fix(tests): silence all remaining noisy test suites

ReadAlongs · Nov 7, 2024 · 65149e5 · 65149e5
1 parent 99da050
commit 65149e5
Show file tree

Hide file tree

Showing 7 changed files with 131 additions and 78 deletions.
diff --git a/test/test_api.py b/test/test_api.py
@@ -5,6 +5,8 @@
 """
 
 import os
+from contextlib import redirect_stderr
+from io import StringIO
 from unittest import main
 
 import click
@@ -23,13 +25,14 @@ def test_call_align(self):
         # API accepts them too.
         langs = ("fra",)  # make sure language can be an iterable, not just a list.
         with SoundSwallowerStub("t0b0d0p0s0w0:920:1520", "t0b0d0p0s1w0:1620:1690"):
-            (status, exception, log) = api.align(
-                self.data_dir / "ej-fra.txt",
-                self.data_dir / "ej-fra.m4a",
-                self.tempdir / "output",
-                langs,
-                output_formats=["html", "TextGrid", "srt"],
-            )
+            with redirect_stderr(StringIO()):
+                (status, exception, log) = api.align(
+                    self.data_dir / "ej-fra.txt",
+                    self.data_dir / "ej-fra.m4a",
+                    self.tempdir / "output",
+                    langs,
+                    output_formats=["html", "TextGrid", "srt"],
+                )
         self.assertEqual(status, 0)
         self.assertTrue(exception is None)
         self.assertIn("Words (<w>) not present; tokenizing", log)
@@ -53,16 +56,18 @@ def test_call_align(self):
             "Make sure the API call doesn't not modify my variables",
         )
 
-        (status, exception, log) = api.align("", "", self.tempdir / "errors")
+        with redirect_stderr(StringIO()):
+            (status, exception, log) = api.align("", "", self.tempdir / "errors")
         self.assertNotEqual(status, 0)
         self.assertFalse(exception is None)
 
     def test_call_make_xml(self):
-        (status, exception, log) = api.make_xml(
-            self.data_dir / "ej-fra.txt",
-            self.tempdir / "prepared.readalong",
-            ("fra", "eng"),
-        )
+        with redirect_stderr(StringIO()):
+            (status, exception, log) = api.make_xml(
+                self.data_dir / "ej-fra.txt",
+                self.tempdir / "prepared.readalong",
+                ("fra", "eng"),
+            )
         self.assertEqual(status, 0)
         self.assertTrue(exception is None)
         self.assertIn("Wrote ", log)

diff --git a/test/test_audio.py b/test/test_audio.py
@@ -41,9 +41,6 @@ def align(self, input_text_path, input_audio_path, output_path, flags):
             input_audio_path,
             output_path,
         ] + flags
-        LOGGER.info(
-            f"Aligning {input_text_path} and {input_audio_path}, outputting to {output_path}"
-        )
         return run(args, capture_output=True, check=False, encoding="utf-8")
 
     def test_mute_section(self):

diff --git a/test/test_dna_text.py b/test/test_dna_text.py
@@ -2,6 +2,8 @@
 
 """Test handling of DNA text in tokenization"""
 
+from contextlib import redirect_stderr
+from io import StringIO
 from unittest import main
 
 from basic_test_case import BasicTestCase
@@ -23,7 +25,8 @@ def test_tok_all_words(self):
 <s>Voici une deuxième phrase.</s>
 </document>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         as_txt = etree.tounicode(tokenized)
         # print(etree.tounicode(tokenized))
 
@@ -54,7 +57,8 @@ def test_tok_some_words(self):
 <s>Un <foo do-not-align="1">mot ou deux</foo> à exclure.</s>
 </document>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         as_txt = etree.tounicode(tokenized)
         # print('as_txt="' + as_txt +'"')
 
@@ -96,7 +100,8 @@ def test_tok_div_p_s(self):
 </div>
 </document>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         as_txt = etree.tounicode(tokenized)
         # print('as_txt="' + as_txt +'"')
 
@@ -143,15 +148,17 @@ def test_dna_word(self):
 
         txt = """<s xml:lang="fra">Une <w do-not-align="true">exclude</w> phrase.</s>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         self.assertRaises(RuntimeError, add_ids, tokenized)
 
     def test_dna_word_nested(self):
         """You also can't have a <w> element inside a DNA element"""
 
         txt = """<s xml:lang="fra">Une <foo do-not-align="true"><bar><w>exclude</w></bar></foo> phrase.</s>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         self.assertRaises(RuntimeError, add_ids, tokenized)
 
 

diff --git a/test/test_g2p_cli.py b/test/test_g2p_cli.py
@@ -4,6 +4,8 @@
 
 import os
 import re
+from contextlib import redirect_stderr
+from io import StringIO
 from unittest import main
 
 from basic_test_case import BasicTestCase
@@ -303,8 +305,9 @@ def test_align_with_invalid_preg2p(self):
         self.assertIn('<w ARPABET="NOT ARPABET" id="s0w2">error</w>', results.output)
 
         audio_file = os.path.join(self.data_dir, "ej-fra.m4a")
-        with self.assertRaises(RuntimeError) as e:
-            results = align_audio(input_file, audio_file)
+        with redirect_stderr(StringIO()):
+            with self.assertRaises(RuntimeError) as e:
+                results = align_audio(input_file, audio_file)
         self.assertIn("could not be g2p'd", str(e.exception))
 
     def test_align_with_preg2p(self):
@@ -330,9 +333,10 @@ def test_align_with_preg2p(self):
             "t0b0d0p0s3w2:15:16",
             "t0b0d0p0s3w3:16:17",
         ):
-            _ = align_audio(
-                text_file, audio_file, save_temps=os.path.join(self.tempdir, "foo")
-            )
+            with redirect_stderr(StringIO()):
+                _ = align_audio(
+                    text_file, audio_file, save_temps=os.path.join(self.tempdir, "foo")
+                )
         with open(os.path.join(self.tempdir, "foo.dict"), "r", encoding="utf8") as f:
             dict_file = f.read()
             self.assertIn("S AH S IY", dict_file)  # "ceci" in fra
@@ -452,7 +456,8 @@ def test_convert_xml_invalid(self):
         self.assertTrue(valid, "convert_xml with valid pre-g2p'd text")
 
         xml = parse_xml('<s><w ARPABET="invalid">invalid</w></s>')
-        c_xml, valid = convert_xml(xml)
+        with redirect_stderr(StringIO()):
+            c_xml, valid = convert_xml(xml)
         self.assertEqual(
             etree.tounicode(c_xml), '<s><w ARPABET="invalid">invalid</w></s>'
         )

diff --git a/test/test_misc.py b/test/test_misc.py
@@ -293,9 +293,9 @@ def test_capture_logs_some_more(self):
         with capture_logs() as captured_logs:
             LOGGER.info("this will be captured")
         self.assertIn("this will be captured", captured_logs.getvalue())
-        with self.assertLogs():
+        with self.assertLogs(LOGGER):
             LOGGER.info("blah")
-        with self.assertLogs() as cm:
+        with self.assertLogs(LOGGER) as cm:
             with capture_logs() as captured_logs:
                 LOGGER.info("This text does not propagate to root")
             LOGGER.info("This text is included in root")

diff --git a/test/test_tokenize_xml.py b/test/test_tokenize_xml.py
@@ -2,6 +2,8 @@
 
 """Unit test suite for our XML tokenizer module"""
 
+from contextlib import redirect_stderr
+from io import StringIO
 from unittest import TestCase, main
 
 from lxml import etree
@@ -23,7 +25,8 @@ def test_simple(self):
 <s xml:lang="atj"><w>Kwei</w>! <w>Tan</w> <w>e</w> <w>ici</w> <w>matisihin</w>?</s>
 </document>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         # print(etree.tounicode(tokenized))
         self.assertEqual(etree.tounicode(tokenized), ref)
 
@@ -39,7 +42,8 @@ def test_mixed_lang(self):
 <s xml:lang="fra"><w>Bonjour</w>! <w>Comment</w> <w>ça</w> <w>va</w>?</s>
 </document>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         # print(etree.tounicode(tokenized))
         self.assertEqual(etree.tounicode(tokenized), ref)
 
@@ -55,7 +59,8 @@ def test_mixed_words(self):
 <s xml:lang="atj">Tan e ici matisihin?</s>
 </document>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         # print(etree.tounicode(tokenized))
         self.assertEqual(etree.tounicode(tokenized), ref)
 
@@ -73,7 +78,8 @@ def test_comments(self):
 <s xml:lang="atj"><w>Tan</w> <w>e</w> <w>ici</w> <w>matisihin</w>?</s>
 </document>"""
         xml = parse_xml(txt)
-        tokenized = tokenize_xml.tokenize_xml(xml)
+        with redirect_stderr(StringIO()):
+            tokenized = tokenize_xml.tokenize_xml(xml)
         # print(etree.tounicode(tokenized))
         self.assertEqual(etree.tounicode(tokenized), ref)