natural-pdf 0.1.4__py3-none-any.whl → 0.1.5__py3-none-any.whl

pdfs/01-practice.pdf

@@ -0,0 +1,543 @@
+%PDF-1.7
+%µ¶
+
+1 0 obj
+<</Type/Pages/Count 1/Kids[2 0 R]>>
+endobj
+
+2 0 obj
+<</Type/Page/MediaBox[0 0 612 792]/Resources 3 0 R/Parent 1 0 R/Contents 6 0 R>>
+endobj
+
+3 0 obj
+<</Font<</F0<</Type/Font/BaseFont/Helvetica/Subtype/Type1>>/F1<</Type/Font/BaseFont/Helvetica-Bold/Subtype/Type1>>>>>>
+endobj
+
+4 0 obj
+<</Length 4482>>
+stream
+q
+    1 0 0 1 50 700 cm
+    1 g
+    325 65 180 -35 re
+    S
+Q
+q
+    1 0 0 1 50 700 cm
+    BT
+        335 50 Td
+        /F0 8 Tf
+        (Jungle Health and Safety Inspection Service) Tj
+        0 -10 Td
+        1 0 0 rg
+        (INS-UP70N51NCL41R) Tj
+    ET
+Q
+q
+    1 0 0 1 50 700 cm
+    BT
+        /F1 10 Tf
+        (Site: ) Tj
+
+        /F0 10 Tf
+        (Durham's Meatpacking  ) Tj
+
+        0.5 g
+        (Chicago, Ill.) Tj
+        0 g
+
+        0 -20 Td
+
+        /F1 10 Tf
+        (Date:  ) Tj
+
+        /F0 10 Tf
+        (February 3, 1905) Tj
+
+        0 -20 Td
+
+        /F1 10 Tf
+        (Violation Count: ) Tj
+
+        /F0 10 Tf
+        (7) Tj
+
+        0 -20 Td
+
+        /F1 10 Tf
+        (Summary: ) Tj
+
+        % 0 -20 Td
+
+        /F0 10 Tf
+        (Worst of any, however, were the fertilizer men, and those who served in the cooking rooms.) Tj
+        0 -16 Td
+        (These people could not be shown to the visitor - for the odor of a fertilizer man would scare any ordinary ) Tj
+        0 -16 Td
+        (visitor at a hundred yards, and as for the other men, who worked in tank rooms full of steam, and in ) Tj
+        0 -16 Td
+        (some of which there were open vats near the level of the floor, their peculiar trouble was that they fell) Tj
+        0 -16 Td
+        (into the vats; and when they were fished out, there was never enough of them left to be worth ) Tj
+        0 -16 Td
+        (exhibiting - sometimes they would be overlooked for days, till all but the bones of them had gone out) Tj
+        0 -16 Td
+        (to the world as Durham's Pure Leaf Lard!) Tj
+    ET
+Q
+
+q
+    1 0 0 1 50 440 cm
+    0 G
+    2 w
+    0 0 m
+    500 0 l
+    S
+
+    BT
+    0 -30 Td
+    /F1 12 Tf
+    (Violations) Tj
+    ET
+Q
+
+
+q
+    1 0 0 1 50 400 cm
+    0.5 G
+    0 0 m
+    500 0 l
+    S
+    0 -20 m
+    500 -20 l
+    S
+    0 -40 m
+    500 -40 l
+    S
+    0 -60 m
+    500 -60 l
+    S
+    0 -80 m
+    500 -80 l
+    S
+    0 -100 m
+    500 -100 l
+    S
+    0 -120 m
+    500 -120 l
+    S
+    0 -140 m
+    500 -140 l
+    S
+    0 -160 m
+    500 -160 l
+    S
+
+    0 0 m
+    0 -160 l
+    S
+
+    50 0 m
+    50 -160 l
+    S
+
+    400 0 m
+    400 -160 l
+    S
+
+    450 0 m
+    450 -160 l
+    S
+
+    500 0 m
+    500 -160 l
+    S
+Q
+
+q
+    1 0 0 1 55 386 cm
+    BT
+        q
+            /F1 10 Tf
+            (Statute) Tj
+            50 0 Td
+            (Description) Tj
+            350 0 Td
+            (Level) Tj
+            50 0 Td
+            (Repeat?) Tj
+        Q
+
+        /F0 10 Tf
+        q
+            0 -20 Td
+            (4.12.7) Tj
+            50 0 Td
+            (Unsanitary Working Conditions.) Tj
+            350 0 Td
+            (Critical) Tj
+        Q
+        q
+            0 -40 Td
+            (5.8.3) Tj
+            50 0 Td
+            (Inadequate Protective Equipment.) Tj
+            350 0 Td
+            (Serious) Tj
+        Q
+        q
+            0 -60 Td
+            (6.3.9) Tj
+            50 0 Td
+            (Ineffective Injury Prevention.) Tj
+            350 0 Td
+            (Serious) Tj
+        Q
+        q
+            0 -80 Td
+            (7.1.5) Tj
+            50 0 Td
+            (Failure to Properly Store Hazardous Materials.) Tj
+            350 0 Td
+            (Critical) Tj
+        Q
+        q
+            0 -100 Td
+            (8.9.2) Tj
+            50 0 Td
+            (Lack of Adequate Fire Safety Measures.) Tj
+            350 0 Td
+            (Serious) Tj
+        Q
+        q
+            0 -120 Td
+            (9.6.4) Tj
+            50 0 Td
+            (Inadequate Ventilation Systems.) Tj
+            350 0 Td
+            (Serious) Tj
+        Q
+        q
+            0 -140 Td
+            (10.2.7) Tj
+            50 0 Td
+            (Insufficient Employee Training for Safe Work Practices.) Tj
+            350 0 Td
+            (Serious) Tj
+        Q
+        
+    ET
+
+    q
+        0 G
+        0.5 w
+
+        465 -20 8 8 re
+        S
+        465 -20 m
+        473 -12 l
+        s
+        465 -12 m
+        473 -20 l
+        s
+
+        465 -40 8 8 re
+        S
+        465 -40 m
+        473 -32 l
+        s
+        465 -32 m
+        473 -40 l
+        s
+
+        465 -60 8 8 re
+        S
+
+        465 -80 8 8 re
+        S
+        
+        465 -100 8 8 re
+        S
+
+        465 -120 8 8 re
+        S
+        465 -120 m
+        473 -112 l
+        s
+        465 -112 m
+        473 -120 l
+        s
+
+        465 -140 8 8 re
+        S
+    Q
+Q
+
+q
+    1 0 0 1 230 20 cm
+    BT
+        /F0 8 Tf
+        (Jungle Health and Safety Inspection Service) Tj
+    ET
+Q
+
+
+endstream
+endobj
+
+5 0 obj
+<</Type/Catalog/Pages 1 0 R>>
+endobj
+
+6 0 obj
+<</Length 2774>>
+stream
+q
+1 0 0 1 50 700 cm
+1 g
+325 65 180 -35 re
+S
+Q
+q
+1 0 0 1 50 700 cm
+BT
+335 50 Td
+/F0 8 Tf
+(Jungle Health and Safety Inspection Service) Tj
+0 -10 Td
+1 0 0 rg
+(INS-UP70N51NCL41R) Tj
+ET
+Q
+q
+1 0 0 1 50 700 cm
+BT
+/F1 10 Tf
+(Site: ) Tj
+/F0 10 Tf
+(Durham's Meatpacking  ) Tj
+.5 g
+(Chicago, Ill.) Tj
+0 g
+0 -20 Td
+/F1 10 Tf
+(Date:  ) Tj
+/F0 10 Tf
+(February 3, 1905) Tj
+0 -20 Td
+/F1 10 Tf
+(Violation Count: ) Tj
+/F0 10 Tf
+(7) Tj
+0 -20 Td
+/F1 10 Tf
+(Summary: ) Tj
+/F0 10 Tf
+(Worst of any, however, were the fertilizer men, and those who served in the cooking rooms.) Tj
+0 -16 Td
+(These people could not be shown to the visitor - for the odor of a fertilizer man would scare any ordinary ) Tj
+0 -16 Td
+(visitor at a hundred yards, and as for the other men, who worked in tank rooms full of steam, and in ) Tj
+0 -16 Td
+(some of which there were open vats near the level of the floor, their peculiar trouble was that they fell) Tj
+0 -16 Td
+(into the vats; and when they were fished out, there was never enough of them left to be worth ) Tj
+0 -16 Td
+(exhibiting - sometimes they would be overlooked for days, till all but the bones of them had gone out) Tj
+0 -16 Td
+(to the world as Durham's Pure Leaf Lard!) Tj
+ET
+Q
+q
+1 0 0 1 50 440 cm
+0 G
+2 w
+0 0 m
+500 0 l
+S
+BT
+0 -30 Td
+/F1 12 Tf
+(Violations) Tj
+ET
+Q
+q
+1 0 0 1 50 400 cm
+.5 G
+0 0 m
+500 0 l
+S
+0 -20 m
+500 -20 l
+S
+0 -40 m
+500 -40 l
+S
+0 -60 m
+500 -60 l
+S
+0 -80 m
+500 -80 l
+S
+0 -100 m
+500 -100 l
+S
+0 -120 m
+500 -120 l
+S
+0 -140 m
+500 -140 l
+S
+0 -160 m
+500 -160 l
+S
+0 0 m
+0 -160 l
+S
+50 0 m
+50 -160 l
+S
+400 0 m
+400 -160 l
+S
+450 0 m
+450 -160 l
+S
+500 0 m
+500 -160 l
+S
+Q
+q
+1 0 0 1 55 386 cm
+BT
+q
+/F1 10 Tf
+(Statute) Tj
+50 0 Td
+(Description) Tj
+350 0 Td
+(Level) Tj
+50 0 Td
+(Repeat?) Tj
+Q
+/F0 10 Tf
+q
+0 -20 Td
+(4.12.7) Tj
+50 0 Td
+(Unsanitary Working Conditions.) Tj
+350 0 Td
+(Critical) Tj
+Q
+q
+0 -40 Td
+(5.8.3) Tj
+50 0 Td
+(Inadequate Protective Equipment.) Tj
+350 0 Td
+(Serious) Tj
+Q
+q
+0 -60 Td
+(6.3.9) Tj
+50 0 Td
+(Ineffective Injury Prevention.) Tj
+350 0 Td
+(Serious) Tj
+Q
+q
+0 -80 Td
+(7.1.5) Tj
+50 0 Td
+(Failure to Properly Store Hazardous Materials.) Tj
+350 0 Td
+(Critical) Tj
+Q
+q
+0 -100 Td
+(8.9.2) Tj
+50 0 Td
+(Lack of Adequate Fire Safety Measures.) Tj
+350 0 Td
+(Serious) Tj
+Q
+q
+0 -120 Td
+(9.6.4) Tj
+50 0 Td
+(Inadequate Ventilation Systems.) Tj
+350 0 Td
+(Serious) Tj
+Q
+q
+0 -140 Td
+(10.2.7) Tj
+50 0 Td
+(Insufficient Employee Training for Safe Work Practices.) Tj
+350 0 Td
+(Serious) Tj
+Q
+ET
+q
+0 G
+.5 w
+465 -20 8 8 re
+S
+465 -20 m
+473 -12 l
+s
+465 -12 m
+473 -20 l
+s
+465 -40 8 8 re
+S
+465 -40 m
+473 -32 l
+s
+465 -32 m
+473 -40 l
+s
+465 -60 8 8 re
+S
+465 -80 8 8 re
+S
+465 -100 8 8 re
+S
+465 -120 8 8 re
+S
+465 -120 m
+473 -112 l
+s
+465 -112 m
+473 -120 l
+s
+465 -140 8 8 re
+S
+Q
+Q
+q
+1 0 0 1 230 20 cm
+BT
+/F0 8 Tf
+(Jungle Health and Safety Inspection Service) Tj
+ET
+Q
+
+endstream
+endobj
+
+xref
+0 7
+0000000000 65536 f 
+0000000016 00000 n 
+0000000068 00000 n 
+0000000165 00000 n 
+0000000300 00000 n 
+0000004833 00000 n 
+0000004879 00000 n 
+
+trailer
+<</Size 7/Root 5 0 R>>
+startxref
+7704
+%%EOF

tests/test_loading.py

@@ -0,0 +1,50 @@
+import os
+
+import pytest
+
+from natural_pdf import PDF
+
+# URL for the test PDF used in the tutorial
+TEST_PDF_URL = "https://github.com/jsoma/natural-pdf/raw/refs/heads/main/pdfs/01-practice.pdf"
+
+
+def test_pdf_loading_from_url():
+    """Tests if a PDF can be loaded successfully from a URL."""
+    try:
+        pdf = PDF(TEST_PDF_URL)
+        # Basic assertions after loading
+        assert pdf is not None
+        assert len(pdf.pages) > 0, "PDF should have at least one page"
+        assert os.path.exists(pdf.path), "PDF file should be downloaded locally"
+        # Check if metadata (like Title) is accessible, even if None
+        assert "Title" in pdf.metadata or pdf.metadata.get("Title") is None
+
+    except Exception as e:
+        pytest.fail(f"PDF loading from URL failed: {e}")
+
+
+def test_page_text_extraction():
+    """Tests if text can be extracted from the first page."""
+    try:
+        pdf = PDF(TEST_PDF_URL)
+        assert len(pdf.pages) > 0, "PDF has no pages"
+        page = pdf.pages[0]
+        text = page.extract_text()
+        assert isinstance(text, str), "Extracted text should be a string"
+        assert len(text) > 50, "Extracted text seems too short or empty"
+        # Add a more specific assertion if you know some expected text
+        # assert "Expected sample text" in text
+
+    except Exception as e:
+        pytest.fail(f"Text extraction failed: {e}")
+
+
+# Clean up downloaded file if necessary (optional, depends on PDF class behavior)
+# You might want a fixture to handle setup/teardown of the downloaded file
+# @pytest.fixture(scope="module")
+# def downloaded_pdf():
+#     pdf = PDF(TEST_PDF_URL)
+#     yield pdf
+#     # Cleanup code here if PDF() doesn't handle it
+#     if os.path.exists(pdf.path):
+#         os.remove(pdf.path)