ci: enhance CI testing by summarizing tests durations

.ci/collect_mapdl_logs_locals.sh

@@ -9,6 +9,8 @@ cp *pymapdl.apdl ./"$LOG_NAMES"/ || echo "No PYMAPDL APDL log files could be fou
 echo "Copying the profiling files..."
 cp -r prof ./"$LOG_NAMES"/prof || echo "No profile files could be found"
 
+echo "Copying the JSONL files..."
+cp *.jsonl ./"$LOG_NAMES"/ || echo "No JSONL files could be found"
 
 ls -la ./"$LOG_NAMES"
 

.ci/collect_mapdl_logs_remote.sh

@@ -38,6 +38,9 @@ cp mapdl_launch_1.log ./"$LOG_NAMES"/mapdl_launch_1.log || echo "MAPDL launch do
 echo "Copying the profiling files..."
 cp -r prof ./"$LOG_NAMES"/prof || echo "No profile files could be found"
 
+echo "Copying the JSONL files..."
+cp *.jsonl ./"$LOG_NAMES"/ || echo "No JSONL files could be found"
+
 echo "Collecting file structure..."
 ls -R > ./"$LOG_NAMES"/files_structure.txt || echo "Failed to copy file structure to a file"
 

.ci/pytest_summary.py

@@ -0,0 +1,251 @@
+import json
+import os
+
+import click
+import numpy as np
+
+BIG_WIDTH = 80
+SMALL_WIDTH = 8
+
+
+def find_json_files(base_dir):
+    """Recursively find all JSON files in subdirectories."""
+    json_files = []
+    for root, _, files in os.walk(base_dir):
+        for file in files:
+            if file.endswith(".jsonl"):
+                json_files.append(os.path.join(root, file))
+    return json_files
+
+
+def read_json_file(file_path):
+    """Read a JSON file and return its content as a list of test configurations."""
+    with open(file_path, "r", encoding="utf-8") as f:
+        try:
+            data = [json.loads(line) for line in f]
+            return data
+        except json.JSONDecodeError as e:
+            print(f"Error reading {file_path}: {e}")
+            return []
+
+
+def extract_tests_with_tags(json_files):
+    """Extract test data and assign a tag based on the directory name."""
+    tests = []
+
+    for file_path in json_files:
+        directory_name = os.path.basename(os.path.dirname(file_path))
+        test_data = read_json_file(file_path)
+
+        for test in test_data:
+            if test.get("outcome", "").lower() == "passed" and test.get("duration"):
+                nodeid = test.get("nodeid")
+                if nodeid.startswith("tests/"):
+                    nodeid = nodeid[6:]
+
+                when = test.get("when")
+                duration = test["duration"]
+                tags = directory_name.split("-")
+                tags.remove("logs")
+                id_ = f"{nodeid}({when})"
+
+                tests.append(
+                    {
+                        "tags": tags,
+                        "id": id_,
+                        "nodeid": nodeid,
+                        "duration": duration,
+                        "when": when,
+                    }
+                )
+    return tests
+
+
+def compute_statistics(tests):
+    """Compute average duration and standard deviation per test ID."""
+    test_stats = {}
+
+    for test in tests:
+        test_id = test["id"]
+        if test_id not in test_stats:
+            test_stats[test_id] = {
+                "durations": [],
+                "n_tests": 0,
+            }
+
+        test_stats[test_id]["durations"].append(test["duration"])
+        test_stats[test_id]["n_tests"] += 1
+
+    summary = []
+
+    for test_id, data in test_stats.items():
+        durations = np.array(data["durations"])
+
+        if durations.size == 0:
+            continue
+
+        avg_duration = np.mean(durations)
+        std_dev = np.std(durations)
+
+        mask_99 = durations < np.percentile(durations, 99)
+        if mask_99.sum() == 0:
+            avg_duration_minus_one = np.nan
+            std_dev_minus_one = np.nan
+        else:
+            avg_duration_minus_one = np.mean(durations[mask_99])
+            std_dev_minus_one = np.std(durations[mask_99])
+
+        mask_75 = durations < np.percentile(durations, 75)
+        if mask_75.sum() == 0:
+            avg_duration_minus_34 = np.nan
+            std_dev_minus_34 = np.nan
+        else:
+            avg_duration_minus_34 = np.mean(durations[mask_75])
+            std_dev_minus_34 = np.std(durations[mask_75])
+
+        summary.append(
+            {
+                "id": test_id,
+                "n_tests": data["n_tests"],
+                "average_duration": avg_duration,
+                "std_dev": std_dev,
+                "avg_duration_minus_one": avg_duration_minus_one,
+                "std_dev_minus_one": std_dev_minus_one,
+                "avg_duration_minus_34": avg_duration_minus_34,
+                "std_dev_minus_34": std_dev_minus_34,
+            }
+        )
+
+    return summary
+
+
+def print_table(data, keys, headers, title=""):
+    JUNCTION = "|"
+
+    def make_bold(s):
+        return click.style(s, bold=True)
+
+    h = [headers[0].ljust(BIG_WIDTH)]
+    h.extend([each.center(SMALL_WIDTH)[:SMALL_WIDTH] for each in headers[1:]])
+
+    len_h = len("| " + " | ".join(h) + " |")
+
+    sep = (
+        f"{JUNCTION}-"
+        + f"-{JUNCTION}-".join(["-" * len(each) for each in h])
+        + f"-{JUNCTION}"
+    )
+    top_sep = f"{JUNCTION}" + "-" * (len_h - 2) + f"{JUNCTION}"
+
+    if title:
+        # click.echo(top_sep)
+        click.echo(
+            "| " + make_bold(f"Top {len(data)} {title}".center(len_h - 4)) + " |"
+        )
+        click.echo(sep)
+
+    click.echo("| " + " | ".join([make_bold(each) for each in h]) + " |")
+    click.echo(sep)
+
+    for test in data:
+        s = []
+        for i, each_key in enumerate(keys):
+
+            if i == 0:
+                id_ = test[each_key]
+
+                id_ = (
+                    id_.replace("(", "\(")
+                    .replace(")", "\)")
+                    .replace("[", "\[")
+                    .replace("]", "\]")
+                )
+                if len(id_) >= BIG_WIDTH:
+                    id_ = id_[: BIG_WIDTH - 15] + "..." + id_[-12:]
+
+                s.append(f"{id_}".ljust(BIG_WIDTH)[0:BIG_WIDTH])
+
+            elif each_key == "n_tests":
+                s.append(f"{int(test[each_key])}".center(SMALL_WIDTH))
+            else:
+                if np.isnan(test[each_key]):
+                    s.append("N/A".center(SMALL_WIDTH))
+                else:
+                    s.append(f"{test[each_key]:.4f}".center(SMALL_WIDTH))
+
+        click.echo("| " + " | ".join(s) + " |")
+    # click.echo(sep)
+
+
+def print_summary(summary, num=10):
+    """Print the top N longest tests and the top N most variable tests."""
+    longest_tests = sorted(summary, key=lambda x: -x["average_duration"])[:num]
+    most_variable_tests = sorted(summary, key=lambda x: -x["std_dev"])[:num]
+
+    print(f"\n## Top {num} Longest Running Tests\n")
+    print_table(
+        longest_tests,
+        ["id", "n_tests", "average_duration", "std_dev"],
+        ["Test ID", "N. tests", "Avg", "STD"],
+        # "Longest Running Tests",
+    )
+
+    print("")
+    print(f"\n## Top {num} Most Variable Running Tests\n")
+    print_table(
+        most_variable_tests,
+        [
+            "id",
+            "n_tests",
+            "std_dev",
+            "average_duration",
+            "std_dev_minus_one",
+            "avg_duration_minus_one",
+            "std_dev_minus_34",
+            "avg_duration_minus_34",
+        ],
+        [
+            "Test ID",
+            "N. tests",
+            "Std",
+            "Avg",
+            "Std-99%",
+            "Avg-99%",
+            "Std-75%",
+            "Avg-75%",
+        ],
+        # "Most Variable Running Tests",
+    )
+
+
+@click.command()
+@click.option(
+    "--directory",
+    type=click.Path(exists=True, file_okay=False, dir_okay=True),
+    default=None,
+)
+@click.option(
+    "--num", default=10, help="Number of top tests to display.", show_default=True
+)
+@click.option(
+    "--save-file",
+    default=None,
+    help="File to save the test durations. Default 'tests_durations.json'.",
+    show_default=True,
+)
+def analyze_tests(directory, num, save_file):
+    directory = directory or os.getcwd()  # Change this to your base directory
+    json_files = find_json_files(directory)
+    tests = extract_tests_with_tags(json_files)
+
+    if save_file:
+        with open(save_file, "a+") as f:
+            for each_line in tests:
+                json.dump(each_line, f, indent=2)
+
+    summary = compute_statistics(tests)
+    print_summary(summary, num=num)
+
+
+if __name__ == "__main__":
+    analyze_tests()

.ci/requirements_minimal.txt → .ci/requirements_testing_minimal.txt

@@ -2,6 +2,7 @@ pyfakefs==5.7.2
 pytest-cov==6.0.0
 pytest-profiling==1.8.1
 pytest-random-order==1.1.1
+pytest-reportlog==0.4.0
 pytest-rerunfailures==15.0
 pytest-timeout==2.3.1
 pytest==8.3.4