natural-pdf 0.2.19__tar.gz → 0.2.20__tar.gz

{natural_pdf-0.2.19 → natural_pdf-0.2.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: natural-pdf
-Version: 0.2.19
+Version: 0.2.20
 Summary: A more intuitive interface for working with PDFs
 Author-email: Jonathan Soma <jonathan.soma@gmail.com>
 License-Expression: MIT

{natural_pdf-0.2.19 → natural_pdf-0.2.20}/natural_pdf/analyzers/guides.py

@@ -943,7 +943,7 @@ class GuidesList(UserList):
 
     def from_headers(
         self,
-        headers: Union["ElementCollection", List["Element"]],
+        headers: Union["ElementCollection", List["Element"], List[str]],
         obj: Optional[Union["Page", "Region"]] = None,
         method: Literal["min_crossings", "seam_carving"] = "min_crossings",
         min_width: Optional[float] = None,
@@ -960,7 +960,10 @@ class GuidesList(UserList):
         between headers that minimize text crossings, regardless of text alignment.
 
         Args:
-            headers: Column header elements (ElementCollection or list of Elements)
+            headers: Column header elements. Can be:
+                - ElementCollection: collection of header elements
+                - List[Element]: list of header elements
+                - List[str]: list of header text to search for
             obj: Page/Region to analyze (uses parent's context if None)
             method: Detection method:
                 - 'min_crossings': Fast vector-based minimum intersection count
@@ -980,6 +983,9 @@ class GuidesList(UserList):
             headers = page.find_all('text[size=16]')
             guides.vertical.from_headers(headers)
 
+            # From header text strings
+            guides.vertical.from_headers(["Statute", "Description", "Level", "Repeat"])
+
             # With width constraints
             guides.vertical.from_headers(headers, min_width=50, max_width=200)
 
@@ -997,6 +1003,24 @@ class GuidesList(UserList):
         # Convert headers to list if ElementCollection
         if hasattr(headers, "elements"):
             header_elements = list(headers.elements)
+        # Check if headers is a list of strings
+        elif isinstance(headers, list) and headers and isinstance(headers[0], str):
+            # Find elements for each header text with exact matching
+            header_elements = []
+            for header_text in headers:
+                # Find all text elements and filter for exact match
+                all_text = target_obj.find_all("text")
+                exact_matches = [elem for elem in all_text if elem.extract_text() == header_text]
+
+                if exact_matches:
+                    # Use the first exact match
+                    header_elements.append(exact_matches[0])
+                else:
+                    logger.warning(f"Could not find header text: {header_text}")
+
+            if not header_elements:
+                logger.warning("No header elements found from provided text strings")
+                return self._parent
         else:
             header_elements = list(headers)
 

{natural_pdf-0.2.19 → natural_pdf-0.2.20}/natural_pdf.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: natural-pdf
-Version: 0.2.19
+Version: 0.2.20
 Summary: A more intuitive interface for working with PDFs
 Author-email: Jonathan Soma <jonathan.soma@gmail.com>
 License-Expression: MIT

{natural_pdf-0.2.19 → natural_pdf-0.2.20}/natural_pdf.egg-info/SOURCES.txt

@@ -319,6 +319,7 @@ tests/test_guides_extract_table_collections.py
 tests/test_guides_extract_table_exclusions.py
 tests/test_guides_extract_table_real.py
 tests/test_guides_from_headers.py
+tests/test_guides_from_headers_strings.py
 tests/test_guides_from_stripes.py
 tests/test_guides_integration.py
 tests/test_guides_marker_sorting.py

natural_pdf-0.2.20/tests/test_guides_from_headers_strings.py

@@ -0,0 +1,76 @@
+"""Test from_headers() with string headers."""
+
+import pytest
+
+from natural_pdf import PDF
+from natural_pdf.analyzers.guides import Guides
+
+
+def test_from_headers_with_strings(practice_pdf):
+    """Test from_headers with list of strings."""
+    page = practice_pdf[0]
+
+    # Get some text to use as test headers
+    all_text = page.find_all("text")[:5]
+    if len(all_text) >= 3:
+        # Extract the actual text content
+        header_texts = [elem.extract_text() for elem in all_text[:3]]
+
+        # Create guides using string headers
+        guides = Guides(page)
+        guides.vertical.from_headers(header_texts)
+
+        # Should work similar to using elements directly
+        guides_from_elements = Guides(page)
+        guides_from_elements.vertical.from_headers(all_text[:3])
+
+        # Both approaches should produce similar results
+        # (may not be identical due to search differences)
+        assert len(guides.vertical) >= 0
+        assert len(guides_from_elements.vertical) >= 0
+
+
+def test_from_headers_with_missing_strings(practice_pdf):
+    """Test from_headers when some strings are not found."""
+    page = practice_pdf[0]
+
+    # Use mix of real and fake headers
+    all_text = page.find_all("text")[:2]
+    if len(all_text) >= 2:
+        real_texts = [elem.extract_text() for elem in all_text]
+        fake_texts = ["NONEXISTENT_HEADER_1", "NONEXISTENT_HEADER_2"]
+        mixed_texts = real_texts + fake_texts
+
+        # Should still work with found headers
+        guides = Guides(page)
+        result = guides.vertical.from_headers(mixed_texts)
+
+        # Should return the parent for chaining
+        assert result is guides
+
+
+def test_from_headers_all_strings_missing(practice_pdf):
+    """Test from_headers when no strings are found."""
+    page = practice_pdf[0]
+
+    # All fake headers
+    fake_headers = ["FAKE_HEADER_1", "FAKE_HEADER_2", "FAKE_HEADER_3"]
+
+    guides = Guides(page)
+    result = guides.vertical.from_headers(fake_headers)
+
+    # Should return parent but no guides created
+    assert result is guides
+    assert len(guides.vertical) == 0
+
+
+def test_from_headers_empty_string_list(practice_pdf):
+    """Test from_headers with empty string list."""
+    page = practice_pdf[0]
+
+    guides = Guides(page)
+    result = guides.vertical.from_headers([])
+
+    # Should handle gracefully
+    assert result is guides
+    assert len(guides.vertical) == 0