☰

ファイルaからファイルbとその括弧全体を検索して抽出する

tag-icon • tag-icon bash shell-script awk
ファイルaからファイルbとその括弧全体を検索して抽出する

2つのファイルがあり、私がやりたいのはファイルおよび検索番号ファイルb検索された行を含む括弧全体を出力します。それが何と呼ばれているのかはわかりません。

ファイル:

300
302
303

ファイルb:

    [300] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },
    [301] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },
    [302] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },
    [303] = {
        name = "John",
        age = "12",
        address = {
        "Person street address"
        },
        job = "Marketing",
        job a = "some job",
        job b = {
        "His job description"
        },
        car = 0,
        salary = 0
    },

期待される出力:

    [300] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },
    [302] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },
    [303] = {
        name = "John",
        age = "12",
        address = {
        "Person street address"
        },
        job = "Marketing",
        job a = "some job",
        job b = {
        "His job description"
        },
        car = 0,
        salary = 0
    },

awk を使ってみましたが、動作しません。

awk 'NR==FNR{a[$1]=1;next} $1 in a && /\[$1\]/,/^\t\},/ {print}' a b > c

助けてくれてありがとう..

答え1

この特定の形式のパーサーにアクセスできず、レコードが実際に示されているように単純で規則的であると仮定すると、これはすべての UNIX ボックスの任意のシェルで任意の awk を使用して機能します。

$ cat tst.awk
NR == FNR {
    vals["["$1"]"]
    next
}
rec == "" { key = $1 }
{ rec = rec $0 ORS }
/^[[:blank:]]*},$/ {
    if ( key in vals ) {
        printf "%s", rec
    }
    key = rec = ""
}

。

$ awk -f tst.awk file_a file_b
    [300] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },
    [302] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },

答え2

以下のコマンドでテストし、正常に動作しました

指示

for i in `cat filea`; do sed -n '/'$i'/,/^[[:space:]]\{4\}\},$/p' fileb; done

出力

 [300] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },
    [302] = {
        name = "John",
        age = "12",
        address = {""},
        job = "Marketing",
        job a = "some job",
        job b = {"some job"},
        car = 0,
        salary = 0
    },
    [303] = {
        name = "John",
        age = "12",
        address = {
        "Person street address"
        },
        job = "Marketing",
        job a = "some job",
        job b = {
        "His job description"
        },
        car = 0,
        salary = 0
    },

答え3

sed次の行を使用すると役立つかもしれません。

while read n; do sed -n -e "/^\[$n\]/,/^},$/p" b.txt; done < a.txt

関連情報