natural-pdf 0.2.12__py3-none-any.whl → 0.2.15__py3-none-any.whl

temp/debug_exclusions_guides.py

@@ -1,67 +0,0 @@
-"""Debug why exclusions aren't working with guides.extract_table()"""
-from natural_pdf import PDF
-from natural_pdf.analyzers.guides import Guides
-
-pdf = PDF("pdfs/m27.pdf")
-page = pdf.pages[0]
-
-# Check initial text
-print("Initial text:")
-print(page.extract_text()[:200])
-print()
-
-# Add exclusions
-pdf.add_exclusion(lambda page: page.find(text="PREMISE").above())
-pdf.add_exclusion(lambda page: page.find("text:regex(Page \d+ of)").expand())
-
-# Check text after exclusions
-print("Text after exclusions:")
-print(page.extract_text()[:100])
-print()
-
-# Debug exclusion regions
-print("Checking exclusion regions:")
-exclusions = page._get_exclusion_regions(debug=True)
-print(f"\nTotal exclusions: {len(exclusions)}")
-for i, exc in enumerate(exclusions):
-    print(f"  {i}: {exc.bbox}")
-print()
-
-# Create guides
-headers = (
-    page
-    .find(text="NUMBER")
-    .right(include_source=True)
-    .expand(top=3, bottom=3)
-    .find_all('text')
-)
-
-guides = Guides(page)
-guides.vertical.from_content(headers, align='left')
-guides.horizontal.from_stripes()
-
-# Build grid to see what regions are created
-print("\nBuilding grid...")
-grid_result = guides.build_grid(include_outer_boundaries=True)
-table_region = grid_result["regions"]["table"]
-print(f"Table region: {table_region}")
-print(f"Table bbox: {table_region.bbox if table_region else 'None'}")
-
-# Check if table region respects exclusions
-if table_region:
-    print("\nExtracting text from table region directly:")
-    table_text = table_region.extract_text()[:200]
-    print(f"Table text: {table_text}")
-    
-    # Now extract table
-    print("\nExtracting table with apply_exclusions=True:")
-    result = guides.extract_table(include_outer_boundaries=True, apply_exclusions=True, header=False)
-    df = result.to_df()
-    print(df.head())
-    
-    # Check if excluded content is in the table
-    table_str = df.to_string()
-    has_feb = "FEBRUARY 2014" in table_str or "FEBR" in table_str
-    has_alphabetic = "ALPHABETIC LISTING" in table_str
-    print(f"\nContains 'FEBRUARY': {has_feb}")
-    print(f"Contains 'ALPHABETIC LISTING': {has_alphabetic}")

temp/debug_extra_guide.py

@@ -1,41 +0,0 @@
-"""Debug the extra guide issue."""
-from natural_pdf import PDF
-from natural_pdf.analyzers.guides import Guides
-
-pdf = PDF("pdfs/m27.pdf")
-page = pdf.pages[0]
-
-# Get headers
-headers = (
-    page
-    .find("text:contains(NUMBER)")
-    .right(include_source=True)
-    .expand(top=3, bottom=3)
-    .find_all('text')
-)
-
-print("Headers 3-5:")
-for i, h in enumerate(headers[3:5]):
-    print(f"  {i}: '{h.text}' bbox={h.bbox}")
-
-# Create guides with just these two headers
-guides = Guides(page)
-guides.vertical.from_content(headers[3:5], align='left', outer=False)
-
-print(f"\nResulting guides: {guides.vertical}")
-print(f"Expected: [328.32012, 539.63316]")
-
-# Let's also check what happens with each header individually
-print("\nTesting each header individually:")
-for i, h in enumerate(headers[3:5]):
-    g = Guides(page)
-    g.vertical.from_content([h], align='left', outer=False)
-    print(f"  Header {i} guides: {g.vertical}")
-
-# Check if it's related to the ElementCollection
-print("\nTesting with manual list of text:")
-text_list = [h.text for h in headers[3:5]]
-print(f"Text list: {text_list}")
-guides2 = Guides(page)
-guides2.vertical.from_content(text_list, align='left', outer=False)
-print(f"Guides from text list: {guides2.vertical}")

temp/debug_outer_boundaries.py

@@ -1,46 +0,0 @@
-"""Debug outer boundaries issue with exclusions"""
-from natural_pdf import PDF
-from natural_pdf.analyzers.guides import Guides
-
-pdf = PDF("pdfs/m27.pdf")
-page = pdf.pages[0]
-
-# Add exclusions
-pdf.add_exclusion(lambda page: page.find(text="PREMISE").above())
-
-# Create guides
-headers = (
-    page
-    .find(text="NUMBER")
-    .right(include_source=True)
-    .expand(top=3, bottom=3)
-    .find_all('text')
-)
-
-guides = Guides(page)
-guides.vertical.from_content(headers, align='left')
-guides.horizontal.from_stripes()
-
-print("Horizontal guides (sorted):")
-for i, h in enumerate(sorted(guides.horizontal)):
-    print(f"  {i}: {h:.2f}")
-
-print(f"\nFirst content guide: {sorted(guides.horizontal)[0]:.2f}")
-print(f"Page height: {page.height}")
-
-# Test without outer boundaries
-print("\n\nWithout outer boundaries:")
-result1 = guides.extract_table(include_outer_boundaries=False, apply_exclusions=True, header=False)
-df1 = result1.to_df()
-print(f"Shape: {df1.shape}")
-print("First row, first column:", df1.iloc[0, 0] if not df1.empty else "Empty")
-
-# Test with outer boundaries
-print("\n\nWith outer boundaries:")
-result2 = guides.extract_table(include_outer_boundaries=True, apply_exclusions=True, header=False)
-df2 = result2.to_df()
-print(f"Shape: {df2.shape}")
-print("First row, first column:", df2.iloc[0, 0] if not df2.empty else "Empty")
-
-# The issue: include_outer_boundaries adds guides at 0 and 612,
-# which creates cells that span into the exclusion zone

temp/debug_st_search.py

@@ -1,33 +0,0 @@
-"""Debug searching for 'ST' text."""
-from natural_pdf import PDF
-
-pdf = PDF("pdfs/m27.pdf")
-page = pdf.pages[0]
-
-# Get the original ST element
-headers = (
-    page
-    .find("text:contains(NUMBER)")
-    .right(include_source=True)
-    .expand(top=3, bottom=3)
-    .find_all('text')
-)
-original_st = headers[4]
-print(f"Original 'ST' element: '{original_st.text}' at {original_st.bbox}")
-
-# Search for 'ST' using find
-found_st = page.find('text:contains("ST")')
-print(f"\nFound 'ST' using find: '{found_st.text}' at {found_st.bbox}")
-
-# Find all elements containing 'ST'
-all_st = page.find_all('text:contains("ST")')
-print(f"\nAll elements containing 'ST':")
-for i, elem in enumerate(all_st[:10]):  # First 10
-    print(f"  {i}: '{elem.text}' at x={elem.x0:.2f}, bbox={elem.bbox}")
-
-# Check what's at position 332.88
-print(f"\nLooking for element at x≈332.88:")
-all_text = page.find_all('text')
-for elem in all_text:
-    if 332 < elem.x0 < 334:
-        print(f"  Found: '{elem.text}' at x={elem.x0:.5f}, bbox={elem.bbox}")