natural-pdf 25.3.16__py3-none-any.whl

natural_pdf/elements/line.py

@@ -0,0 +1,124 @@
+"""
+Line element class for natural-pdf.
+"""
+from typing import Dict, Any, Optional, Tuple, TYPE_CHECKING
+
+from natural_pdf.elements.base import Element
+
+if TYPE_CHECKING:
+    from natural_pdf.core.page import Page
+
+
+class LineElement(Element):
+    """
+    Represents a line element in a PDF.
+    
+    This class is a wrapper around pdfplumber's line objects,
+    providing additional functionality for analysis and extraction.
+    """
+    
+    def __init__(self, obj: Dict[str, Any], page: 'Page'):
+        """
+        Initialize a line element.
+        
+        Args:
+            obj: The underlying pdfplumber object
+            page: The parent Page object
+        """
+        super().__init__(obj, page)
+    
+    @property
+    def type(self) -> str:
+        """Element type."""
+        return 'line'
+        
+    @property
+    def color(self) -> Tuple:
+        """Get the line color (RGB tuple)."""
+        # PDFs often use non-RGB values, so we handle different formats
+        color = self._obj.get('stroking_color', (0, 0, 0))
+        
+        # If it's a single value, treat as grayscale
+        if isinstance(color, (int, float)):
+            return (color, color, color)
+        
+        # If it's a tuple of 3 values, treat as RGB
+        if isinstance(color, tuple) and len(color) == 3:
+            return color
+        
+        # If it's a tuple of 4 values, treat as CMYK and convert to approximate RGB
+        if isinstance(color, tuple) and len(color) == 4:
+            c, m, y, k = color
+            r = 1 - min(1, c + k)
+            g = 1 - min(1, m + k)
+            b = 1 - min(1, y + k)
+            return (r, g, b)
+        
+        # Default to black
+        return (0, 0, 0)
+    
+    @property
+    def width(self) -> float:
+        """Get the line width."""
+        return self._obj.get('linewidth', 0)
+    
+    @property
+    def is_horizontal(self) -> bool:
+        """Check if this is a horizontal line."""
+        return self.height < 1 and self.width > 1
+    
+    @property
+    def is_vertical(self) -> bool:
+        """Check if this is a vertical line."""
+        return self.width < 1 and self.height > 1
+    
+    def text_above(self, distance: float = 5, **kwargs) -> Any:
+        """
+        Get text elements above this line.
+        
+        Args:
+            distance: Maximum distance above the line in points
+            **kwargs: Additional filter parameters
+            
+        Returns:
+            ElementCollection of text elements above this line
+        """
+        from natural_pdf.elements.collections import ElementCollection
+        
+        # TODO: Implement proper filtering of elements above this line
+        return ElementCollection([])  # Placeholder
+    
+    def text_below(self, distance: float = 5, **kwargs) -> Any:
+        """
+        Get text elements below this line.
+        
+        Args:
+            distance: Maximum distance below the line in points
+            **kwargs: Additional filter parameters
+            
+        Returns:
+            ElementCollection of text elements below this line
+        """
+        from natural_pdf.elements.collections import ElementCollection
+        
+        # TODO: Implement proper filtering of elements below this line
+        return ElementCollection([])  # Placeholder
+    
+    def extract_text(self, keep_blank_chars=True, apply_exclusions=True, **kwargs) -> str:
+        """
+        Lines don't have text, so this returns an empty string.
+        
+        Args:
+            keep_blank_chars: Whether to keep blank characters (default: True)
+            apply_exclusions: Whether to apply exclusion regions (default: True)
+            **kwargs: Additional extraction parameters
+            
+        Returns:
+            Empty string
+        """
+        return ""
+    
+    def __repr__(self) -> str:
+        """String representation of the line element."""
+        line_type = "horizontal" if self.is_horizontal else "vertical" if self.is_vertical else "diagonal"
+        return f"<LineElement type={line_type} width={self.width:.1f} bbox={self.bbox}>"

natural_pdf/elements/rect.py

@@ -0,0 +1,122 @@
+"""
+Rectangle element class for natural-pdf.
+"""
+from typing import Dict, Any, Optional, Tuple, TYPE_CHECKING
+
+from natural_pdf.elements.base import Element
+
+if TYPE_CHECKING:
+    from natural_pdf.core.page import Page
+
+
+class RectangleElement(Element):
+    """
+    Represents a rectangle element in a PDF.
+    
+    This class is a wrapper around pdfplumber's rectangle objects,
+    providing additional functionality for analysis and extraction.
+    """
+    
+    def __init__(self, obj: Dict[str, Any], page: 'Page'):
+        """
+        Initialize a rectangle element.
+        
+        Args:
+            obj: The underlying pdfplumber object
+            page: The parent Page object
+        """
+        super().__init__(obj, page)
+    
+    @property
+    def type(self) -> str:
+        """Element type."""
+        return 'rect'
+    
+    @property
+    def fill(self) -> Tuple:
+        """Get the fill color of the rectangle (RGB tuple)."""
+        # PDFs often use non-RGB values, so we handle different formats
+        color = self._obj.get('non_stroking_color', (0, 0, 0))
+        
+        # If it's a single value, treat as grayscale
+        if isinstance(color, (int, float)):
+            return (color, color, color)
+        
+        # If it's a tuple of 3 values, treat as RGB
+        if isinstance(color, tuple) and len(color) == 3:
+            return color
+        
+        # If it's a tuple of 4 values, treat as CMYK and convert to approximate RGB
+        if isinstance(color, tuple) and len(color) == 4:
+            c, m, y, k = color
+            r = 1 - min(1, c + k)
+            g = 1 - min(1, m + k)
+            b = 1 - min(1, y + k)
+            return (r, g, b)
+        
+        # Default to black
+        return (0, 0, 0)
+    
+    @property
+    def stroke(self) -> Tuple:
+        """Get the stroke color of the rectangle (RGB tuple)."""
+        # PDFs often use non-RGB values, so we handle different formats
+        color = self._obj.get('stroking_color', (0, 0, 0))
+        
+        # If it's a single value, treat as grayscale
+        if isinstance(color, (int, float)):
+            return (color, color, color)
+        
+        # If it's a tuple of 3 values, treat as RGB
+        if isinstance(color, tuple) and len(color) == 3:
+            return color
+        
+        # If it's a tuple of 4 values, treat as CMYK and convert to approximate RGB
+        if isinstance(color, tuple) and len(color) == 4:
+            c, m, y, k = color
+            r = 1 - min(1, c + k)
+            g = 1 - min(1, m + k)
+            b = 1 - min(1, y + k)
+            return (r, g, b)
+        
+        # Default to black
+        return (0, 0, 0)
+    
+    @property
+    def stroke_width(self) -> float:
+        """Get the stroke width of the rectangle."""
+        return self._obj.get('linewidth', 0)
+    
+    def text_inside(self, **kwargs) -> Any:
+        """
+        Get text elements inside this rectangle.
+        
+        Args:
+            **kwargs: Additional filter parameters
+            
+        Returns:
+            ElementCollection of text elements inside this rectangle
+        """
+        from natural_pdf.elements.collections import ElementCollection
+        
+        # TODO: Implement proper filtering of elements inside this rectangle
+        return ElementCollection([])  # Placeholder
+    
+    def extract_text(self, **kwargs) -> str:
+        """
+        Extract text from inside this rectangle.
+        
+        Args:
+            **kwargs: Additional extraction parameters
+            
+        Returns:
+            Extracted text as string
+        """
+        # Use the region to extract text
+        from natural_pdf.elements.region import Region
+        region = Region(self.page, self.bbox)
+        return region.extract_text(**kwargs)
+    
+    def __repr__(self) -> str:
+        """String representation of the rectangle element."""
+        return f"<RectangleElement fill={self.fill} stroke={self.stroke} bbox={self.bbox}>"