今年食ったサバ缶をまとめるためのPythonスクリプト

1
# prompt: dates.tsvからmackerelの一覧を取得して、mackerels.tsvにある情報を付け加えて、サバ缶のレビュー記事を書きたい。
2
from typing import Union
3
import pandas as pd
4

5
# load data from dates.tsv
6
dates_df = pd.read_csv("dates.tsv", sep="\t")
7

8
# load data from mackrels.tsv
9
mackerels_df = pd.read_csv("mackerels.tsv", sep="\t")
10
mackerels_df = mackerels_df.rename(columns={"No": "mackerel"})
11
print(dates_df)
12
print(mackerels_df)
13

14
# mackerelの情報を取得
15
dates_df = dates_df.query("mackerel != 0")
16
dates_df.drop_duplicates(subset="mackerel")
17
dates_df.dropna(subset=["mackerel"], inplace=True)
18
mackerel_info = pd.merge(dates_df, mackerels_df, on="mackerel", how="left")
19

20
# Create Heading 2
21
def generate_h2(row: Union[pd.Series, pd.DataFrame]) -> str:
22
    """
23
     Generates a level 2 heading for a mackerel review article.
24

25
    :param row: A pandas Series containing the data for the review.
26
                It should have columns "number_for_article", "品名", and "メーカー".
27
    :type row: pd.Series
28
    :raises TypeError: if row is not a pd.Series
29
    :raises KeyError: if row is missing any of the required columns
30
    :returns: The level 2 heading string.
31
    :rtype: str
32

33
    :Example:
34
    >>> generate_h2(pd.Series({'number_for_article': 1, '品名': 'サバ缶', 'メーカー': 'メーカーA'}))
35
    '## 1. サバ缶（メーカーA）'
36
    """
37
    if not isinstance(row, pd.Series):
38
        raise TypeError("row must be a pandas Series")
39
    number_for_article = row["number_for_article"]
40
    name = row["品名"]
41
    brand = row["メーカー"]
42
    return f"## {number_for_article}. {name}（{brand}）"
43

44
# レビューを作成
45
def generate_review(row: Union[pd.Series, pd.DataFrame]) -> str:
46
    """Generates a review article for canned mackerel.
47

48
    :param row: Data used to create the review article.
49
    :type row: Union[pd.Series, pd.DataFrame]
50
    :raises TypeError: if row is not a pd.Series or pd.DataFrame
51
    :returns: The review article.
52
    :rtype: str
53

54
    :Example:
55
    >>> generate_review(pd.Series({'date': '2023-10-27', 'name': 'Canned Mackerel', 'brand': 'Maker'}))
56
    '## 1. Canned Mackerel (Maker)\\n\\n![]()\\n\\nEaten on 2023-10-27, "Canned Mackerel" made by Maker.\\nThe fish is...\\n'
57
    """
58
    if not isinstance(row, (pd.Series, pd.DataFrame)):
59
        raise TypeError("row must be a pandas Series or DataFrame")
60
    date = row["date"]
61
    name = row["品名"]
62
    brand = row["メーカー"]
63
    fish_category = row["魚種"]
64
    flavor = row["風味"]
65
    can_color = row["缶の色"]
66
    delicious = row["美味さ"]
67
    dha = row["DHA_content(mg)"]
68
    epa = row["EPA_content(mg)"]
69
    ease_of_open = "開けやすい" if row["開け易い"] else "開けにくい"
70
    comment = row["コメント"]
71

72
    review = f"{date}に食べた{brand}製の「{name}」です。\n魚は{fish_category}で、{flavor}な風味と{can_color}の缶の色が特徴です。{delicious}で、缶は{ease_of_open}です。\nオメガ3脂肪酸の含有量の目安としては、DHAが{dha}mgで、EPAが{epa}mgでした。\n{comment}"
73
    h2 = row["h2"]
74
    image = "![]()"
75
    review = f"{h2}\n\n{image}\n\n{review}"
76
    return review
77

78

79
mackerel_info["number_for_article"] = range(1, len(mackerel_info.index) + 1)
80
mackerel_info["h2"] = mackerel_info.apply(generate_h2, axis=1)
81
mackerel_info["review"] = mackerel_info.apply(generate_review, axis=1)
82
mackerel_info["md_item"] = mackerel_info.apply(generate_review, axis=1)
83

84
# 結果を出力
85
print("micsellanious info of mackerel_info is the following...")
86
print(mackerel_info[["number_for_article", "date", "name", "h2", "review"]])
87
print("mackerel_info_image is the following...")
88
print(mackerel_info["画像"])