Add compare-multi command for N-file comparison

src/vmstat_visualizer/cli/__main__.py

@@ -4,6 +4,7 @@
 import click
 
 import vmstat_visualizer.cli.visualizer as viz
+from vmstat_visualizer.cli.multi_compare import compare_multi
 
 
 @click.group(no_args_is_help=True)
@@ -17,6 +18,7 @@ def main():
     "main is the entrypoint of vmstat-visualizer"
     cli.add_command(viz.visualize)
     cli.add_command(viz.compare)
+    cli.add_command(compare_multi)
     cli(prog_name="vmstat-visualizer")
 
 

src/vmstat_visualizer/cli/multi_compare.py

@@ -0,0 +1,49 @@
+"""CLI command for multi-file comparison."""
+
+import click
+from vmstat_visualizer.parser.parser import Parser
+from vmstat_visualizer.multi_compare import plot_multi_compare
+
+
+@click.command("compare-multi", no_args_is_help=True)
+@click.argument("files", nargs=-1, type=click.Path(exists=True), required=True)
+@click.option(
+    "-m",
+    "--metric",
+    type=click.Choice(["cpu", "memory", "system_load", "swap", "io", "all"],
+                       case_sensitive=False),
+    required=True,
+    help="Metric to compare.",
+)
+@click.option(
+    "-o",
+    "--output-prefix",
+    default="vmstat",
+    help="Output filename prefix.",
+)
+@click.option(
+    "-e",
+    "--output-extension",
+    default="png",
+    help="Output format: png or svg.",
+)
+def compare_multi(files, metric, output_prefix, output_extension):
+    """Compare N vmstat log files on a given metric.
+
+    Pass two or more log files as arguments.
+    """
+    if len(files) < 2:
+        raise click.UsageError("At least 2 files are required for comparison.")
+
+    parsers = []
+    for f in files:
+        p = Parser(f)
+        p.parse()
+        print(f">>> Parsed {len(p.timeseries)} entries from {f}")
+        parsers.append(p)
+
+    plot_multi_compare(
+        parsers, metric,
+        output_file_prefix=output_prefix,
+        output_format=output_extension,
+    )

src/vmstat_visualizer/multi_compare.py

@@ -0,0 +1,136 @@
+"""Multi-file comparison plotting for N vmstat log files."""
+
+import datetime
+import os
+import time
+import matplotlib.pyplot as plt
+
+
+LINE_STYLES = ['-', '--', '-.', ':', (0, (3, 1, 1, 1))]
+COLORS = ['#e74c3c', '#3498db', '#2ecc71', '#f39c12', '#9b59b6',
+          '#1abc9c', '#e67e22', '#34495e', '#16a085', '#c0392b']
+
+
+def _style_for(idx):
+    """Return (color, linestyle) for file index."""
+    color = COLORS[idx % len(COLORS)]
+    style = LINE_STYLES[idx % len(LINE_STYLES)]
+    return color, style
+
+
+def plot_multi_compare(parsers, metric, output_file_prefix='vmstat',
+                       output_format='png'):
+    """Compare N parsed files on a given metric.
+
+    Args:
+        parsers: list of Parser instances (already parsed)
+        metric: one of 'cpu', 'memory', 'system_load', 'swap', 'io', 'all'
+        output_file_prefix: filename prefix
+        output_format: 'png' or 'svg'
+    """
+    if metric == 'all':
+        for m in ['cpu', 'memory', 'system_load', 'swap', 'io']:
+            plot_multi_compare(parsers, m, output_file_prefix, output_format)
+        return
+
+    from vmstat_visualizer.parser.parser import Parser
+
+    all_metrics = []
+    filenames = []
+    for p in parsers:
+        pm = Parser.PlotMetrics(p.timeseries, relative_time=True)
+        all_metrics.append(pm)
+        filenames.append(os.path.basename(p.filename))
+
+    max_len = max(len(pm.t) for pm in all_metrics)
+    t = list(range(max_len))
+
+    now_unix = int(time.mktime(
+        datetime.datetime.now().replace(second=0, microsecond=0).timetuple()
+    ))
+
+    metric_configs = {
+        'cpu': {
+            'title': 'CPU Usage (%) - Multi Comparison',
+            'ylabel': 'Percent',
+            'series': [
+                ('User CPU', 'user_cpu_percent'),
+                ('System CPU', 'system_cpu_percent'),
+                ('Idle CPU', 'idle_cpu_percent'),
+                ('Wait CPU', 'wait_cpu_percent'),
+                ('Steal CPU', 'steal_cpu_percent'),
+            ]
+        },
+        'memory': {
+            'title': 'Memory Usage (KB) - Multi Comparison',
+            'ylabel': 'Memory (KB)',
+            'series': [
+                ('Free', 'free_memory_kb'),
+                ('Inactive', 'inactive_memory_kb'),
+                ('Active', 'active_memory_kb'),
+                ('Swapped', 'swapped_memory_kb'),
+            ]
+        },
+        'system_load': {
+            'title': 'System Load - Multi Comparison',
+            'ylabel': 'Processes',
+            'series': [
+                ('Run Queue', 'run_queue'),
+                ('Blocked', 'blocked_processes'),
+            ]
+        },
+        'swap': {
+            'title': 'Swap Usage - Multi Comparison',
+            'ylabel': 'KB',
+            'series': [
+                ('Swap In', 'swap_in_kb'),
+                ('Swap Out', 'swap_out_kb'),
+                ('Swapped', 'swapped_memory_kb'),
+            ]
+        },
+        'io': {
+            'title': 'Block I/O - Multi Comparison',
+            'ylabel': 'Blocks',
+            'series': [
+                ('Blocks In', 'blocks_in'),
+                ('Blocks Out', 'blocks_out'),
+            ]
+        },
+    }
+
+    config = metric_configs[metric]
+    fig, ax = plt.subplots(figsize=(12, 5))
+
+    for file_idx, (pm, fname) in enumerate(zip(all_metrics, filenames)):
+        color, style = _style_for(file_idx)
+        for series_name, attr in config['series']:
+            data = getattr(pm, attr, [])
+            if not data or all(v is None for v in data):
+                continue
+            values = [int(v) for v in data]
+            ax.plot(
+                t[:len(values)], values,
+                label=f'{series_name} ({fname})',
+                color=color, linestyle=style, linewidth=1.5,
+                alpha=0.8,
+            )
+
+    ax.set_title(config['title'])
+    ax.set_xlabel('Relative Time (seconds)')
+    ax.set_ylabel(config['ylabel'])
+    ax.set_ylim(bottom=0)
+    ax.legend(loc='best', fontsize=7, ncol=2)
+    plt.tight_layout()
+
+    starts = []
+    for p in parsers:
+        if p.timeseries:
+            starts.append(f"{os.path.basename(p.filename)}: {p.timeseries[0].time}")
+    start_text = " | ".join(starts)
+    plt.figtext(0.99, 0.01, f"Start times: {start_text}",
+                horizontalalignment='right', fontsize=6, color='gray')
+
+    outfile = f'{output_file_prefix}_{metric}_multi_{now_unix}.{output_format}'
+    plt.savefig(outfile, bbox_inches='tight')
+    plt.close()
+    print(f">>> Saved {outfile}")