split-gmail.pl: extract bank statements: 把 zimbra mail server 下載回來的 *.eml 檔當中的銀行/信用卡/電費/電信費帳單抓出來、移除密碼，成為單純的 pdf 檔 + split-gmail.pl 把 gmail 的一個 *.mbox 拆成每封信一個檔案

split-gmail.pl

#!/usr/bin/perl -w

# usage: split-gmail.pl test.mbox
# split test.mbox, a file exported from gmail, into one mail per file, named m0001.mbox, m0002.mbox, ...

$fn = sprintf("m%04d.mbox", 1);
open(FH, '>', $fn) or die $!;
$N = 0;
while (<>) {
    # a sample string we are matching against:
    # From 1789598446958704853@xxx Wed Jan 31 09:41:15 +0000 2024^M
    if (/^From\s+\S+\s+\w+\s+\w+\s+\d+\s+[\d:]+\s+\S+\s+\d+\s*$/) {
	close(FH);
	++$N;
	$fn = sprintf("m%04d.mbox", $N);
	open(FH, '>', $fn) or die $!;
	print(FH $_);
    } else {
	print(FH $_);
    }
}
close(FH);

xtbankst.sh

#!/bin/bash
# extract bank statements
# 用法： xtbankst.sh 來源目錄 目的地目錄

src_dir=$(realpath $1)
dst_dir=$(realpath $2)
echo "$src_dir => $dst_dir"

echo -n "Password: "
read -s PASSWORD
echo ''

shopt -s nullglob
# https://www.endpointdev.com/blog/2016/12/bash-loop-wildcards-nullglob-failglob/

all_files=$(echo \
    "$src_dir/*-玩很大銀行*月份『綜合對帳單』.eml" \
    "$src_dir/*-真好貸銀行*月信用卡電子帳單.eml" \
    "$src_dir/*-抄底高手證券電子綜合月對帳單\(20*_*\).eml" \
    "$src_dir/*-台電e-Bill*電費繳費憑證*.eml" \
    "$src_dir/*-中華電信*年*月電信費用通知單*.eml" \
    "$src_dir/*.mbox" \
)

mkdir -p $dst_dir
pushd $dst_dir
for MF in $all_files ; do
    ym=$( echo "$MF" | perl -ne '
        ($y, $m) = /(\d+)年(\d+)月/ or
        ($y, $m) = /\b(20\d\d)_(\d\d)\b/ or
        ($y, $m) = (2099,99);
        $y = $y>2000 ? $y-2000 : $y-89;
	printf("%02d%02d\n", $y, $m)' )
    echo "[$ym] [$MF]" 
    rm -rf xtbankst-tmp/ ; mkdir -p xtbankst-tmp/
    pushd xtbankst-tmp/
    munpack -qf "$MF"
    for att in *.pdf ; do
	qpdf --decrypt --password="$PASSWORD" "$att" "$dst_dir/$ym-$att"
    done
    for att in *X[Uu][Tt][Ff]-8XBX*X= ; do 
	dan=$(echo $att | sed 's/=X[Uu][Tt][Ff]-8XBX//; s/X=//' | base64 -d)
	# decoded attachment name
	if [[ "$dan" =~ \.pdf$ ]]; then
	    qpdf --decrypt --password="$PASSWORD" "$att" "$dst_dir/$ym-$dan"
	fi
	rm $att
    done
    popd
    rm -rf xtbankst-tmp/
done
popd