Improve test summary

* Group failures together at the end
* Align summary with `fmt` and `lint` style

# Rust tests and lints will be skipped. Delete if not intended.
[ci skip-rust]

# Building wheels and fs_util will be skipped. Delete if not intended.
[ci skip-build-wheels]

src/python/pants/core/goals/test.py

@@ -7,7 +7,7 @@
 from dataclasses import dataclass
 from enum import Enum
 from pathlib import PurePath
-from typing import Dict, Iterable, List, Optional, Tuple, Type, TypeVar, cast
+from typing import Any, Dict, Iterable, List, Optional, Tuple, Type, TypeVar, Union, cast
 
 from pants.core.util_rules.filter_empty_sources import (
     FieldSetsWithSources,
@@ -136,6 +136,19 @@ def message(self) -> str:
             output = f"{output.rstrip()}\n\n"
         return f"{message}{output}"
 
+    def __lt__(self, other: Union[Any, "EnrichedTestResult"]) -> bool:
+        """We sort first by status (skipped vs failed vs succeeded), then alphanumerically within
+        each group."""
+        if not isinstance(other, EnrichedTestResult):
+            return NotImplemented
+        if self.exit_code == other.exit_code:
+            return self.address.spec < other.address.spec
+        if self.exit_code is None:
+            return True
+        if other.exit_code is None:
+            return False
+        return self.exit_code < other.exit_code
+
 
 @dataclass(frozen=True)
 class TestDebugRequest:
@@ -371,22 +384,21 @@ async def run_tests(
         Get(EnrichedTestResult, TestFieldSet, field_set) for field_set in field_sets_with_sources
     )
 
-    # Print summary
+    # Print summary.
     exit_code = 0
     if results:
         console.print_stderr("")
-    for result in results:
+    for result in sorted(results):
         if result.skipped:
             continue
-        result_desc = console.green("SUCCESS") if result.exit_code == 0 else console.red("FAILURE")
-        # The right-align logic sees the color control codes as characters, so we have
-        # to account for that. In f-strings the alignment field widths must be literals,
-        # so we have to indirect via a call to .format().
-        right_align = 19 if console.use_colors else 10
-        format_str = f"{{addr:80}}.....{{result:>{right_align}}}"
-        console.print_stderr(format_str.format(addr=result.address.spec, result=result_desc))
-        if result.exit_code is not None and result.exit_code != 0:
-            exit_code = result.exit_code
+        if result.exit_code == 0:
+            sigil = console.green("✓")
+            status = "succeeded"
+        else:
+            sigil = console.red("𐄂")
+            status = "failed"
+            exit_code = cast(int, result.exit_code)
+        console.print_stderr(f"{sigil} {result.address} {status}.")
 
     merged_xml_results = await Get(
         Digest, MergeDigests(result.xml_results for result in results if result.xml_results),

src/python/pants/core/goals/test_test.py

@@ -217,10 +217,10 @@ def test_summary(self) -> None:
         )
         assert exit_code == ConditionallySucceedsFieldSet.exit_code(bad_address)
         assert stderr == dedent(
-            f"""\
+            """\
 
-            {good_address}                                                                         .....   SUCCESS
-            {bad_address}                                                                          .....   FAILURE
+            ✓ //:good succeeded.
+            𐄂 //:bad failed.
             """
         )
 
@@ -242,6 +242,26 @@ def test_coverage(self) -> None:
         assert stderr.strip().endswith(f"Ran coverage on {addr1.spec}, {addr2.spec}")
 
 
+def sort_results() -> None:
+    create_test_result = partial(
+        EnrichedTestResult, stdout="", stderr="", output_setting=ShowOutput.ALL
+    )
+    skip1 = create_test_result(exit_code=None, address=Address("t1"))
+    skip2 = create_test_result(exit_code=None, address=Address("t2"))
+    success1 = create_test_result(exit_code=0, address=Address("t1"))
+    success2 = create_test_result(exit_code=0, address=Address("t2"))
+    fail1 = create_test_result(exit_code=1, address=Address("t1"))
+    fail2 = create_test_result(exit_code=1, address=Address("t2"))
+    assert sorted([fail2, success2, skip2, fail1, success1, skip1]) == [
+        skip1,
+        skip2,
+        success1,
+        success2,
+        fail1,
+        fail2,
+    ]
+
+
 def assert_streaming_output(
     *,
     exit_code: Optional[int],