[PATCH v2 01/11] scripts/coverage: initial coverage comparison script

["Alex Bennée" <alex.bennee@linaro.org>]

This is a very rough and ready first pass at comparing gcovr's json
output between two different runs. At the moment it will give you a
file level diff between two runs but hopefully it wont be too hard to
extend to give better insight.

After generating the coverage results you run with something like:

  ./scripts/coverage/compare_gcov_json.py \
    -a ./builds/gcov.config1/coverage.json \
    -b ./builds/gcov.config2/coverage.json

My hope is we can use this to remove some redundancy from testing as
well as evaluate if new tests are actually providing additional
coverage or just burning our precious CI time.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Kautuk Consul <kconsul@linux.vnet.ibm.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20230330101141.30199-2-alex.bennee@linaro.org>
---
 MAINTAINERS                           |   5 ++
 scripts/coverage/compare_gcov_json.py | 119 ++++++++++++++++++++++++++
 2 files changed, 124 insertions(+)
 create mode 100755 scripts/coverage/compare_gcov_json.py

diff --git a/MAINTAINERS b/MAINTAINERS
index ef45b5e71e..9e1a60ea24 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3908,3 +3908,8 @@ Performance Tools and Tests
 M: Ahmed Karaman <ahmedkhaledkaraman@gmail.com>
 S: Maintained
 F: scripts/performance/
+
+Code Coverage Tools
+M: Alex Bennée <alex.bennee@linaro.org>
+S: Odd Fixes
+F: scripts/coverage/
diff --git a/scripts/coverage/compare_gcov_json.py b/scripts/coverage/compare_gcov_json.py
new file mode 100755
index 0000000000..1b92dc2c8c
--- /dev/null
+++ b/scripts/coverage/compare_gcov_json.py
@@ -0,0 +1,119 @@
+#!/usr/bin/env python3
+#
+# Compare output of two gcovr JSON reports and report differences. To
+# generate the required output first:
+#   - create two build dirs with --enable-gcov
+#   - run set of tests in each
+#   - run make coverage-html in each
+#   - run gcovr --json --exclude-unreachable-branches \
+#           --print-summary -o coverage.json --root ../../ . *.p
+#
+# Author: Alex Bennée <alex.bennee@linaro.org>
+#
+# SPDX-License-Identifier: GPL-2.0-or-later
+#
+
+import argparse
+import json
+import sys
+from pathlib import Path
+
+def create_parser():
+    parser = argparse.ArgumentParser(
+        prog='compare_gcov_json',
+        description='analyse the differences in coverage between two runs')
+
+    parser.add_argument('-a', type=Path, default=None,
+                        help=('First file to check'))
+
+    parser.add_argument('-b', type=Path, default=None,
+                        help=('Second file to check'))
+
+    parser.add_argument('--verbose', action='store_true', default=False,
+                        help=('A minimal verbosity level that prints the '
+                              'overall result of the check/wait'))
+    return parser
+
+
+# See https://gcovr.com/en/stable/output/json.html#json-format-reference
+def load_json(json_file_path: Path, verbose = False) -> dict[str, set[int]]:
+
+    with open(json_file_path) as f:
+        data = json.load(f)
+
+    root_dir = json_file_path.absolute().parent
+    covered_lines = dict()
+
+    for filecov in data["files"]:
+        file_path = Path(filecov["file"])
+
+        # account for generated files - map into src tree
+        resolved_path = Path(file_path).absolute()
+        if resolved_path.is_relative_to(root_dir):
+            file_path = resolved_path.relative_to(root_dir)
+            # print(f"remapped {resolved_path} to {file_path}")
+
+        lines = filecov["lines"]
+
+        executed_lines = set(
+            linecov["line_number"]
+            for linecov in filecov["lines"]
+            if linecov["count"] != 0 and not linecov["gcovr/noncode"]
+        )
+
+        # if this file has any coverage add it to the system
+        if len(executed_lines) > 0:
+            if verbose:
+                print(f"file {file_path} {len(executed_lines)}/{len(lines)}")
+            covered_lines[str(file_path)] = executed_lines
+
+    return covered_lines
+
+def find_missing_files(first, second):
+    """
+    Return a list of files not covered in the second set
+    """
+    missing_files = []
+    for f in sorted(first):
+        file_a = first[f]
+        try:
+            file_b = second[f]
+        except KeyError:
+            missing_files.append(f)
+
+    return missing_files
+
+def main():
+    """
+    Script entry point
+    """
+    parser = create_parser()
+    args = parser.parse_args()
+
+    if not args.a or not args.b:
+        print("We need two files to compare")
+        sys.exit(1)
+
+    first_coverage = load_json(args.a, args.verbose)
+    second_coverage = load_json(args.b, args.verbose)
+
+    first_missing = find_missing_files(first_coverage,
+                                       second_coverage)
+
+    second_missing = find_missing_files(second_coverage,
+                                        first_coverage)
+
+    a_name = args.a.parent.name
+    b_name = args.b.parent.name
+
+    print(f"{b_name} missing coverage in {len(first_missing)} files")
+    for f in first_missing:
+        print(f"  {f}")
+
+    print(f"{a_name} missing coverage in {len(second_missing)} files")
+    for f in second_missing:
+        print(f"  {f}")
+
+
+if __name__ == '__main__':
+    main()
-- 
2.39.2