こんにちは。
複数ページ構成のweb記事に対して、その url の次のページの url を作るシェルスクリプトを作りました。ただし万全ではないと思うのでご注意願います。
$ ./next_page.sh https://gendai.media/articles/-/163037
https://gendai.media/articles/-/163037?page=2
next_page.sh
#!/bin/sh
url="$1"
prog_name=$(basename "$0")
# functions
print_usage_f() {
cat << EOS
usage:
$prog_name https://gendai.media/articles/-/163037
$prog_name https://ascii.jp/elem/000/004/366/4366125
$prog_name https://natgeo.nikkeibp.co.jp/atcl/web/15/403964/011400177
$prog_name https://www.newsweekjapan.jp/kimura/2026/01/post-343.php
EOS
}
page_num_from_param_f() {
url="$1"
n=$(echo "$url" | grep -oE '[?&](page|P)=[0-9]+' | grep -oE '[0-9]+')
[ -n "$n" ] || n=1
echo "$n"
}
page_num_from_file_f() {
url="$1"
n=1
echo "$url" | grep -oq '_' && n=$(echo "$url" | sed 's/.*_\([0-9]*\)\..*/\1/')
echo "$n"
}
url_from_file_f() {
url="$1"
page_num=$2
file="${url##*/}"
base="${file%%.*}"; base="${base%%_*}"
ext="${file##*.}"
echo "${url%/*}/${base}_${page_num}.$ext"
}
page_num_from_subdir_f() {
url="$1"
n=$(echo "${url%/}" | grep -oE '[0-9]+$')
{ [ -z "$n" ] || [ "$n" -gt 12 ]; } && n=1
echo "$n"
}
url_from_subdir_f() {
url="$1"
page_num=$2
base="${url%/}"
n=$(echo "$base" | grep -oE '[0-9]+$')
[ "$n" -le 12 ] && base=${base%/*}
echo "$base/${page_num}/"
}
page_num_f() {
url="$1"
case "$url" in
*newsweekjapan.jp*)
page_num_from_file_f "$url" ;;
*ascii.jp*)
page_num_from_subdir_f "$url" ;;
*nikkeibp.co.jp*)
page_num_from_param_f "$url" ;;
*gendai.media*)
page_num_from_param_f "$url" ;;
esac
}
url_f() {
url="$1"
page_num=$2
case "$url" in
*newsweekjapan.jp*)
url_from_file_f "$url" "$page_num" ;;
*ascii.jp*)
url_from_subdir_f "$url" "$page_num" ;;
*nikkeibp.co.jp*)
echo "${url%%\?*}?P=${page_num}" ;;
*gendai.media*)
echo "${url%%\?*}?page=${page_num}" ;;
esac
}
# usage
case "$1" in -h|--help) print_usage_f; exit;; esac
# main
n=$(page_num_f "$url")
url_f "$url" $((n + 1))