analysis/capt_validation_schema.py

from __future__ import annotations

import csv
from dataclasses import asdict, dataclass, field
from pathlib import Path
from typing import Dict, List, Optional


BOOKS_MASTER_FIELDS = [
    "book_id",
    "asin",
    "isbn13",
    "title",
    "author_name",
    "publisher",
    "publisher_tier",
    "publication_date",
    "format",
    "list_price",
    "genre_primary",
    "genre_secondary",
    "sample_group",
    "author_central_present",
    "author_central_links_count",
    "youtube_link_present",
    "goodreads_presence",
    "social_presence_score",
    "author_best_known_bsr",
    "author_prior_top100_title_count",
    "nyt_list_any",
    "nyt_list_count",
]


BOOK_OBSERVATION_FIELDS = [
    "book_id",
    "observation_date",
    "source_url",
    "days_since_publication",
    "amazon_bsr_overall",
    "amazon_bsr_primary_category",
    "amazon_bsr_secondary_category",
    "best_visible_category_bsr",
    "primary_category_name",
    "secondary_category_name",
    "category_count",
    "category_top10_count",
    "category_top100_count",
    "category_top1000_count",
    "launch_window_30d_flag",
    "hot_new_releases_flag",
    "hot_new_releases_position",
    "movers_shakers_flag",
    "movers_shakers_position",
    "rating_average",
    "rating_count",
    "review_count",
    "also_bought_visible_flag",
    "also_bought_count",
    "also_bought_bestseller_count",
    "also_bought_shared_category_count",
    "review_velocity_7d",
    "review_velocity_30d",
    "rank_change_24h",
    "top100_flag",
    "threshold_50_reviews_flag",
    "threshold_event_flag",
    "sustained_top100_60d_flag",
    "ar_proxy_score",
    "csd_proxy_score",
    "cpa_proxy_score",
    "arm_observed_score",
]


NYT_WEEKLY_FIELDS = [
    "published_date",
    "list_name",
    "list_name_encoded",
    "rank",
    "rank_last_week",
    "weeks_on_list",
    "title",
    "author",
    "publisher",
    "primary_isbn13",
    "primary_isbn10",
    "amazon_product_url",
    "book_image",
    "book_uri",
]


NYT_HISTORY_FIELDS = [
    "query_value",
    "query_type",
    "title",
    "author",
    "publisher",
    "description",
    "contributor",
    "contributor_note",
    "price",
    "age_group",
    "primary_isbn13",
    "primary_isbn10",
    "isbn13",
    "isbn10",
    "history_list_name",
    "history_display_name",
    "bestsellers_date",
    "published_date",
    "rank",
    "rank_last_week",
    "weeks_on_list",
    "asterisk",
    "dagger",
]


AMAZON_OBSERVATION_INPUT_FIELDS = [
    "book_id",
    "observation_date",
    "source_url",
    "asin",
    "title",
    "author_name",
    "days_since_publication",
    "amazon_bsr_overall",
    "amazon_bsr_primary_category",
    "amazon_bsr_secondary_category",
    "primary_category_name",
    "secondary_category_name",
    "category_count",
    "category_top10_count",
    "category_top100_count",
    "category_top1000_count",
    "hot_new_releases_flag",
    "hot_new_releases_position",
    "movers_shakers_flag",
    "movers_shakers_position",
    "rating_average",
    "rating_count",
    "review_count",
    "also_bought_visible_flag",
    "also_bought_count",
    "also_bought_bestseller_count",
    "also_bought_shared_category_count",
    "notes",
]


@dataclass
class BookMasterRecord:
    book_id: str
    asin: str = ""
    isbn13: str = ""
    title: str = ""
    author_name: str = ""
    publisher: str = ""
    publisher_tier: str = ""
    publication_date: str = ""
    format: str = ""
    list_price: str = ""
    genre_primary: str = ""
    genre_secondary: str = ""
    sample_group: str = ""
    author_central_present: str = ""
    author_central_links_count: str = ""
    youtube_link_present: str = ""
    goodreads_presence: str = ""
    social_presence_score: str = ""
    author_best_known_bsr: str = ""
    author_prior_top100_title_count: str = ""
    nyt_list_any: str = ""
    nyt_list_count: str = ""

    def to_row(self) -> Dict[str, str]:
        return {field_name: str(asdict(self).get(field_name, "")) for field_name in BOOKS_MASTER_FIELDS}


@dataclass
class BookObservationRecord:
    book_id: str
    observation_date: str
    source_url: str = ""
    days_since_publication: str = ""
    amazon_bsr_overall: str = ""
    amazon_bsr_primary_category: str = ""
    amazon_bsr_secondary_category: str = ""
    best_visible_category_bsr: str = ""
    primary_category_name: str = ""
    secondary_category_name: str = ""
    category_count: str = ""
    category_top10_count: str = ""
    category_top100_count: str = ""
    category_top1000_count: str = ""
    launch_window_30d_flag: str = ""
    hot_new_releases_flag: str = ""
    hot_new_releases_position: str = ""
    movers_shakers_flag: str = ""
    movers_shakers_position: str = ""
    rating_average: str = ""
    rating_count: str = ""
    review_count: str = ""
    also_bought_visible_flag: str = ""
    also_bought_count: str = ""
    also_bought_bestseller_count: str = ""
    also_bought_shared_category_count: str = ""
    review_velocity_7d: str = ""
    review_velocity_30d: str = ""
    rank_change_24h: str = ""
    top100_flag: str = ""
    threshold_50_reviews_flag: str = ""
    threshold_event_flag: str = ""
    sustained_top100_60d_flag: str = ""
    ar_proxy_score: str = ""
    csd_proxy_score: str = ""
    cpa_proxy_score: str = ""
    arm_observed_score: str = ""

    def to_row(self) -> Dict[str, str]:
        return {field_name: str(asdict(self).get(field_name, "")) for field_name in BOOK_OBSERVATION_FIELDS}


@dataclass
class TemplateBundle:
    books_master_path: Path
    observations_path: Path
    nyt_weekly_path: Path
    nyt_history_path: Path
    amazon_input_path: Path
    metadata: Dict[str, str] = field(default_factory=dict)


def ensure_directory(path: str | Path) -> Path:
    directory = Path(path)
    directory.mkdir(parents=True, exist_ok=True)
    return directory


def write_csv_template(path: str | Path, fieldnames: List[str], example_rows: Optional[List[Dict[str, str]]] = None) -> Path:
    target = Path(path)
    ensure_directory(target.parent)
    with target.open("w", newline="", encoding="utf-8") as handle:
        writer = csv.DictWriter(handle, fieldnames=fieldnames)
        writer.writeheader()
        for row in example_rows or []:
            writer.writerow({field_name: row.get(field_name, "") for field_name in fieldnames})
    return target


def create_validation_templates(output_dir: str | Path = "validation_templates") -> TemplateBundle:
    output_path = ensure_directory(output_dir)

    books_master_example = BookMasterRecord(
        book_id="philo_001",
        asin="B0EXAMPLE01",
        isbn13="9780000000001",
        title="Example Philosophy Title",
        author_name="Example Author",
        publisher="Example Press",
        publisher_tier="mid",
        publication_date="2026-01-15",
        format="hardcover",
        list_price="28.00",
        genre_primary="philosophy",
        genre_secondary="self-help",
        sample_group="candidate",
        author_central_present="1",
        author_central_links_count="2",
        youtube_link_present="1",
        goodreads_presence="1",
        social_presence_score="2",
        author_best_known_bsr="41",
        author_prior_top100_title_count="2",
        nyt_list_any="0",
        nyt_list_count="0",
    ).to_row()

    observation_example = BookObservationRecord(
        book_id="philo_001",
        observation_date="2026-04-12",
        source_url="https://www.amazon.com/dp/B0EXAMPLE01",
        days_since_publication="88",
        amazon_bsr_overall="154",
        amazon_bsr_primary_category="12",
        amazon_bsr_secondary_category="47",
        best_visible_category_bsr="12",
        primary_category_name="Philosophy",
        secondary_category_name="Ethics & Morality",
        category_count="2",
        category_top10_count="0",
        category_top100_count="2",
        category_top1000_count="2",
        launch_window_30d_flag="0",
        hot_new_releases_flag="0",
        hot_new_releases_position="",
        movers_shakers_flag="1",
        movers_shakers_position="18",
        rating_average="4.6",
        rating_count="73",
        review_count="73",
        also_bought_visible_flag="1",
        also_bought_count="12",
        also_bought_bestseller_count="4",
        also_bought_shared_category_count="7",
        review_velocity_7d="5",
        review_velocity_30d="19",
        rank_change_24h="23",
        top100_flag="0",
        threshold_50_reviews_flag="1",
        threshold_event_flag="0",
        sustained_top100_60d_flag="",
        ar_proxy_score="0.61",
        csd_proxy_score="0.48",
        cpa_proxy_score="0.67",
        arm_observed_score="0.35",
    ).to_row()

    nyt_example = {
        "published_date": "2026-04-12",
        "list_name": "Hardcover Nonfiction",
        "list_name_encoded": "hardcover-nonfiction",
        "rank": "8",
        "rank_last_week": "0",
        "weeks_on_list": "1",
        "title": "Example Philosophy Title",
        "author": "Example Author",
        "publisher": "Example Press",
        "primary_isbn13": "9780000000001",
        "primary_isbn10": "0000000000",
        "amazon_product_url": "https://www.amazon.com/dp/B0EXAMPLE01",
        "book_image": "https://example.com/cover.jpg",
        "book_uri": "nyt://books/example",
    }

    nyt_history_example = {
        "query_value": "9780000000001",
        "query_type": "isbn",
        "title": "Example Philosophy Title",
        "author": "Example Author",
        "publisher": "Example Press",
        "description": "Example retrospective history row.",
        "contributor": "by Example Author",
        "contributor_note": "",
        "price": "28.00",
        "age_group": "",
        "primary_isbn13": "9780000000001",
        "primary_isbn10": "0000000000",
        "isbn13": "9780000000001",
        "isbn10": "0000000000",
        "history_list_name": "hardcover-nonfiction",
        "history_display_name": "Hardcover Nonfiction",
        "bestsellers_date": "2026-04-05",
        "published_date": "2026-04-12",
        "rank": "8",
        "rank_last_week": "10",
        "weeks_on_list": "2",
        "asterisk": "0",
        "dagger": "0",
    }

    amazon_input_example = {
        "book_id": "philo_001",
        "observation_date": "2026-04-12",
        "source_url": "https://www.amazon.com/dp/B0EXAMPLE01",
        "asin": "B0EXAMPLE01",
        "title": "Example Philosophy Title",
        "author_name": "Example Author",
        "days_since_publication": "88",
        "amazon_bsr_overall": "154",
        "amazon_bsr_primary_category": "12",
        "amazon_bsr_secondary_category": "47",
        "primary_category_name": "Philosophy",
        "secondary_category_name": "Ethics & Morality",
        "category_count": "2",
        "category_top10_count": "0",
        "category_top100_count": "2",
        "category_top1000_count": "2",
        "hot_new_releases_flag": "0",
        "hot_new_releases_position": "",
        "movers_shakers_flag": "1",
        "movers_shakers_position": "18",
        "rating_average": "4.6",
        "rating_count": "73",
        "review_count": "73",
        "also_bought_visible_flag": "1",
        "also_bought_count": "12",
        "also_bought_bestseller_count": "4",
        "also_bought_shared_category_count": "7",
        "notes": "Manual observation example row.",
    }

    books_master_path = write_csv_template(output_path / "books_master_template.csv", BOOKS_MASTER_FIELDS, [books_master_example])
    observations_path = write_csv_template(output_path / "book_observations_template.csv", BOOK_OBSERVATION_FIELDS, [observation_example])
    nyt_weekly_path = write_csv_template(output_path / "nyt_weekly_lists_template.csv", NYT_WEEKLY_FIELDS, [nyt_example])
    nyt_history_path = write_csv_template(output_path / "nyt_bestseller_history_template.csv", NYT_HISTORY_FIELDS, [nyt_history_example])
    amazon_input_path = write_csv_template(output_path / "amazon_observation_input_template.csv", AMAZON_OBSERVATION_INPUT_FIELDS, [amazon_input_example])

    return TemplateBundle(
        books_master_path=books_master_path,
        observations_path=observations_path,
        nyt_weekly_path=nyt_weekly_path,
        nyt_history_path=nyt_history_path,
        amazon_input_path=amazon_input_path,
        metadata={
            "books_master_fields": str(len(BOOKS_MASTER_FIELDS)),
            "book_observation_fields": str(len(BOOK_OBSERVATION_FIELDS)),
            "nyt_weekly_fields": str(len(NYT_WEEKLY_FIELDS)),
            "nyt_history_fields": str(len(NYT_HISTORY_FIELDS)),
            "amazon_input_fields": str(len(AMAZON_OBSERVATION_INPUT_FIELDS)),
        },
    )


if __name__ == "__main__":
    bundle = create_validation_templates()
    print("Created validation templates:")
    print(f"- books_master: {bundle.books_master_path}")
    print(f"- book_observations: {bundle.observations_path}")
    print(f"- nyt_weekly_lists: {bundle.nyt_weekly_path}")
    print(f"- nyt_bestseller_history: {bundle.nyt_history_path}")
    print(f"- amazon_observation_input: {bundle.amazon_input_path}")