BabelDOC 0.5.22__tar.gz → 0.5.24__tar.gz

{babeldoc-0.5.22 → babeldoc-0.5.24}/.github/ISSUE_TEMPLATE/bug_report.yaml

@@ -23,6 +23,8 @@ body:
           required: true
         - label: I confirm that the latest version of BabelDOC is being used.
           required: true
+        - label: I am aware that the issue section of this project is only for submitting bugs that are clearly related to the BabelDOC core, with complete reproduction steps and relevant logs attached.** Otherwise, issues will be closed directly.
+          required: true
 
   - type: markdown
     attributes:
@@ -33,6 +35,12 @@ body:
         Please also note:
         - If you are using a downstream project like pdf2zh-next, please submit an issue directly to the downstream application. Only when you confirm that this issue is a problem with the core library of BabelDOC, submit this issue.
         - The CLI is only used for debugging purposes, we do not provide any technical support for CLI usage.
+
+  - type: markdown
+    attributes:
+      value: |
+        Please note! Users of immersive translate online services should contact customer service and provide their translation ID. **Feedback related to online services is not handled here.**
+
   - type: textarea
     id: environment
     attributes:
@@ -58,7 +66,7 @@ body:
     id: reproduce
     attributes:
       label: Steps to Reproduce
-      description: Help us reproduce the issue
+      description: Help us reproduce the issue. Issues that do not provide reproduction steps will be closed directly.
       value: |
         1. Go to '...'
         2. Click on '...'

{babeldoc-0.5.22 → babeldoc-0.5.24}/.github/workflows/codeql.yml

@@ -55,7 +55,7 @@ jobs:
         # your codebase is analyzed, see https://docs.github.com/en/code-security/code-scanning/creating-an-advanced-setup-for-code-scanning/codeql-code-scanning-for-compiled-languages
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v5
+      uses: actions/checkout@v6
 
     # Initializes the CodeQL tools for scanning.
     - name: Initialize CodeQL

{babeldoc-0.5.22 → babeldoc-0.5.24}/.github/workflows/docs.yml

@@ -9,7 +9,7 @@ jobs:
   deploy:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           fetch-depth: 0
       - name: Configure Git Credentials
@@ -24,7 +24,7 @@ jobs:
           cache-dependency-glob: "uv.lock"
           activate-environment: true
       - run: echo "cache_id=$(date --utc '+%V')" >> $GITHUB_ENV 
-      - uses: actions/cache@v4
+      - uses: actions/cache@v5
         with:
           key: mkdocs-material-${{ env.cache_id }}
           path: .cache

{babeldoc-0.5.22 → babeldoc-0.5.24}/.github/workflows/labeler.yml

@@ -22,10 +22,10 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Check out the repository
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: Run Labeler
-        uses: crazy-max/ghaction-github-labeler@24d110aa46a59976b8a7f35518cb7f14f434c916 # v5.3.0
+        uses: crazy-max/ghaction-github-labeler@548a7c3603594ec17c819e1239f281a3b801ab4d # v6.0.0
         with:
           skip-delete: true
           dry-run: ${{ github.event_name == 'pull_request' }}

{babeldoc-0.5.22 → babeldoc-0.5.24}/.github/workflows/lint.yml

@@ -10,7 +10,7 @@ jobs:
       fail-fast: false
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
       - name: Ruff
         uses: astral-sh/ruff-action@v3
       - name: AutoCorrect

{babeldoc-0.5.22 → babeldoc-0.5.24}/.github/workflows/pr-lint.yml

@@ -9,7 +9,7 @@ jobs:
     name: runner / ruff
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
       
       - name: Install Python
         uses: actions/setup-python@v6
@@ -20,7 +20,7 @@ jobs:
         run: pip install ruff
         
       - name: Install reviewdog
-        uses: reviewdog/action-setup@d8edfce3dd5e1ec6978745e801f9c50b5ef80252 # v1.4.0
+        uses: reviewdog/action-setup@d8a7baabd7f3e8544ee4dbde3ee41d0011c3a93f # v1.5.0
         with:
           reviewdog_version: latest
           
@@ -34,7 +34,7 @@ jobs:
     name: runner / autocorrect
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
       - name: AutoCorrect
         uses: huacnlee/autocorrect-action@bf91ab3904c2908dd8e71312a8a83ed1eb632997 # v2.13.3
       - name: Report ReviewDog

{babeldoc-0.5.22 → babeldoc-0.5.24}/.github/workflows/publish-to-pypi.yml

@@ -28,7 +28,7 @@ jobs:
     outputs:
       is_release: ${{ steps.check-version.outputs.tag }}
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           persist-credentials: true
           fetch-depth: 2
@@ -69,7 +69,7 @@ jobs:
         run: "uv build"
 
       - name: Store the distribution packages
-        uses: actions/upload-artifact@v4.6.2
+        uses: actions/upload-artifact@v7.0.0
         with:
           name: python-package-distributions
           path: dist/
@@ -139,7 +139,7 @@ jobs:
       contents: write
       pull-requests: write
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           persist-credentials: true
           fetch-depth: 2

{babeldoc-0.5.22 → babeldoc-0.5.24}/.github/workflows/test.yml

@@ -18,12 +18,12 @@ jobs:
         python-version: ["3.10", "3.11", "3.12", "3.13"]
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           persist-credentials: false
       - name: Cached Assets
         id: cache-assets
-        uses: actions/cache@v4.2.0
+        uses: actions/cache@v5
         with:
           path: ~/.cache/babeldoc
           key: babeldoc-assets-${{ hashFiles('babeldoc/assets/embedding_assets_metadata.py') }}

{babeldoc-0.5.22 → babeldoc-0.5.24}/.gitignore

@@ -47,3 +47,4 @@ CLAUDE.md
 babeldoc/format/playground
 temp.jpg
 AGENTS.md
+.serena

{babeldoc-0.5.22 → babeldoc-0.5.24}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BabelDOC
-Version: 0.5.22
+Version: 0.5.24
 Summary: Yet Another Document Translator
 Project-URL: Homepage, https://github.com/funstory-ai/BabelDOC
 Project-URL: Issues, https://github.com/funstory-ai/BabelDOC/issues
@@ -21,6 +21,7 @@ Requires-Dist: freetype-py>=2.5.1
 Requires-Dist: httpx[socks]>=0.27.0
 Requires-Dist: huggingface-hub>=0.27.0
 Requires-Dist: hyperscan>=0.7.13
+Requires-Dist: levenshtein>=0.27.1
 Requires-Dist: msgpack>=1.1.0
 Requires-Dist: numpy>=2.0.2
 Requires-Dist: onnx>=1.18.0
@@ -32,7 +33,6 @@ Requires-Dist: peewee>=3.17.8
 Requires-Dist: psutil>=7.0.0
 Requires-Dist: pydantic>=2.10.6
 Requires-Dist: pymupdf>=1.25.1
-Requires-Dist: python-levenshtein>=0.27.1
 Requires-Dist: pyzstd>=0.17.0
 Requires-Dist: rapidocr-onnxruntime>=1.4.4
 Requires-Dist: rich>=13.9.4
@@ -251,6 +251,7 @@ uv run babeldoc --files example.pdf --files example2.pdf --openai --openai-model
 - `--openai`: Use OpenAI for translation (default: False)
 - `--custom-system-prompt`: Custom system prompt for translation.
 - `--add-formula-placehold-hint`: Add formula placeholder hint for translation. (Currently not recommended, it may affect translation quality, default: False)
+- `--disable-same-text-fallback`: Disable fallback translation when LLM output matches input text. (default: False)
 - `--pool-max-workers`: Maximum number of worker threads for internal task processing pools. If not specified, defaults to QPS value. This parameter directly sets the worker count, replacing previous QPS-based dynamic calculations.
 - `--no-auto-extract-glossary`: Disable automatic term extraction. If this flag is present, the step is skipped. Defaults to enabled.
 
@@ -361,6 +362,7 @@ openai-model = "gpt-4o-mini"
 openai-base-url = "https://api.openai.com/v1"
 openai-api-key = "your-api-key-here"
 enable-json-mode-if-requested = false  # Enable JSON mode when requested (default: false)
+disable_same_text_fallback = false # Disable fallback translation when LLM output matches input text (default: false)
 pool-max-workers = 8  # Maximum worker threads for task processing (defaults to QPS value if not set)
 
 # Glossary Options (Optional)

{babeldoc-0.5.22 → babeldoc-0.5.24}/README.md

@@ -195,6 +195,7 @@ uv run babeldoc --files example.pdf --files example2.pdf --openai --openai-model
 - `--openai`: Use OpenAI for translation (default: False)
 - `--custom-system-prompt`: Custom system prompt for translation.
 - `--add-formula-placehold-hint`: Add formula placeholder hint for translation. (Currently not recommended, it may affect translation quality, default: False)
+- `--disable-same-text-fallback`: Disable fallback translation when LLM output matches input text. (default: False)
 - `--pool-max-workers`: Maximum number of worker threads for internal task processing pools. If not specified, defaults to QPS value. This parameter directly sets the worker count, replacing previous QPS-based dynamic calculations.
 - `--no-auto-extract-glossary`: Disable automatic term extraction. If this flag is present, the step is skipped. Defaults to enabled.
 
@@ -305,6 +306,7 @@ openai-model = "gpt-4o-mini"
 openai-base-url = "https://api.openai.com/v1"
 openai-api-key = "your-api-key-here"
 enable-json-mode-if-requested = false  # Enable JSON mode when requested (default: false)
+disable_same_text_fallback = false # Disable fallback translation when LLM output matches input text (default: false)
 pool-max-workers = 8  # Maximum worker threads for task processing (defaults to QPS value if not set)
 
 # Glossary Options (Optional)

babeldoc-0.5.24/babeldoc/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.5.24"

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/const.py

@@ -6,7 +6,7 @@ import subprocess
 import threading
 from pathlib import Path
 
-__version__ = "0.5.22"
+__version__ = "0.5.24"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/docvision/doclayout.py

@@ -37,6 +37,8 @@ os_name = platform.system()
 
 
 class OnnxModel(DocLayoutModel):
+    _FIXED_IMGSZ = 1024  # fixed input size for static-shape CoreML
+
     def __init__(self, model_path: str):
         self.model_path = model_path
 
@@ -47,18 +49,52 @@ class OnnxModel(DocLayoutModel):
         providers = []
 
         available_providers = onnxruntime.get_available_providers()
-        for provider in available_providers:
-            # disable dml|cuda|
-            # directml/cuda may encounter problems under special circumstances
-            if re.match(r"cpu", provider, re.IGNORECASE):
-                logger.info(f"Available Provider: {provider}")
-                providers.append(provider)
+        use_coreml = os_name == "Darwin" and any(
+            re.match(r"coreml", p, re.IGNORECASE) for p in available_providers
+        )
+        if use_coreml:
+            # Fix input shapes to [1, 3, 1024, 1024] so CoreML can take over
+            # 96%+ of the graph nodes (658/681) instead of just 3/823.
+            model = self._make_static_model(model, self._FIXED_IMGSZ)
+            providers = [
+                "CoreMLExecutionProvider",
+                "CPUExecutionProvider",
+            ]
+            logger.info(
+                "Using CoreMLExecutionProvider with static input "
+                f"[1, 3, {self._FIXED_IMGSZ}, {self._FIXED_IMGSZ}]"
+            )
+        else:
+            for provider in available_providers:
+                # disable dml|cuda|
+                # directml/cuda may encounter problems under special circumstances
+                if re.match(r"cpu", provider, re.IGNORECASE):
+                    logger.info(f"Available Provider: {provider}")
+                    providers.append(provider)
         self.model = onnxruntime.InferenceSession(
             model.SerializeToString(),
             providers=providers,
         )
         self.lock = threading.Lock()
 
+    @staticmethod
+    def _make_static_model(model, imgsz):
+        """Rewrite dynamic input dims to fixed [1, 3, imgsz, imgsz] and run
+        shape inference so all intermediate shapes become static."""
+        from onnx import shape_inference
+
+        for inp in model.graph.input:
+            if inp.name == "images":
+                dim = inp.type.tensor_type.shape.dim
+                for i, val in enumerate([1, 3, imgsz, imgsz]):
+                    dim[i].ClearField("dim_param")
+                    dim[i].dim_value = val
+        try:
+            model = shape_inference.infer_shapes(model)
+        except Exception:
+            pass  # best-effort; CoreML still benefits from fixed input
+        return model
+
     @staticmethod
     def from_pretrained():
         pth = get_doclayout_onnx_model_path()
@@ -97,9 +133,9 @@ class OnnxModel(DocLayoutModel):
             interpolation=cv2.INTER_LINEAR,
         )
 
-        # Calculate padding size and align to stride multiple
-        pad_w = (new_w - resized_w) % self.stride
-        pad_h = (new_h - resized_h) % self.stride
+        # Pad to full target size (enables static-shape CoreML acceleration)
+        pad_w = new_w - resized_w
+        pad_h = new_h - resized_h
         top, bottom = pad_h // 2, pad_h - pad_h // 2
         left, right = pad_w // 2, pad_w - pad_w // 2
 

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/format/pdf/babelpdf/cidfont.py

@@ -43,8 +43,12 @@ def get_descendant_fonts(doc, xref):
 def get_glyph_bbox(face, g):
     try:
         face.load_glyph(g, freetype.FT_LOAD_NO_SCALE)
-        cbox = face.glyph.outline.get_bbox()
-        return cbox.xMin, cbox.yMin, cbox.xMax, cbox.yMax
+        outline = face.glyph.outline
+        if outline.contours:
+            cbox = outline.get_bbox()
+            return cbox.xMin, cbox.yMin, cbox.xMax, cbox.yMax
+        else:
+            return 0, 0, 0, 0
     except Exception:
         return 0, 0, 0, 0
 

babeldoc-0.5.24/babeldoc/format/pdf/babelpdf/type3.py

@@ -0,0 +1,41 @@
+import io
+import re
+
+import pymupdf
+
+
+def merge_bbox(bbox_list, factor=1):
+    if bbox_list:
+        base = bbox_list[0]
+        for bbox in bbox_list[1:]:
+            base.include_rect(bbox)
+        x0, y0, x1, y1 = [v / factor for v in tuple(base)]
+        return x0, -y1, x1, -y0
+
+
+def get_type3_bbox(doc, obj):
+    bbox_list = [(0, 0, 0, 0)] * 256
+    first = int(doc.xref_get_key(obj, "FirstChar")[1])
+    last = int(doc.xref_get_key(obj, "LastChar")[1])
+    factor_text = doc.xref_get_key(obj, "FontMatrix")[1]
+    factor = 1
+    if factor_m := re.search(r"(\d+)?\.\d+", factor_text):
+        factor = float(factor_m.group(0))
+    page = doc.new_page(width=10, height=10)
+    doc.xref_set_key(page.xref, "Resources", "<<>>")
+    doc.xref_set_key(page.xref, "Resources/Font", f"<</T0 {obj} 0 R>>")
+    text = doc.get_new_xref()
+    doc.update_object(text, "<<>>")
+    for x in range(first, last + 1):
+        doc.update_stream(text, b"1 0 0 1 0 10 cm BT /T0 1 Tf <%02X> Tj ET" % x)
+        doc.xref_set_key(page.xref, "Contents", f"{text} 0 R")
+        char_data = page.get_svg_image(text_as_path=True)
+        char_doc = pymupdf.Document(stream=io.BytesIO(char_data.encode("U8")))
+        char_bbox = []
+        for element in char_doc:
+            for item in element.get_drawings():
+                char_bbox.append(item["rect"])
+        if char_bbox_merged := merge_bbox(char_bbox, factor):
+            bbox_list[x] = char_bbox_merged
+    doc.delete_page(-1)
+    return bbox_list

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/format/pdf/document_il/backend/pdf_creater.py

@@ -485,10 +485,13 @@ def reproduce_cmap(doc):
     assert doc
     font_set = set()
     for page in doc:
-        font_list = page.get_fonts()
-        for font in font_list:
-            if font[1] == "ttf" and font[3] in FONT_NAMES and ".ttf" in font[4]:
-                font_set.add(font)
+        try:
+            font_list = page.get_fonts()
+            for font in font_list:
+                if font[1] == "ttf" and font[3] in FONT_NAMES and ".ttf" in font[4]:
+                    font_set.add(font)
+        except Exception as e:
+            logger.error(f"Error in getting page fonts: {e}")
     for font in font_set:
         reproduce_one_font(doc, font[0])
     return doc
@@ -1399,7 +1402,7 @@ class PDFCreater:
                 auto_extracted_glossary_path = self.translation_config.get_output_file_path(
                     f"{basename}{debug_suffix}.{translation_config.lang_out}.glossary.csv"
                 )
-                with auto_extracted_glossary_path.open("w", encoding="utf-8") as f:
+                with auto_extracted_glossary_path.open("w", encoding="utf-8-sig") as f:
                     logger.info(
                         f"save auto extracted glossary to {auto_extracted_glossary_path}"
                     )

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/format/pdf/document_il/frontend/il_creater.py

@@ -15,8 +15,10 @@ import tiktoken
 import babeldoc.pdfminer.pdfinterp
 from babeldoc.format.pdf.babelpdf.base14 import get_base14_bbox
 from babeldoc.format.pdf.babelpdf.cidfont import get_cidfont_bbox
+from babeldoc.format.pdf.babelpdf.cidfont import get_glyph_bbox
 from babeldoc.format.pdf.babelpdf.encoding import WinAnsiEncoding
 from babeldoc.format.pdf.babelpdf.encoding import get_type1_encoding
+from babeldoc.format.pdf.babelpdf.type3 import get_type3_bbox
 from babeldoc.format.pdf.babelpdf.utils import guarded_bbox
 from babeldoc.format.pdf.document_il import il_version_1
 from babeldoc.format.pdf.document_il.utils import zstd_helper
@@ -111,15 +113,9 @@ def indirect(obj):
         return int(obj[1].split(" ")[0])
 
 
-def get_glyph_cbox(face, g):
-    face.load_glyph(g, freetype.FT_LOAD_NO_SCALE)
-    cbox = face.glyph.outline.get_bbox()
-    return cbox.xMin, cbox.yMin, cbox.xMax, cbox.yMax
-
-
 def get_char_cbox(face, idx):
     g = face.get_char_index(idx)
-    return get_glyph_cbox(face, g)
+    return get_glyph_bbox(face, g)
 
 
 def get_name_cbox(face, name):
@@ -127,7 +123,7 @@ def get_name_cbox(face, name):
         if isinstance(name, str):
             name = name.encode("utf-8")
         g = face.get_name_index(name)
-        return get_glyph_cbox(face, g)
+        return get_glyph_bbox(face, g)
     return (0, 0, 0, 0)
 
 
@@ -798,6 +794,8 @@ class ILCreater:
                 bbox_list = get_base14_bbox(obj_val[1:])
         if cid_bbox := get_cidfont_bbox(self.mupdf, xobj_id):
             bbox_list = cid_bbox
+        if self.mupdf.xref_get_key(xobj_id, "Subtype")[1] == "/Type3":
+            bbox_list = get_type3_bbox(self.mupdf, xobj_id)
         return bbox_list, cmap
 
     def create_graphic_state(

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/format/pdf/document_il/midend/automatic_term_extractor.py

@@ -413,7 +413,7 @@ class AutomaticTermExtractor:
                 "auto_extractor_glossary.csv"
             )
             logger.debug(f"save auto extracted glossary to {path}")
-            with Path(path).open("w", encoding="utf-8") as f:
+            with Path(path).open("w", encoding="utf-8-sig") as f:
                 auto_extracted_glossary = self.shared_context.auto_extracted_glossary
                 if auto_extracted_glossary:
                     f.write(auto_extracted_glossary.to_csv())

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/format/pdf/document_il/midend/il_translator_llm_only.py

@@ -761,7 +761,12 @@ class ILTranslatorLLMOnly:
                     input_token_count = self.calc_token_count(trimed_input)
                     output_token_count = self.calc_token_count(output_unicode)
 
-                    if trimed_input == output_unicode and input_token_count > 10:
+                    same_as_input = trimed_input == output_unicode
+                    if (
+                        same_as_input
+                        and input_token_count > 10
+                        and not self.translation_config.disable_same_text_fallback
+                    ):
                         llm_translate_tracker.set_error_message(
                             "Translation result is the same as input, fallback."
                         )
@@ -781,16 +786,19 @@ class ILTranslatorLLMOnly:
                         llm_translate_tracker.set_placeholder_full_match()
                         continue
 
-                    edit_distance = Levenshtein.distance(input_unicode, output_unicode)
-                    if edit_distance < 5 and input_token_count > 20:
-                        llm_translate_tracker.set_error_message(
-                            f"Translation result edit distance is too small. distance: {edit_distance}, input: {input_unicode}, output: {output_unicode}"
-                        )
-                        logger.warning(
-                            f"Translation result edit distance is too small. distance: {edit_distance}, input: {input_unicode}, output: {output_unicode}"
+                    if not self.translation_config.disable_same_text_fallback:
+                        edit_distance = Levenshtein.distance(
+                            input_unicode, output_unicode
                         )
-                        llm_translate_tracker.set_placeholder_full_match()
-                        continue
+                        if edit_distance < 5 and input_token_count > 20:
+                            llm_translate_tracker.set_error_message(
+                                f"Translation result edit distance is too small. distance: {edit_distance}, input: {input_unicode}, output: {output_unicode}"
+                            )
+                            logger.warning(
+                                f"Translation result edit distance is too small. distance: {edit_distance}, input: {input_unicode}, output: {output_unicode}"
+                            )
+                            llm_translate_tracker.set_placeholder_full_match()
+                            continue
                     # Apply the translation to the paragraph
                     self.il_translator.post_translate_paragraph(
                         inputs[id_][2],

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/format/pdf/result_merger.py

@@ -135,7 +135,7 @@ class ResultMerger:
             auto_extracted_glossary_path = self.config.get_output_file_path(
                 f"{basename}{debug_suffix}.{self.config.lang_out}.glossary.csv"
             )
-            with auto_extracted_glossary_path.open("w", encoding="utf-8") as f:
+            with auto_extracted_glossary_path.open("w", encoding="utf-8-sig") as f:
                 logger.info(
                     f"save auto extracted glossary to {auto_extracted_glossary_path}"
                 )

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/format/pdf/translation_config.py

@@ -199,6 +199,7 @@ class TranslationConfig:
         term_extraction_translator: BaseTranslator | None = None,
         metadata_extra_data: str | None = None,
         term_pool_max_workers: int | None = None,
+        disable_same_text_fallback: bool = False,
     ):
         self.translator = translator
         self.term_extraction_translator = term_extraction_translator or translator
@@ -354,6 +355,7 @@ class TranslationConfig:
             "completion_tokens": 0,
             "cache_hit_prompt_tokens": 0,
         }
+        self.disable_same_text_fallback = disable_same_text_fallback
 
         if self.ocr_workaround:
             self.remove_non_formula_lines = False

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/main.py

@@ -26,7 +26,7 @@ from babeldoc.translator.translator import OpenAITranslator
 from babeldoc.translator.translator import set_translate_rate_limiter
 
 logger = logging.getLogger(__name__)
-__version__ = "0.5.22"
+__version__ = "0.5.24"
 
 
 def create_parser():
@@ -269,6 +269,12 @@ def create_parser():
         default=False,
         help="Add formula placeholder hint for translation. (Currently not recommended, it may affect translation quality, default: False)",
     )
+    translation_group.add_argument(
+        "--disable-same-text-fallback",
+        action="store_true",
+        default=False,
+        help="Disable fallback translation when LLM output matches input text. (default: False)",
+    )
     translation_group.add_argument(
         "--glossary-files",
         type=str,
@@ -703,6 +709,7 @@ async def main():
             custom_system_prompt=args.custom_system_prompt,
             working_dir=working_dir,
             add_formula_placehold_hint=args.add_formula_placehold_hint,
+            disable_same_text_fallback=args.disable_same_text_fallback,
             glossaries=loaded_glossaries,
             pool_max_workers=args.pool_max_workers,
             auto_extract_glossary=args.auto_extract_glossary,

{babeldoc-0.5.22 → babeldoc-0.5.24}/babeldoc/translator/translator.py

@@ -344,9 +344,15 @@ class OpenAITranslator(BaseTranslator):
                 self.prompt_token_count.inc(response.usage.prompt_tokens)
             if response.usage and response.usage.completion_tokens:
                 self.completion_token_count.inc(response.usage.completion_tokens)
-            if response.usage and (
-                hit_count := getattr(response.usage, "prompt_cache_hit_tokens", 0)
+            # Support both response.usage.prompt_cache_hit_tokens and response.prompt_tokens_details.cached_tokens
+            hit_count = 0
+            if response.usage and hasattr(response.usage, "prompt_cache_hit_tokens"):
+                hit_count = getattr(response.usage, "prompt_cache_hit_tokens", 0)
+            if hasattr(response, "prompt_tokens_details") and getattr(
+                response.prompt_tokens_details, "cached_tokens", 0
             ):
+                hit_count += getattr(response.prompt_tokens_details, "cached_tokens", 0)
+            if hit_count:
                 self.cache_hit_prompt_token_count.inc(hit_count)
         except Exception as e:
             logger.exception("Error updating token count")

{babeldoc-0.5.22 → babeldoc-0.5.24}/docs/CONTRIBUTING.md

@@ -46,6 +46,12 @@ Please pay special attention to:
 > 
 > 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.
 >
+> In addition, we do not accept PRs involving the following changes:
+> 1. PRs that modify prompts.
+> 2. Adding GUI or other features directly targeting end users to this project. (Exceptions granted by maintainers in issues are excluded.)
+> 3. PRs that do not comply with this specification.
+> 4. Other PRs that maintainers deem inappropriate.
+>
 > **This project cannot accept all PRs. We recommend that you discuss with the maintainers via [Issue](https://github.com/funstory-ai/BabelDOC/issues) before submitting a PR.**
 
 [//]: # (> We welcome pull requests and will review your contributions.)

{babeldoc-0.5.22 → babeldoc-0.5.24}/mkdocs.yml

@@ -30,6 +30,7 @@ site_description: >-
 repo_name: funstory-ai/BabelDOC
 repo_url: https://github.com/funstory-ai/BabelDOC
 edit_uri: edit/main/docs/
+docs_dir: docs
 
 # Copyright
 copyright: Copyright © 2025 funstory.ai

{babeldoc-0.5.22 → babeldoc-0.5.24}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "BabelDOC"
-version = "0.5.22"
+version = "0.5.24"
 description = "Yet Another Document Translator"
 license = "AGPL-3.0"
 readme = "README.md"
@@ -42,7 +42,7 @@ dependencies = [
     "scikit-image>=0.25.2",
     "freetype-py>=2.5.1",
     "tiktoken>=0.9.0",
-    "python-levenshtein>=0.27.1",
+    "Levenshtein>=0.27.1",
     "opencv-python-headless>=4.10.0.84",
     "rapidocr-onnxruntime>=1.4.4",
     "pyzstd>=0.17.0",
@@ -162,7 +162,7 @@ pythonpath = [".", "src"]
 testpaths = ["tests"]
 
 [bumpver]
-current_version = "0.5.22"
+current_version = "0.5.24"
 version_pattern = "MAJOR.MINOR.PATCH[.PYTAGNUM]"
 
 [bumpver.file_patterns]

babeldoc-0.5.22/.cursorignore

@@ -1,2 +0,0 @@
-# Project notes and templates
-xnotes/

babeldoc-0.5.22/babeldoc/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.5.22"