Last active
January 11, 2022 06:23
-
-
Save dictcp/cd9e3028b9b873663ff0 to your computer and use it in GitHub Desktop.
Convert Hang Seng Bank eStatement To CSV file
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
<?php | |
// working command | |
// pdftotext -layout eStatementFile-2014-01-30.pdf - | grep -ozP 'TRANSACTION HISTORY(.|\n)*?\n\n\n' | php ~/a.php | fgrep '.' | grep -v "B/F BALANCE" | sed 's/,//g' | sed 's/;/,/g' | |
function trim_value(&$value) | |
{ | |
$value=str_replace(',', '', $value); | |
$value = trim($value); | |
} | |
$fields=[]; | |
$fp = fopen('php://stdin','r') or die ("can't open file"); | |
while ($s = fgets($fp,1024)) { | |
$s=str_pad($s,142); | |
$target = unpack('A11date/A58title/A27deposit/A24withdraw/A21balance',$s); | |
$last_i = count($fields)-1; | |
if ($last_i>1 && empty($target['date'])) { | |
$target['date'] = $fields[$last_i]['date']; | |
} | |
if ( | |
$last_i>1 && | |
(empty($fields[$last_i]['deposit']) && empty($fields[$last_i]['withdraw']) && empty($fields[$last_i]['balance'])) && | |
(empty($target['deposit']) || empty($target['withdraw']) || empty($target['balance'])) | |
) { | |
foreach ($fields[$last_i] as $k => &$v) { | |
if($k=='date')continue; | |
$v.=$target[$k]; | |
} | |
array_walk($fields[$last_i], 'trim_value'); | |
} else { | |
array_walk($target, 'trim_value'); | |
$fields[]=$target; | |
} | |
} | |
array_map(function(array $x){ | |
$x['date']=$x['date'] . " 2014"; | |
$x['date']=strftime('%Y-%m-%d',strtotime($x['date'])); | |
$x['amount']=$x['deposit']-$x['withdraw']; | |
unset($x['deposit']); | |
unset($x['withdraw']); | |
unset($x['balance']); | |
echo implode(';',$x)."\n"; | |
} | |
,$fields); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment