Honor pdftotree's loglevel when calling tabula (#103)

Author: HiromuHota

CHANGELOG.md

@@ -10,6 +10,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Embed base64-encoded images inline. Support starting with JPEG and BMP.
   ([#99](https://github.com/HazyResearch/pdftotree/pull/99), [@HiromuHota][HiromuHota])
 
+### Changed
+- Suppress tabula-java's log messages unless pdftotree's logger is set logging.DEBUG.
+  ([#103](https://github.com/HazyResearch/pdftotree/pull/103), [@HiromuHota][HiromuHota])
+
 ### Fixed
 - List a missing "ocrx_line" in the ocr-capabilities metadata field.
   ([#94](https://github.com/HazyResearch/pdftotree/issues/94), [@HiromuHota][HiromuHota])

README.rst

@@ -43,9 +43,11 @@ pdftotree as a Python package
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. code:: python
+    # Uncomment the followings if tabula should not be silent.
+    # import logging
+    # logging.getLogger("pdftotree").setLevel(logging.DEBUG)
 
     import pdftotree
-
     pdftotree.parse(pdf_file, html_path=None, model_type=None, model_path=None, visualize=False):
 
 pdftotree
@@ -73,7 +75,7 @@ This takes a PDF file as input and produces an hOCR file as output::
                             printed to stdout.
       -V, --visualize       Whether to output visualization images
       -v, --verbose         Output INFO level logging.
-      -vv, --veryverbose    Output DEBUG level logging.
+      -vv, --veryverbose    Output DEBUG level logging. Use this if tabula should not be silent.
 
 extract\_tables
 ~~~~~~~~~~~~~~~

bin/pdftotree

@@ -58,7 +58,7 @@ if __name__ == "__main__":
         "--veryverbose",
         dest="debug",
         action="store_true",
-        help="Output DEBUG level logging.",
+        help="Output DEBUG level logging. Use this if tabula should not be silent.",
     )
     parser.set_defaults(visualize=False)
     args = parser.parse_args()

pdftotree/TreeExtract.py

@@ -446,8 +446,13 @@ def get_html_table(self, table: List[float], page_num) -> Optional[Element]:
         :return: DOM element for a table
         """
         logger.debug(f"Calling tabula at page: {page_num} and area: {table}.")
+        loglevel = logging.getLogger("pdftotree").getEffectiveLevel()
         table_json = tabula.read_pdf(
-            self.pdf_file, pages=page_num, area=table, output_format="json"
+            self.pdf_file,
+            pages=page_num,
+            area=table,
+            output_format="json",
+            silent=False if loglevel <= logging.DEBUG else True,
         )
         logger.debug(f"Tabula recognized {len(table_json)} table(s).")
         if len(table_json) == 0:

tests/test_basic.py

@@ -1,3 +1,4 @@
+import logging
 import os
 from subprocess import PIPE, Popen
 from typing import Optional
@@ -73,6 +74,19 @@ def test_no_out_of_order(caplog):
     assert "Out of order" not in caplog.text
 
 
+def test_tabula_warning_suppressed(caplog):
+    """Test if tabula warnings are suppressed."""
+    # Warnings suppressed by default
+    pdftotree.parse("tests/input/112823.pdf")
+    assert "org.apache.pdfbox" not in caplog.text
+
+    # Not to suppress warnings
+    log = logging.getLogger("pdftotree")
+    log.setLevel(logging.DEBUG)
+    pdftotree.parse("tests/input/112823.pdf")
+    assert "org.apache.pdfbox" in caplog.text
+
+
 def test_visualize_output(tmp_path):
     """Test if an output can be visualzied."""
     html_path = os.path.join(tmp_path, "md.html")