{
  "tasks": [
    {
      "id": "train_str_normalize_strip_bug",
      "split": "train",
      "title": "String normalization: missing strip",
      "instruction": "The normalize_string function does not remove leading and trailing whitespace. It should return a lowercase, whitespace-trimmed string.",
      "files": {
        "utils/string_utils.py": "def normalize_string(s):\n    return s.lower()",
        "tests/test_string_utils.py": "from utils.string_utils import normalize_string\n\ndef test_normalize_string():\n    assert normalize_string(' Test ') == 'test'\n    assert normalize_string('\\tTabbed\\n') == 'tabbed'\n"
      },
      "checks": [
        {
          "type": "contains",
          "path": "utils/string_utils.py",
          "text": ".strip()"
        },
        {
          "type": "regex",
          "path": "utils/string_utils.py",
          "pattern": "return\\s+s\\.lower\\(\\)\\.strip\\(\\)"
        }
      ],
      "failure_messages": [
        "normalize_string did not remove whitespace",
        "normalize_string does not use both lower and strip"
      ],
      "canonical_solution": "Call strip() after lower() in normalize_string."
    },
    {
      "id": "train_slugify_spaces_to_hyphen",
      "split": "train",
      "title": "Slugify: spaces not replaced with hyphens",
      "instruction": "The slugify function does not replace spaces with hyphens. It should return a lowercase, hyphen-separated string with no spaces.",
      "files": {
        "core/slug.py": "def slugify(text):\n    return text.lower()",
        "tests/test_slug.py": "from core.slug import slugify\n\ndef test_slugify():\n    assert slugify('Hello World') == 'hello-world'\n    assert slugify('Python Slug') == 'python-slug'\n"
      },
      "checks": [
        {
          "type": "contains",
          "path": "core/slug.py",
          "text": ".replace(' ', '-')"
        },
        {
          "type": "regex",
          "path": "core/slug.py",
          "pattern": "return\\s+text\\.lower\\(\\)\\.replace\\(' ', '- '\\)"
        }
      ],
      "failure_messages": [
        "slugify did not convert spaces to hyphens",
        "slugify does not replace spaces using .replace"
      ],
      "canonical_solution": "Add .replace(' ', '-') after lower() in slugify."
    },
    {
      "id": "train_label_parse_extra_whitespace",
      "split": "train",
      "title": "Label parser: extra whitespace not handled",
      "instruction": "The parse_labels function splits labels by comma but does not strip whitespace from each label. It should return a list of trimmed labels.",
      "files": {
        "label_parser.py": "def parse_labels(s):\n    return s.split(',')",
        "tests/test_label_parser.py": "from label_parser import parse_labels\n\ndef test_parse_labels():\n    assert parse_labels('a, b, c') == ['a', 'b', 'c']\n    assert parse_labels('foo,bar') == ['foo', 'bar']\n"
      },
      "checks": [
        {
          "type": "contains",
          "path": "label_parser.py",
          "text": "label.strip()"
        },
        {
          "type": "regex",
          "path": "label_parser.py",
          "pattern": "\\[label\\.strip\\(\\)\\s*for\\s*label\\s*in\\s*s\\.split\\("
        }
      ],
      "failure_messages": [
        "parse_labels returned labels with spaces",
        "parse_labels does not strip whitespace from labels"
      ],
      "canonical_solution": "Use a list comprehension to strip whitespace from each split label."
    },
    {
      "id": "train_logging_missing_logger_name",
      "split": "train",
      "title": "Logging: missing logger name",
      "instruction": "The get_logger function creates a logger but does not set its name as the argument. Returned loggers have the default root logger name instead of the module name.",
      "files": {
        "logging_utils.py": "import logging\n\ndef get_logger(name):\n    logger = logging.getLogger()\n    return logger",
        "tests/test_logging_utils.py": "from logging_utils import get_logger\n\ndef test_get_logger():\n    logger = get_logger('my_mod')\n    assert logger.name == 'my_mod'\n"
      },
      "checks": [
        {
          "type": "contains",
          "path": "logging_utils.py",
          "text": "logging.getLogger(name)"
        },
        {
          "type": "not_contains",
          "path": "logging_utils.py",
          "text": "logging.getLogger()"
        }
      ],
      "failure_messages": [
        "get_logger did not set the logger name",
        "get_logger should not use getLogger() without arguments"
      ],
      "canonical_solution": "Pass the name to logging.getLogger in get_logger."
    },
    {
      "id": "train_metric_name_dot_instead_of_underscore",
      "split": "train",
      "title": "Metric name: uses dot instead of underscore",
      "instruction": "The format_metric_name function joins metric parts with a dot. It should join them with an underscore instead.",
      "files": {
        "metrics/naming.py": "def format_metric_name(*args):\n    return '.'.join(args)",
        "tests/test_naming.py": "from metrics.naming import format_metric_name\n\ndef test_metric_name():\n    assert format_metric_name('accuracy', 'top1') == 'accuracy_top1'\n    assert format_metric_name('loss', 'test') == 'loss_test'\n"
      },
      "checks": [
        {
          "type": "contains",
          "path": "metrics/naming.py",
          "text": "'_'"
        },
        {
          "type": "regex",
          "path": "metrics/naming.py",
          "pattern": "return\\s*'_'.join\\("
        }
      ],
      "failure_messages": [
        "format_metric_name did not use underscore",
        "format_metric_name does not join with '_'"
      ],
      "canonical_solution": "Change the join separator from '.' to '_' in format_metric_name."
    },
    {
      "id": "train_score_parser_non_int_bug",
      "split": "train",
      "title": "Score parser: does not handle non-integer input",
      "instruction": "The parse_score function crashes when the input is not a valid integer string. It should return None if parsing fails.",
      "files": {
        "core/score_parser.py": "def parse_score(s):\n    return int(s)",
        "tests/test_score_parser.py": "from core.score_parser import parse_score\n\ndef test_parse_score():\n    assert parse_score('3') == 3\n    assert parse_score('abc') is None\n"
      },
      "checks": [
        {
          "type": "contains",
          "path": "core/score_parser.py",
          "text": "try:"
        },
        {
          "type": "contains",
          "path": "core/score_parser.py",
          "text": "except"
        },
        {
          "type": "regex",
          "path": "core/score_parser.py",
          "pattern": "return\\s+None"
        }
      ],
      "failure_messages": [
        "parse_score crashed for non-integer input",
        "parse_score does not handle ValueError",
        "parse_score did not return None for bad input"
      ],
      "canonical_solution": "Wrap int(s) in try/except and return None if ValueError is raised."
    },
    {
      "id": "eval_pathlib_write_missing_parents",
      "split": "eval",
      "title": "Pathlib file write: parent directories not created",
      "instruction": "The write_to_file function writes a file using Path, but fails if the parent directory does not exist. It should create missing directories before writing.",
      "files": {
        "fs/write.py": "from pathlib import Path\n\ndef write_to_file(path, data):\n    p = Path(path)\n    p.write_text(data)",
        "tests/test_write.py": "from fs.write import write_to_file\nimport tempfile\nimport os\n\ndef test_write_to_file():\n    with tempfile.TemporaryDirectory() as d:\n        f = os.path.join(d, 'a/b/c.txt')\n        write_to_file(f, 'hello')\n        assert open(f).read() == 'hello'\n"
      },
      "checks": [
        {
          "type": "contains",
          "path": "fs/write.py",
          "text": "p.parent.mkdir"
        },
        {
          "type": "contains",
          "path": "fs/write.py",
          "text": "exist_ok=True"
        },
        {
          "type": "contains",
          "path": "fs/write.py",
          "text": "parents=True"
        }
      ],
      "failure_messages": [
        "write_to_file did not create parent directories",
        "write_to_file did not use exist_ok=True",
        "write_to_file did not set parents=True"
      ],
      "canonical_solution": "Call p.parent.mkdir(parents=True, exist_ok=True) before writing the file."
    },
    {
      "id": "train_js_package_script_node_typo",
      "split": "train",
      "title": "JS package script: node command typo",
      "instruction": "The run.js script is referenced in package.json scripts with a typo 'nod' instead of 'node'. It should use 'node'.",
      "files": {
        "package.json": "{\n  \"scripts\": {\n    \"start\": \"nod run.js\"\n  }\n}",
        "run.js": "console.log('Hello');",
        "tests/test_script.py": "import json\n\npkg = json.load(open('package.json'))\nassert pkg['scripts']['start'].startswith('node ')"
      },
      "checks": [
        {
          "type": "contains",
          "path": "package.json",
          "text": "\"start\": \"node"
        },
        {
          "type": "not_contains",
          "path": "package.json",
          "text": "nod "
        }
      ],
      "failure_messages": [
        "Start script does not use 'node'",
        "Start script should not contain 'nod '"
      ],
      "canonical_solution": "Fix the script from 'nod run.js' to 'node run.js' in package.json."
    }
  ]
}