markitdown-paddleocr 0.2.2__tar.gz → 0.2.3__tar.gz

{markitdown_paddleocr-0.2.2 → markitdown_paddleocr-0.2.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: markitdown-paddleocr
-Version: 0.2.2
+Version: 0.2.3
 Summary: Intelligent PDF/Image to Markdown converter using PaddleOCR cloud API
 Project-URL: Documentation, https://github.com/microsoft/markitdown#readme
 Project-URL: Issues, https://github.com/microsoft/markitdown/issues
@@ -51,7 +51,7 @@ pip install markitdown-paddleocr
 export BAIDU_PADDLE_TOKEN="your-paddle-token"
 
 # 可选
-export PADDLE_OCR_MODEL="PaddleOCR-VL-1.5"   # 模型名称
+export PADDLE_OCR_MODEL="PaddleOCR-VL-1.6"   # 模型名称
 ```
 
 ### 配置优先级
@@ -129,7 +129,7 @@ print(markdown)
 | 参数 | 类型 | 默认值 | 说明 |
 |------|------|--------|------|
 | `token` | str | 环境变量 `BAIDU_PADDLE_TOKEN` | PaddleOCR Token |
-| `model` | str | `PaddleOCR-VL-1.5` | OCR 模型名称 |
+| `model` | str | `PaddleOCR-VL-1.6` | OCR 模型名称 |
 | `poll_interval` | float | 2.0 | 轮询间隔（秒） |
 | `poll_timeout` | float | 300.0 | 轮询超时（秒） |
 | `force_ai` | bool | False | 强制所有页面使用 OCR |
@@ -142,7 +142,7 @@ print(markdown)
 | 变量 | 说明 | 示例 |
 |------|------|------|
 | `BAIDU_PADDLE_TOKEN` | Token（必需） | `7963b85a...` |
-| `PADDLE_OCR_MODEL` | 模型名称 | `PaddleOCR-VL-1.5` |
+| `PADDLE_OCR_MODEL` | 模型名称 | `PaddleOCR-VL-1.6` |
 
 ## 工作原理
 

{markitdown_paddleocr-0.2.2 → markitdown_paddleocr-0.2.3}/README.md

@@ -25,7 +25,7 @@ pip install markitdown-paddleocr
 export BAIDU_PADDLE_TOKEN="your-paddle-token"
 
 # 可选
-export PADDLE_OCR_MODEL="PaddleOCR-VL-1.5"   # 模型名称
+export PADDLE_OCR_MODEL="PaddleOCR-VL-1.6"   # 模型名称
 ```
 
 ### 配置优先级
@@ -103,7 +103,7 @@ print(markdown)
 | 参数 | 类型 | 默认值 | 说明 |
 |------|------|--------|------|
 | `token` | str | 环境变量 `BAIDU_PADDLE_TOKEN` | PaddleOCR Token |
-| `model` | str | `PaddleOCR-VL-1.5` | OCR 模型名称 |
+| `model` | str | `PaddleOCR-VL-1.6` | OCR 模型名称 |
 | `poll_interval` | float | 2.0 | 轮询间隔（秒） |
 | `poll_timeout` | float | 300.0 | 轮询超时（秒） |
 | `force_ai` | bool | False | 强制所有页面使用 OCR |
@@ -116,7 +116,7 @@ print(markdown)
 | 变量 | 说明 | 示例 |
 |------|------|------|
 | `BAIDU_PADDLE_TOKEN` | Token（必需） | `7963b85a...` |
-| `PADDLE_OCR_MODEL` | 模型名称 | `PaddleOCR-VL-1.5` |
+| `PADDLE_OCR_MODEL` | 模型名称 | `PaddleOCR-VL-1.6` |
 
 ## 工作原理
 

markitdown_paddleocr-0.2.3/src/markitdown_paddleocr/__about__.py

@@ -0,0 +1 @@
+__version__ = "0.2.3"

{markitdown_paddleocr-0.2.2 → markitdown_paddleocr-0.2.3}/src/markitdown_paddleocr/_config.py

@@ -12,6 +12,7 @@ class ScanDetectionMode(str, Enum):
     - FIRST_PAGE_HINT: 首页是扫描件则全文档使用OCR
     - SAMPLING: 抽样前N页，多数是扫描件则全部OCR
     """
+
     PAGE_BY_PAGE = "page_by_page"
     FIRST_PAGE_HINT = "first_page_hint"
     SAMPLING = "sampling"
@@ -31,7 +32,7 @@ class PaddleOcrConfig:
     token: str = ""  # Reads from BAIDU_PADDLE_TOKEN by default
 
     # OCR model
-    model: str = "PaddleOCR-VL-1.5"
+    model: str = "PaddleOCR-VL-1.6"
 
     # API endpoint
     job_url: str = "https://paddleocr.aistudio-app.com/api/v2/ocr/jobs"
@@ -58,7 +59,7 @@ class PaddleOcrConfig:
         """Create config from environment variables with optional overrides."""
         defaults = {
             "token": os.environ.get("BAIDU_PADDLE_TOKEN", ""),
-            "model": os.environ.get("PADDLE_OCR_MODEL", "PaddleOCR-VL-1.5"),
+            "model": os.environ.get("PADDLE_OCR_MODEL", "PaddleOCR-VL-1.6"),
         }
         defaults.update(overrides)
         return cls(**defaults)

{markitdown_paddleocr-0.2.2 → markitdown_paddleocr-0.2.3}/src/markitdown_paddleocr/_converter.py

@@ -51,7 +51,7 @@ class PaddleOcrConverter(DocumentConverter):
     def __init__(
         self,
         token: Optional[str] = None,
-        model: str = "PaddleOCR-VL-1.5",
+        model: str = "PaddleOCR-VL-1.6",
         poll_interval: float = 2.0,
         poll_timeout: float = 300.0,
         force_ai: bool = False,
@@ -67,7 +67,7 @@ class PaddleOcrConverter(DocumentConverter):
 
         Args:
             token: Baidu PaddleOCR token (reads from BAIDU_PADDLE_TOKEN env var if not provided)
-            model: OCR model name (default: PaddleOCR-VL-1.5)
+            model: OCR model name (default: PaddleOCR-VL-1.6)
             poll_interval: Seconds between status polls (default: 2.0)
             poll_timeout: Max seconds to wait for job completion (default: 300.0)
             force_ai: Force all pages to use OCR (default: False)
@@ -82,7 +82,7 @@ class PaddleOcrConverter(DocumentConverter):
         # Build config from explicit params or provided config
         if config:
             self.token = token or config.token
-            self.model = model if model != "PaddleOCR-VL-1.5" else config.model
+            self.model = model if model != "PaddleOCR-VL-1.6" else config.model
             self.poll_interval = (
                 poll_interval if poll_interval != 2.0 else config.poll_interval
             )
@@ -126,7 +126,9 @@ class PaddleOcrConverter(DocumentConverter):
                 if scan_detection_mode is not None
                 else ScanDetectionMode.SAMPLING
             )
-            self.scan_sample_pages = scan_sample_pages if scan_sample_pages is not None else 3
+            self.scan_sample_pages = (
+                scan_sample_pages if scan_sample_pages is not None else 3
+            )
             self.scan_text_threshold = (
                 scan_text_threshold if scan_text_threshold is not None else 50
             )
@@ -357,7 +359,9 @@ class PaddleOcrConverter(DocumentConverter):
         Returns:
             Markdown text from all pages.
         """
-        logger.info("PaddleOcrConverter: 批量上传PDF到OCR API, 大小=%d bytes", len(pdf_bytes))
+        logger.info(
+            "PaddleOcrConverter: 批量上传PDF到OCR API, 大小=%d bytes", len(pdf_bytes)
+        )
         markdown = self._get_client().ocr(
             file_bytes=pdf_bytes,
             filename="document.pdf",

{markitdown_paddleocr-0.2.2 → markitdown_paddleocr-0.2.3}/src/markitdown_paddleocr/_dual_converter.py

@@ -1,10 +1,14 @@
 """DualOcrConverter - glmocr (primary) → paddleocr (fallback) automatic degradation."""
 
 import logging
-from typing import Optional
+from typing import Any, BinaryIO, Optional
 
-from markitdown import MarkItDown, DocumentConverter, DocumentConverterResult, StreamInfo
-from typing import BinaryIO, Any
+from markitdown import (
+    DocumentConverter,
+    DocumentConverterResult,
+    MarkItDown,
+    StreamInfo,
+)
 
 logger = logging.getLogger(__name__)
 
@@ -28,7 +32,7 @@ class DualOcrConverter(DocumentConverter):
         glmocr_force_ai: bool = False,
         # paddleocr kwargs
         paddleocr_token: Optional[str] = None,
-        paddleocr_model: str = "PaddleOCR-VL-1.5",
+        paddleocr_model: str = "PaddleOCR-VL-1.6",
         paddleocr_poll_interval: float = 2.0,
         paddleocr_poll_timeout: float = 300.0,
         paddleocr_force_ai: bool = False,
@@ -61,6 +65,7 @@ class DualOcrConverter(DocumentConverter):
         """Lazily init both converters."""
         try:
             from markitdown_glmocr import GlmOcrConverter
+
             # Filter out None values
             kwargs = {k: v for k, v in self.glmocr_kwargs.items() if v is not None}
             self._primary = GlmOcrConverter(**kwargs)
@@ -71,6 +76,7 @@ class DualOcrConverter(DocumentConverter):
 
         try:
             from markitdown_paddleocr import PaddleOcrConverter
+
             kwargs = {k: v for k, v in self.paddleocr_kwargs.items() if v is not None}
             self._fallback = PaddleOcrConverter(**kwargs)
             logger.info("paddleocr converter initialized (fallback)")
@@ -155,6 +161,7 @@ class DualOcrConverter(DocumentConverter):
 def io_bytes(data: bytes):
     """Create a seekable BytesIO from bytes."""
     import io
+
     buf = io.BytesIO(data)
     buf.seek(0)
     return buf

{markitdown_paddleocr-0.2.2 → markitdown_paddleocr-0.2.3}/src/markitdown_paddleocr/_plugin.py

@@ -28,7 +28,7 @@ def register_converters(markitdown: MarkItDown, **kwargs: Any) -> None:
     try:
         converter = PaddleOcrConverter(
             token=kwargs.get("token"),
-            model=kwargs.get("model", "PaddleOCR-VL-1.5"),
+            model=kwargs.get("model", "PaddleOCR-VL-1.6"),
             poll_interval=kwargs.get("poll_interval", 2.0),
             poll_timeout=kwargs.get("poll_timeout", 300.0),
             force_ai=kwargs.get("force_ai", False),

markitdown_paddleocr-0.2.2/src/markitdown_paddleocr/__about__.py

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