoc-0.2.2 → babeldoc-0.2.4}/babeldoc/main.py

@@ -23,7 +23,7 @@ from babeldoc.translation_config import TranslationConfig
 from babeldoc.translation_config import WatermarkOutputMode
 
 logger = logging.getLogger(__name__)
-__version__ = "0.2.2"
+__version__ = "0.2.4"
 
 
 def create_parser():

{babeldoc-0.2.2 → babeldoc-0.2.4}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "BabelDOC"
-version = "0.2.2"
+version = "0.2.4"
 description = "Yet Another Document Translator"
 license = "AGPL-3.0"
 readme = "README.md"
@@ -136,7 +136,7 @@ pythonpath = [".", "src"]
 testpaths = ["tests"]
 
 [bumpver]
-current_version = "0.2.2"
+current_version = "0.2.4"
 version_pattern = "MAJOR.MINOR.PATCH[.PYTAGNUM]"
 
 [bumpver.file_patterns]

{babeldoc-0.2.2 → babeldoc-0.2.4}/uv.lock

@@ -48,7 +48,7 @@ wheels = [
 
 [[package]]
 name = "babeldoc"
-version = "0.2.2"
+version = "0.2.4"
 source = { editable = "." }
 dependencies = [
     { name = "bitstring" },

babeldoc-0.2.2/babeldoc/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.2.2"