我正在寻找各种（更好的）方法来解析PHP中的结构化文本数据并将这些数据转换为PHP对象图。我在PHP中看到了很多不同的解析器，用于各种基于文本的文件格式，但几乎所有这些解析器似乎都是一些脆弱的正则表达式链。必须有更好的方法！
在这个特定情况下，我希望解析MT940文件（银行账户交易）。但我也遇到了与其他文件格式相同的问题。我总是最终得到一大批正则表达式，这些正则表达式变得复杂，特别是在需要支持不同格式时。 MT940也有这个问题。 MT940不是严格定义的格式，几乎所有银行都使用稍微不同的方言。
那么，你如何设计更强大和可扩展的解析器来处理不同的方言？
这是MT940的一个示例语句，取自this question：
{1:F01AHHBCH110XXX0000000000}{2:I940X N2}{3:{108:XBS/091502}}{4:
:20:XBS/091202/0001
:25:5887/507004-50
:28C:140/1
:60F:C0914CHF7789,
:61:0912021202D36,80NTRFNONREF//0887-1202-29-941
04392579-0 LUTHY + xxx, ZUR
:86:6034?60LUTHY + xxxx, ZUR vom 01.12.09 um 16:28 Karten-Nr. 2232
2579-0
:62F:C091202CHF52,2
:64:C091302CHF52,2
-}
答案 0 :(得分：3)
您可以使用此免费解析器（GPL 2.0）：
http://www.kingsquare.nl/php-mt940
这是另一个：
http://www.butcher.art.pl/en/2010/09/tutoriale/parser-php-mt940-format-wyciagow-bankowych/
希望这会让你放弃重新发明轮子。
那么，您如何设计更健壮且可扩展的解析器 处理不同的方言？
不幸的是，对此没有简单的答案。您必须紧扣并熟悉您希望支持的所有变体。从国王广场页面：
解析器尝试通过它来确定它来自哪个发起银行 文件的前几行然后加载每个银行的引擎。
这将需要大量的经验和研究。幸运的是，他们的代码可以帮助你。
答案 1 :(得分：-1)
class Mt940{
private $controler = null;
private $content = null;
private $transItem = array();
private $transCount = 0;
private $operationItem = array();
private $operationCount = 0;
private $details = array();
public function __construct(\app\library\_PajaxController $controler)
{
$this->controler = $controler;
}
public function load($file_path, $encodingIn='', $encodingOut='')
{
try{
if(!file_exists($file_path)) throw new \app\library\_PajaxException("File not found!", \app\library\_PajaxException::TEXT, $this->controler->getEnv() );
$this->transItem = array();
$this->transCount = 0;
$this->operationItem = array();
$this->operationCount = 0;
$this->content = file_get_contents($file_path);
if($encodingIn != '' && $encodingOut != '' ) $this->content = @iconv($encodingIn, $encodingOut, $this->content);
$this->splitTransfers();
$this->to_details();
$this->parse();
return $this->details;
}catch(\app\library\_PajaxException $ex)
{
throw $ex;
}catch(\Throwable $thr) { throw new \app\library\_PajaxException($thr, \app\library\_PajaxException::THROWABLE, $this->getEnv()); }
}
private function splitTransfers()
{
try{
$contTmp = $this->content;
$offsetStr = 0;
while(true)
{
$intStr20 = strpos( $contTmp,':20:', $offsetStr);
$intStr62F = -1;
$iniStrEnd = -1;
if( is_numeric($intStr20) && $intStr20 >= 0) //found start transaction..
{
//try find end..
$intStr62F = strpos( $contTmp,':62F:', $intStr20);
//echo $intStr20.' -- > '.$intStr62F;
if( is_numeric($intStr62F) && $intStr62F >= 0)
{
$iniStrEnd = strpos( $contTmp,"\r\n", $intStr62F);
$items = substr( $contTmp ,$intStr20, $iniStrEnd - $intStr20 );
$this->transItem[$this->transCount] = $items;
$this->transCount++;
$offsetStr = $iniStrEnd+2;
}
}else break;
}
//print_r($this->transItem);
}catch(\app\library\_PajaxException $ex)
{
throw $ex;
}catch(\Throwable $thr) { throw new \app\library\_PajaxException($thr, \app\library\_PajaxException::THROWABLE, $this->getEnv()); }
}
private function to_details()
{
try{
foreach ($this->transItem as $ix => $trans)
{
// echo $trans."\n\n\n";
$offsetStr = 0;
while(true)
{
$intStr61 = strpos( $trans,':61:', $offsetStr);
$iniStrEnd = -1;
if( is_numeric($intStr61) && $intStr61 >= 0) //found start transaction..
{
//try find end..
$intStrEndStr = strpos( $trans,':61:', $intStr61+4); //nex trans
$intStrEndStr2 = strpos( $trans,':62F:', $intStr61); //nex trans
// echo "\n".$intStr61." -->> ".$iniStrEnd."\n";
// echo "\n".$intStr61." -->> ".$intStrEndStr2."\n";
if( is_numeric($intStrEndStr) && $intStrEndStr >= 0)
{
$item = substr($trans, $intStr61, $intStrEndStr-$intStr61-4);
// echo "\n{".$item."\n}\n";
$this->operationItem[$this->operationCount] = $item;
$this->operationCount++;
$offsetStr = $intStrEndStr-4;
continue;
}else
if( is_numeric($intStrEndStr2) && $intStrEndStr2 >= 0)
{
$item = substr($trans, $intStr61, $intStrEndStr2-$intStr61-4);
// echo "\n3\{\n".$item."\n}\n";
$this->operationItem[$this->operationCount] = $item;
$this->operationCount++;
$offsetStr = $intStrEndStr2-4;
continue;
}else break;
}else break;
}
// echo "\n---------\n";
}
}catch(\app\library\_PajaxException $ex)
{
throw $ex;
}catch(\Throwable $thr) { throw new \app\library\_PajaxException($thr, \app\library\_PajaxException::THROWABLE, $this->getEnv()); }
}
/**
*
* @throws \app\library\_PajaxException
*/
private function parse()
{
try{
foreach ($this->operationItem as $ix => $trans)
{
// echo "\n".$trans."\n";
$trans = str_replace("\r\n", "", $trans);
$int61 = strpos($trans, ":61:");
$int86 = strpos($trans, ":86:", $int61);
if(is_numeric($int61) && is_numeric($int86) )
{
$str61 = substr($trans, $int61+4, $int86 - $int61-2) ;
$str86 = substr($trans, $int86+4);
// echo "\n[".$str61."]\n";
// echo "\n[".$str86."]\n";
$currencyDate = substr($str61, 0,6); //YYMMDD
$approveDate = substr($str61, 6, 4); //MMDD
$accPage = substr($str61,10, 2); //CN/CD etc.
$amPos = strpos($str61, ",",12);
$amount = ( substr($str61, 12, $amPos-12+3) );
$amount = str_replace(",", ".",$amount);
$amount = number_format($amount,2,'.','');
$exp = explode("^", $str86);
if(count($exp) < 2) $exp = explode(">", $str86);
if(count($exp) < 2) $exp = explode("<", $str86);
if(count($exp) < 2) $exp = explode("&", $str86);
if(count($exp) < 2) $exp = explode("*", $str86);
if(count($exp) < 2) $exp = explode(":", $str86);
$cl_name = '';
$title = '';
$unique = '';
if(count($exp)>0) $title .= $exp[0].' ';
if(count($exp)>1) $title .= $exp[1].' ';
if(count($exp)>2) $cl_name .= $exp[2].' ';
if(count($exp)>3) $cl_name .= $exp[3].' ';
if(count($exp)>4) $cl_name .= $exp[4].' ';
if(count($exp)>5) $cl_name .= $exp[5].' ';
foreach($exp as $xi => $ex_row) $unique .= $ex_row;
$unique = md5($unique);
// echo "\n[".$currencyDate.' | '.$approveDate.' | '.$accPage .' | '.$amount." | ".$title." | ".$cl_name." | ".$unique."]\n";
// echo "\n \n";
$this->details[] = array(
'unique' => $unique,
'amount' => $amount,
'currencyDate' => $currencyDate,
'approveDate' => $approveDate,
'title' => $title,
'name' => $cl_name,
);
}
}
}catch(\app\library\_PajaxException $ex)
{
throw $ex;
}catch(\Throwable $thr) { throw new \app\library\_PajaxException($thr, \app\library\_PajaxException::THROWABLE, $this->getEnv()); }
}
}