BUG: Fix pd.show_versions as_json invalid JSON (pandas-dev#39701)

Author: alexprincel

doc/source/whatsnew/v1.3.0.rst

@@ -461,6 +461,7 @@ Other
 - Bug in :class:`Styler` where ``subset`` arg in methods raised an error for some valid multiindex slices (:issue:`33562`)
 - :class:`Styler` rendered HTML output minor alterations to support w3 good code standard (:issue:`39626`)
 - Bug in :meth:`DataFrame.equals`, :meth:`Series.equals`, :meth:`Index.equals` with object-dtype containing ``np.datetime64("NaT")`` or ``np.timedelta64("NaT")`` (:issue:`39650`)
+- Bug in :func:`pandas.util.show_versions` where console JSON output was not proper JSON (:issue:`39701`)
 
 
 .. ---------------------------------------------------------------------------

pandas/tests/util/test_show_versions.py

@@ -1,7 +1,11 @@
+import json
+import os
 import re
 
 import pytest
 
+from pandas.util._print_versions import _get_dependency_info, _get_sys_info
+
 import pandas as pd
 
 
@@ -26,21 +30,65 @@
     "ignore:Distutils:UserWarning"
 )
 @pytest.mark.filterwarnings("ignore:Setuptools is replacing distutils:UserWarning")
-def test_show_versions(capsys):
+@pytest.mark.parametrize("as_json", [True, False, "test_output.json"])
+def test_show_versions(capsys, as_json, tmpdir):
     # gh-32041
-    pd.show_versions()
+    if isinstance(as_json, str):
+        as_json = os.path.join(tmpdir, as_json)
+
+    pd.show_versions(as_json=as_json)
     captured = capsys.readouterr()
     result = captured.out
 
-    # check header
-    assert "INSTALLED VERSIONS" in result
+    # check header for non-JSON console output
+    if as_json is False:
+        assert "INSTALLED VERSIONS" in result
+
+        # check full commit hash
+        assert re.search(r"commit\s*:\s[0-9a-f]{40}\n", result)
+
+        # check required dependency
+        # 2020-12-09 npdev has "dirty" in the tag
+        assert re.search(r"numpy\s*:\s([0-9\.\+a-g\_]|dev)+(dirty)?\n", result)
+
+        # check optional dependency
+        assert re.search(r"pyarrow\s*:\s([0-9\.]+|None)\n", result)
+
+    # Dictionary-based asserts
+    else:
+        # check valid json is printed to the console if as_json is True
+        if as_json is True:
+            dict_check = json.loads(result)
+        elif isinstance(as_json, str):
+            # make sure that the file was created
+            assert os.path.exists(as_json)
+
+            with open(as_json) as fd:
+                contents = fd.readlines()
+                str_contents = "".join(contents)
+
+                # make sure that there was output to the file
+                assert str_contents
+
+                # check if file output is valid JSON
+                dict_check = json.loads(str_contents)
+
+        # Basic check that each version element is found in output
+        version_elements = {
+            "system": _get_sys_info(),
+            "dependencies": _get_dependency_info(),
+        }
+
+        assert version_elements == dict_check
+
 
-    # check full commit hash
-    assert re.search(r"commit\s*:\s[0-9a-f]{40}\n", result)
+def test_json_output_match(capsys, tmpdir):
+    pd.show_versions(as_json=True)
+    result_console = capsys.readouterr().out
 
-    # check required dependency
-    # 2020-12-09 npdev has "dirty" in the tag
-    assert re.search(r"numpy\s*:\s([0-9\.\+a-g\_]|dev)+(dirty)?\n", result)
+    out_path = os.path.join(tmpdir, "test_json.json")
+    pd.show_versions(as_json=out_path)
+    with open(out_path) as out_fd:
+        result_file = "".join(out_fd.readlines())
 
-    # check optional dependency
-    assert re.search(r"pyarrow\s*:\s([0-9\.]+|None)\n", result)
+    assert result_console == result_file

pandas/util/_print_versions.py

@@ -107,7 +107,7 @@ def show_versions(as_json: Union[str, bool] = False) -> None:
         j = {"system": sys_info, "dependencies": deps}
 
         if as_json is True:
-            print(j)
+            sys.stdout.writelines(json.dumps(j, indent=2))
         else:
             assert isinstance(as_json, str)  # needed for mypy
             with codecs.open(as_json, "wb", encoding="utf8") as f: