Web Profiler

A utility toolkit for profiling, checking, and analyzing web request execution. Log profiling runs, analyze performance data, generate reports, and export results — all from the command line.

Commands

All commands accept optional arguments. Without arguments, they display recent entries from their log.

Data Storage

All data is stored locally in ~/.local/share/web-profiler/. Each command maintains its own .log file with timestamped entries in YYYY-MM-DD HH:MM|value format. A unified history.log tracks all operations across commands.

Export formats supported:

• JSON — Array of objects with type, time, and value fields
• CSV — Standard comma-separated with type,time,value header
• TXT — Human-readable grouped by command type

Requirements

• Bash 4.0+ with set -euo pipefail (strict mode)
• Standard Unix utilities: date, wc, du, grep, tail, sed, cat
• No external dependencies — runs on any POSIX-compliant system

When to Use

1. Profiling web request performance — Log and review timing, memory, and query data for HTTP requests
2. Debugging slow routes — Use analyze and compare to record performance investigations
3. Tracking profiling history — Keep a timestamped log of all profiling runs for trend analysis
4. Generating performance reports — Export accumulated profiling data to JSON/CSV for dashboards
5. Batch profiling operations — Profile multiple endpoints in one session and review results later

Examples

# Run a profiling task
web-profiler run "GET /api/users — 342ms, 12MB memory"

# Analyze query performance
web-profiler analyze "SELECT * FROM orders — 89ms, 1.2k rows"

# Compare two profiling runs
web-profiler compare "v2.1 vs v2.2: 15% latency reduction"

# Search for previous profiling entries
web-profiler search "memory"

# Export all profiling data to CSV
web-profiler export csv

# View summary statistics
web-profiler stats

How It Works

Web Profiler stores all data locally in ~/.local/share/web-profiler/. Each command creates a dedicated log file (e.g., run.log, analyze.log, report.log). Every entry is timestamped and appended, providing a full audit trail. The history.log file aggregates activity across all commands for unified tracking.

When called without arguments, each command displays its most recent 20 entries, making it easy to review past profiling work without manually inspecting log files.

Output

All output goes to stdout. Redirect to a file with:

web-profiler stats > report.txt
web-profiler export json  # writes to ~/.local/share/web-profiler/export.json

Powered by BytesAgain | bytesagain.com | hello@bytesagain.com