{
  "id": "linux-files-and-text",
  "title": "Extract the Signal From a Log File",
  "category": "Linux Fundamentals",
  "isolation_type": "shared_namespace",
  "difficulty": "Easy",
  "type": "task",
  "weight": 4,
  "description": "## Extract the Signal From a Log File\n\nA real application log (`/srv/text-lab/app.log`) has been generated for you, mixing `INFO` and `ERROR` lines. Reading raw logs line-by-line doesn't scale — filtering does.\n\n**Your task:**\n\n1. Use `grep` to extract only the `ERROR` lines from `/srv/text-lab/app.log`\n2. Redirect the result into a new file: `/srv/text-lab/errors.txt`\n3. Count how many ERROR lines there were and save the count to `/srv/text-lab/error-count.txt`",
  "hints": [
    {
      "title": "Filtering and redirecting",
      "body": "`grep PATTERN file > newfile` writes only matching lines to a new file. Pipe to `wc -l` to count them.",
      "command": "grep ERROR /srv/text-lab/app.log > /srv/text-lab/errors.txt\ngrep -c ERROR /srv/text-lab/app.log > /srv/text-lab/error-count.txt"
    }
  ],
  "setup_commands": [
    { "command": "mkdir -p /srv/text-lab && printf 'INFO: boot ok\\nERROR: disk full\\nINFO: retry\\nERROR: timeout\\nINFO: done\\n' > /srv/text-lab/app.log" }
  ],
  "validation": {
    "commands": [
      { "description": "errors.txt contains only ERROR lines", "command": "grep -vc ERROR /srv/text-lab/errors.txt || true", "expected_output": "0", "match": "exact" },
      { "description": "errors.txt has exactly the 2 real ERROR lines", "command": "wc -l < /srv/text-lab/errors.txt | tr -d ' '", "expected_output": "2", "match": "exact" },
      { "description": "error-count.txt records the count", "command": "cat /srv/text-lab/error-count.txt", "expected_output": "2", "match": "exact" }
    ]
  },
  "default_namespace": "default",
  "teardown_commands": [
    { "command": "rm -rf /srv/text-lab" }
  ]
}