Merge pull request #378 from chezou/update-encoding

Update encoding everytime when SubprocessTabule is initialized

tabula/backend.py

@@ -92,6 +92,20 @@ def __init__(
         self.java_options = java_options
         self.encoding = encoding
 
+    def update_encoding(
+        self, encoding: str, java_options: List[str], silent: Optional[bool]
+    ) -> None:
+        self.encoding = encoding
+        self.java_options = java_options
+        if silent:
+            self.java_options.extend(
+                (
+                    "-Dorg.slf4j.simpleLogger.defaultLogLevel=off",
+                    "-Dorg.apache.commons.logging.Log"
+                    "=org.apache.commons.logging.impl.NoOpLog",
+                )
+            )
+
     def call_tabula_java(
         self, options: TabulaOption, path: Optional[str] = None
     ) -> str:

tabula/io.py

@@ -76,6 +76,10 @@ def _run(
             _tabula_vm = SubprocessTabula(
                 java_options=java_options, silent=options.silent, encoding=encoding
             )
+    elif isinstance(_tabula_vm, SubprocessTabula):
+        _tabula_vm.update_encoding(
+            encoding=encoding, java_options=java_options, silent=options.silent
+        )
     elif set(java_options) - IGNORED_JAVA_OPTIONS:
         logger.warning("java_options is ignored until rebooting the Python process.")
 

tests/test_read_pdf_table.py

@@ -41,6 +41,9 @@ def test_read_pdf_with_force_subprocess(self):
         self.assertTrue(len(df), 1)
         self.assertTrue(isinstance(df[0], pd.DataFrame))
         self.assertTrue(df[0].equals(pd.read_csv(self.expected_csv1)))
+        self.assertTrue(tabula.io._tabula_vm.encoding, "utf-8")
+        tabula.read_pdf(self.pdf_path, stream=True, encoding="cp932")
+        self.assertTrue(tabula.io._tabula_vm.encoding, "cp932")
 
     def test_read_pdf_into_json(self):
         expected_json = "tests/resources/data_1.json"