BabelDOC 0.2.12__tar.gz → 0.2.14__tar.gz

{babeldoc-0.2.12 → babeldoc-0.2.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BabelDOC
-Version: 0.2.12
+Version: 0.2.14
 Summary: Yet Another Document Translator
 Project-URL: Homepage, https://github.com/funstory-ai/BabelDOC
 Project-URL: Issues, https://github.com/funstory-ai/BabelDOC/issues

babeldoc-0.2.14/babeldoc/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.2.14"

{babeldoc-0.2.12 → babeldoc-0.2.14}/babeldoc/const.py

@@ -3,7 +3,7 @@ import shutil
 import subprocess
 from pathlib import Path
 
-__version__ = "0.2.12"
+__version__ = "0.2.14"
 
 CACHE_FOLDER = Path.home() / ".cache" / "babeldoc"
 

{babeldoc-0.2.12 → babeldoc-0.2.14}/babeldoc/document_il/backend/pdf_creater.py

@@ -730,7 +730,12 @@ class PDFCreater:
                     draw_op.append(b" Tj ET Q \n")
                 for xobj in page.pdf_xobject:
                     draw_op = xobj_draw_ops[xobj.xobj_id]
-                    pdf.update_stream(xobj.xref_id, draw_op.tobytes())
+                    try:
+                        pdf.update_stream(xobj.xref_id, draw_op.tobytes())
+                    except Exception:
+                        logger.warning(
+                            f"update xref {xobj.xref_id} stream fail, continue"
+                        )
                     # pdf.update_stream(xobj.xref_id, b'')
                 for rect in page.pdf_rectangle:
                     self._debug_render_rectangle(page_op, rect)

{babeldoc-0.2.12 → babeldoc-0.2.14}/babeldoc/high_level.py

@@ -190,10 +190,10 @@ def start_parse_il(
         #         box[y0:y1, x0:x1] = 0
         # layout[page.pageno] = box
         # 新建一个 xref 存放新指令流
-        page.page_xref = doc_zh.get_new_xref()  # hack 插入页面的新 xref
-        doc_zh.update_object(page.page_xref, "<<>>")
-        doc_zh.update_stream(page.page_xref, b"")
-        doc_zh[page.pageno].set_contents(page.page_xref)
+        # page.page_xref = doc_zh.get_new_xref()  # hack 插入页面的新 xref
+        # doc_zh.update_object(page.page_xref, "<<>>")
+        # doc_zh.update_stream(page.page_xref, b"")
+        # doc_zh[page.pageno].set_contents(page.page_xref)
         ops_base = interpreter.process_page(page)
         il_creater.on_page_base_operation(ops_base)
         il_creater.on_page_end()
@@ -353,6 +353,21 @@ class MemoryMonitor:
             time.sleep(self.interval)
 
 
+def fix_null_xref(doc: Document) -> None:
+    """Fix null xref in PDF file by replacing them with empty arrays.
+
+    Args:
+        doc: PyMuPDF Document object to fix
+    """
+    for i in range(1, doc.xref_length()):
+        try:
+            obj = doc.xref_object(i)
+            if obj == "null":
+                doc.update_object(i, "[]")
+        except Exception:
+            doc.update_object(i, "[]")
+
+
 def do_translate(
     pm: ProgressMonitor, translation_config: TranslationConfig
 ) -> TranslateResult:
@@ -507,12 +522,14 @@ def _do_translate_single(
     """Original translation logic for a single document or part"""
     translation_config.progress_monitor = pm
     original_pdf_path = translation_config.input_file
-    doc_input = Document(original_pdf_path)
     if translation_config.debug:
+        doc_input = Document(original_pdf_path)
         logger.debug("debug mode, save decompressed input pdf")
         output_path = translation_config.get_working_file_path(
             "input.decompressed.pdf",
         )
+        # Fix null xref in PDF file
+        fix_null_xref(doc_input)
         doc_input.save(output_path, expand=True, pretty=True)
 
     # Continue with original processing
@@ -521,12 +538,7 @@ def _do_translate_single(
     resfont = "china-ss"
 
     # Fix null xref in PDF file
-    for i in range(1, doc_pdf2zh.xref_length()):
-        try:
-            if doc_pdf2zh.xref_object(i) == "null":
-                doc_pdf2zh.update_object(i, "[]")
-        except Exception:
-            logger.warning(f"try fix xref {i} fail, continue")
+    fix_null_xref(doc_pdf2zh)
 
     for page in doc_pdf2zh:
         page.insert_font(resfont, None)
@@ -534,7 +546,7 @@ def _do_translate_single(
     resfont = None
     doc_pdf2zh.save(temp_pdf_path)
     il_creater = ILCreater(translation_config)
-    il_creater.mupdf = doc_input
+    il_creater.mupdf = doc_pdf2zh
     xml_converter = XMLConverter()
     logger.debug(f"start parse il from {temp_pdf_path}")
     with Path(temp_pdf_path).open("rb") as f:
@@ -569,7 +581,7 @@ def _do_translate_single(
 
     # Generate layouts for all pages
     logger.debug("start generating layouts")
-    docs = LayoutParser(translation_config).process(docs, doc_input)
+    docs = LayoutParser(translation_config).process(docs, doc_pdf2zh)
     logger.debug("finish generating layouts")
     if translation_config.debug:
         xml_converter.write_json(
@@ -625,7 +637,7 @@ def _do_translate_single(
 
     if translation_config.watermark_output_mode == WatermarkOutputMode.Both:
         mono_watermark_first_page_doc_bytes, dual_watermark_first_page_doc_bytes = (
-            generate_first_page_with_watermark(doc_input, translation_config, docs)
+            generate_first_page_with_watermark(doc_pdf2zh, translation_config, docs)
         )
 
     Typesetting(translation_config).typsetting_document(docs)

{babeldoc-0.2.12 → babeldoc-0.2.14}/babeldoc/main.py

@@ -22,7 +22,7 @@ from babeldoc.translation_config import TranslationConfig
 from babeldoc.translation_config import WatermarkOutputMode
 
 logger = logging.getLogger(__name__)
-__version__ = "0.2.12"
+__version__ = "0.2.14"
 
 
 def create_parser():

{babeldoc-0.2.12 → babeldoc-0.2.14}/babeldoc/pdfinterp.py

@@ -337,14 +337,14 @@ class PDFPageInterpreterEx(PDFPageInterpreter):
         ops_base = self.render_contents(page.resources, page.contents, ctm=ctm)
         self.device.fontid = self.fontid
         self.device.fontmap = self.fontmap
-        ops_new = self.device.end_page(page)
+        _ops_new = self.device.end_page(page)
         # 上面渲染的时候会根据 cropbox 减掉页面偏移得到真实坐标，这里输出的时候需要用 cm 把页面偏移加回来
-        self.obj_patch[page.page_xref] = (
-            # f"q {ops_base}Q 1 0 0 1 {x0} {y0} cm {ops_new}"  # ops_base 里可能有图，需要让 ops_new 里的文字覆盖在上面，使用 q/Q 重置位置矩阵
-            ""
-        )
-        for obj in page.contents:
-            self.obj_patch[obj.objid] = ""
+        # self.obj_patch[page.page_xref] = (
+        #     # f"q {ops_base}Q 1 0 0 1 {x0} {y0} cm {ops_new}"  # ops_base 里可能有图，需要让 ops_new 里的文字覆盖在上面，使用 q/Q 重置位置矩阵
+        #     ""
+        # )
+        # for obj in page.contents:
+        #     self.obj_patch[obj.objid] = ""
         return ops_base
 
     def render_contents(

{babeldoc-0.2.12 → babeldoc-0.2.14}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "BabelDOC"
-version = "0.2.12"
+version = "0.2.14"
 description = "Yet Another Document Translator"
 license = "AGPL-3.0"
 readme = "README.md"
@@ -147,7 +147,7 @@ pythonpath = [".", "src"]
 testpaths = ["tests"]
 
 [bumpver]
-current_version = "0.2.12"
+current_version = "0.2.14"
 version_pattern = "MAJOR.MINOR.PATCH[.PYTAGNUM]"
 
 [bumpver.file_patterns]

{babeldoc-0.2.12 → babeldoc-0.2.14}/uv.lock

@@ -48,7 +48,7 @@ wheels = [
 
 [[package]]
 name = "babeldoc"
-version = "0.2.12"
+version = "0.2.14"
 source = { editable = "." }
 dependencies = [
     { name = "bitstring" },
@@ -1399,7 +1399,7 @@ wheels = [
 
 [[package]]
 name = "openai"
-version = "1.66.3"
+version = "1.66.5"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "anyio" },
@@ -1411,9 +1411,9 @@ dependencies = [
     { name = "tqdm" },
     { name = "typing-extensions" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a3/77/5172104ca1df35ed2ed8fb26dbc787f721c39498fc51d666c4db07756a0c/openai-1.66.3.tar.gz", hash = "sha256:8dde3aebe2d081258d4159c4cb27bdc13b5bb3f7ea2201d9bd940b9a89faf0c9", size = 397244 }
+sdist = { url = "https://files.pythonhosted.org/packages/bb/10/b19dc682c806e6735a8387f2003afe2abada9f9e5227318de642c6949524/openai-1.66.5.tar.gz", hash = "sha256:f61b8fac29490ca8fdc6d996aa6926c18dbe5639536f8c40219c40db05511b11", size = 398595 }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/78/5a/e20182f7b6171642d759c548daa0ba20a1d3ac10d2bd0a13fd75704a9ac3/openai-1.66.3-py3-none-any.whl", hash = "sha256:a427c920f727711877ab17c11b95f1230b27767ba7a01e5b66102945141ceca9", size = 567400 },
+    { url = "https://files.pythonhosted.org/packages/c7/3b/1ba418920ecd1eae7cc4d4ac8a01711ee0879b1a57dd81d10551e5b9a2ea/openai-1.66.5-py3-none-any.whl", hash = "sha256:74be528175f8389f67675830c51a15bd51e874425c86d3de6153bf70ed6c2884", size = 571144 },
 ]
 
 [[package]]
@@ -1594,7 +1594,7 @@ wheels = [
 
 [[package]]
 name = "pre-commit"
-version = "4.1.0"
+version = "4.2.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "cfgv" },
@@ -1603,9 +1603,9 @@ dependencies = [
     { name = "pyyaml" },
     { name = "virtualenv" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/2a/13/b62d075317d8686071eb843f0bb1f195eb332f48869d3c31a4c6f1e063ac/pre_commit-4.1.0.tar.gz", hash = "sha256:ae3f018575a588e30dfddfab9a05448bfbd6b73d78709617b5a2b853549716d4", size = 193330 }
+sdist = { url = "https://files.pythonhosted.org/packages/08/39/679ca9b26c7bb2999ff122d50faa301e49af82ca9c066ec061cfbc0c6784/pre_commit-4.2.0.tar.gz", hash = "sha256:601283b9757afd87d40c4c4a9b2b5de9637a8ea02eaff7adc2d0fb4e04841146", size = 193424 }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/43/b3/df14c580d82b9627d173ceea305ba898dca135feb360b6d84019d0803d3b/pre_commit-4.1.0-py2.py3-none-any.whl", hash = "sha256:d29e7cb346295bcc1cc75fc3e92e343495e3ea0196c9ec6ba53f49f10ab6ae7b", size = 220560 },
+    { url = "https://files.pythonhosted.org/packages/88/74/a88bf1b1efeae488a0c0b7bdf71429c313722d1fc0f377537fbe554e6180/pre_commit-4.2.0-py2.py3-none-any.whl", hash = "sha256:a009ca7205f1eb497d10b845e52c838a98b6cdd2102a6c8e4540e94ee75c58bd", size = 220707 },
 ]
 
 [[package]]

babeldoc-0.2.12/babeldoc/__init__.py

@@ -1 +0,0 @@
-__version__ = "0.2.12"