From dac3aeaafb817e2ec987e510542e07fc9db7f74f Mon Sep 17 00:00:00 2001
From: Maxim Vershinin <im@vmn.su>
Date: Mon, 28 Sep 2026 18:52:32 +0400
Subject: [PATCH] feat(project-setup): add CLI entry point and __main__.py

---
 scripts/cli.py               | 97 ++++++++++++++++++++++++++++++++++++
 src/disk_checker/__main__.py |  6 +++
 2 files changed, 103 insertions(+)
 create mode 100644 scripts/cli.py
 create mode 100644 src/disk_checker/__main__.py

diff --git a/scripts/cli.py b/scripts/cli.py
new file mode 100644
index 0000000..0b713f7
--- /dev/null
+++ b/scripts/cli.py
@@ -0,0 +1,97 @@
+#!/usr/bin/env python3
+"""
+CLI entry point for disk-checker.
+
+Usage:
+    python -m disk_checker scan_latency.csv
+    disk-checker scan_latency.csv --metric read_clat --unit ms
+"""
+
+import argparse
+import sys
+from pathlib import Path
+
+from disk_checker.core.extractor import load_latency_csv
+from disk_checker.core.helpers import percentile_table
+from disk_checker.plotting.plots import generate_plot
+
+
+def main():
+    parser = argparse.ArgumentParser(
+        description="Plot fio json+ wide-format latency CSV."
+    )
+    parser.add_argument("csv", help="CSV from fio_jsonplus_clat2csv")
+    parser.add_argument("--metric", default=None,
+                       help="Metric to plot (e.g. read_clat). "
+                            "If omitted, the metric with the most I/Os is used.")
+    parser.add_argument("--all-metrics", action="store_true",
+                       help="Plot every metric that has data")
+    parser.add_argument("--unit", choices=["ns", "us", "ms", "s"], default="us",
+                       help="X-axis latency unit (default: us)")
+    parser.add_argument("--log-y", action="store_true")
+    parser.add_argument("--log-x", action="store_true")
+    parser.add_argument("--cdf", action="store_true",
+                       help="Add a CDF subplot below the histogram")
+    parser.add_argument("--cdf-only", action="store_true")
+    parser.add_argument("--normalize", action="store_true",
+                       help="Normalize histogram to a probability density")
+    parser.add_argument("--percentiles", default="50,90,99,99.9,99.99")
+    parser.add_argument("--no-percentiles", action="store_true")
+    parser.add_argument("--output", "-o", default=None)
+    parser.add_argument("--dpi", type=int, default=150)
+    parser.add_argument("--title", default="")
+    parser.add_argument("--show", action="store_true")
+    args = parser.parse_args()
+
+    if not Path(args.csv).is_file():
+        sys.exit(f"error: file not found: {args.csv}")
+
+    df = load_latency_csv(args.csv)
+    if df.empty:
+        sys.exit("error: no latency data in CSV")
+
+    available = sorted(df["metric"].unique())
+    print("Metrics with data:", ", ".join(available))
+
+    if args.all_metrics:
+        metrics = available
+    elif args.metric:
+        if args.metric not in available:
+            sys.exit(f"error: metric {args.metric!r} not in {available}")
+        metrics = [args.metric]
+    else:
+        totals = df.groupby("metric")["count"].sum().sort_values(ascending=False)
+        metrics = [totals.index[0]]
+        print(f"Auto-selected metric: {metrics[0]}")
+
+    output = args.output or f"{Path(args.csv).stem}_latency.png"
+    generate_plot(
+        df=df,
+        metrics=metrics,
+        output_path=Path(output),
+        unit=args.unit,
+        log_y=args.log_y,
+        log_x=args.log_x,
+        normalize=args.normalize,
+        show_cdf=args.cdf,
+        cdf_only=args.cdf_only,
+        title=args.title,
+        dpi=args.dpi,
+    )
+    print(f"Saved plot to {output}")
+
+    if not args.no_percentiles:
+        try:
+            pcts = [float(x) for x in args.percentiles.split(",") if x.strip()]
+        except ValueError:
+            sys.exit("error: --percentiles must be comma-separated numbers")
+        for m in metrics:
+            print(percentile_table(df, m, args.unit, pcts))
+
+    if args.show:
+        import matplotlib.pyplot as plt
+        plt.show()
+
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/src/disk_checker/__main__.py b/src/disk_checker/__main__.py
new file mode 100644
index 0000000..6e1b3a5
--- /dev/null
+++ b/src/disk_checker/__main__.py
@@ -0,0 +1,6 @@
+"""Allow running disk-checker as a module: python -m disk_checker"""
+
+from disk_checker.scripts.cli import main
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file