Perl 简明教程

Perl - Error Handling

执行和错误总是同时发生的。如果你正在打开一个不存在的文件。那么如果你没有正确处理此情况，你的程序将被视为质量很差。

The execution and the errors always go together. If you are opening a file which does not exist. then if you did not handle this situation properly then your program is considered to be of bad quality.

如果发生错误，程序将停止。因此，使用适当的错误处理来处理各种类型的错误，这些错误可能会在程序执行期间发生并采取适当的操作，而不是完全停止程序。

The program stops if an error occurs. So a proper error handling is used to handle various type of errors, which may occur during a program execution and take appropriate action instead of halting program completely.

你可以用多种不同的方式识别和捕获错误。在 Perl 中捕获错误并正确处理它们非常容易。这里有一些可用的方法。

You can identify and trap an error in a number of different ways. Its very easy to trap errors in Perl and then handling them properly. Here are few methods which can be used.

The if statement

当需要检查语句的返回值时， if statement 是显而易见的选择；例如 -

The if statement is the obvious choice when you need to check the return value from a statement; for example −

if(open(DATA, $file)) {
   ...
} else {
   die "Error: Couldn't open the file - $!";
}

这里变量 $! 返回实际的错误消息。或者，在有意义的情况下，我们可以将语句缩减为一行；例如 -

Here variable $! returns the actual error message. Alternatively, we can reduce the statement to one line in situations where it makes sense to do so; for example −

open(DATA, $file) || die "Error: Couldn't open the file $!";

The unless Function

unless 函数与 if 相反：语句可以完全绕过成功状态，并且仅在表达式返回 false 时执行。例如 -

The unless function is the logical opposite to if: statements can completely bypass the success status and only be executed if the expression returns false. For example −

unless(chdir("/etc")) {
   die "Error: Can't change directory - $!";
}

unless 语句最适合在仅当表达式失败时引发错误或替代项时使用。当在单行语句中使用时，该语句也有意义 -

The unless statement is best used when you want to raise an error or alternative only if the expression fails. The statement also makes sense when used in a single-line statement −

die "Error: Can't change directory!: $!" unless(chdir("/etc"));

我们仅在 chdir 操作失败时才执行，并且它易于理解。

Here we die only if the chdir operation fails, and it reads nicely.

The ternary Operator

对于非常短的测试，可以使用条件运算符 ?:

For very short tests, you can use the conditional operator ?:

print(exists($hash{value}) ? 'There' : 'Missing',"\n");

这里我们尝试实现的目标并不是很清楚，但效果与使用 if 或 unless 语句相同。当希望在一个表达式或语句中快速返回两个值中的一个时，条件运算符最好用。

It’s not quite so clear here what we are trying to achieve, but the effect is the same as using an if or unless statement. The conditional operator is best used when you want to quickly return one of the two values within an expression or statement.

The warn Function

warn 函数只是发出警告，消息会打印到 STDERR，但不会执行进一步的操作。因此，如果你只想为用户打印警告并继续执行其余的操作，则它会更有用 -

The warn function just raises a warning, a message is printed to STDERR, but no further action is taken. So it is more useful if you just want to print a warning for the user and proceed with rest of the operation −

chdir('/etc') or warn "Can't change directory";

The die Function

die 函数的工作方式与 warn 一样，只是它还调用 exit。在正常脚本中，此函数的效果是立即终止执行。如果程序中出现错误并且继续进行毫无用处，则应使用此函数 -

The die function works just like warn, except that it also calls exit. Within a normal script, this function has the effect of immediately terminating execution. You should use this function in case it is useless to proceed if there is an error in the program −

chdir('/etc') or die "Can't change directory";

Errors within Modules

我们应该能够处理两种不同的情况 -

There are two different situations we should be able to handle −

  1. Reporting an error in a module that quotes the module’s filename and line number - this is useful when debugging a module, or when you specifically want to raise a module-related, rather than script-related, error.

  2. Reporting an error within a module that quotes the caller’s information so that you can debug the line within the script that caused the error. Errors raised in this fashion are useful to the end-user, because they highlight the error in relation to the calling script’s origination line.

当从模块中调用时， warn 和 die 的工作方式与你的预期略有不同。例如，简单的模块 -

The warn and die functions work slightly differently than you would expect when called from within a module. For example, the simple module −

package T;

require Exporter;
@ISA = qw/Exporter/;
@EXPORT = qw/function/;
use Carp;

sub function {
   warn "Error in module!";
}
1;

当从一个像下面这样的脚本调用时 -

When called from a script like below −

use T;
function();

它将产生以下结果 −

It will produce the following result −

Error in module! at T.pm line 9.

以下是您可能预期的大致意思，但未必是您所需。从模块编程人员的角度出发，这些信息很有用，因为它有助于指出模块自身中的错误。对于最终用户来说，提供的信息几乎无用，而且对于除经验丰富的程序员之外的所有人来说，它都是毫无意义的。

This is more or less what you might expected, but not necessarily what you want. From a module programmer’s perspective, the information is useful because it helps to point to a bug within the module itself. For an end-user, the information provided is fairly useless, and for all but the hardened programmer, it is completely pointless.

对于此类问题，解决方案是 Carp 模块，它提供了一种简化方法来报告模块中的错误，并返回有关调用脚本的信息。Carp 模块提供四个函数：carp、cluck、croak 和 confess。这些函数在下面讨论。

The solution for such problems is the Carp module, which provides a simplified method for reporting errors within modules that return information about the calling script. The Carp module provides four functions: carp, cluck, croak, and confess. These functions are discussed below.

The carp Function

carp 函数是 warn 的基本等效项，它将消息打印到 STDERR 而不会实际退出脚本和打印脚本名称。

The carp function is the basic equivalent of warn and prints the message to STDERR without actually exiting the script and printing the script name.

package T;

require Exporter;
@ISA = qw/Exporter/;
@EXPORT = qw/function/;
use Carp;

sub function {
   carp "Error in module!";
}
1;

当从一个像下面这样的脚本调用时 -

When called from a script like below −

use T;
function();

它将产生以下结果 −

It will produce the following result −

Error in module! at test.pl line 4

The cluck Function

cluck 函数是一种超强的 carp，它遵循相同的基本原则，但还会打印包含所有导致调用该函数的模块的堆栈跟踪，其中包括原始脚本的信息。

The cluck function is a sort of supercharged carp, it follows the same basic principle but also prints a stack trace of all the modules that led to the function being called, including the information on the original script.

package T;

require Exporter;
@ISA = qw/Exporter/;
@EXPORT = qw/function/;
use Carp qw(cluck);

sub function {
   cluck "Error in module!";
}
1;

当从一个像下面这样的脚本调用时 -

When called from a script like below −

use T;
function();

它将产生以下结果 −

It will produce the following result −

Error in module! at T.pm line 9
   T::function() called at test.pl line 4

The croak Function

功能 croak 等同于 die ，不同之处在于它将调用者报告到上一级。像 die 一样，此函数在向 STDERR 报告错误后也会退出脚本 -

The croak function is equivalent to die, except that it reports the caller one level up. Like die, this function also exits the script after reporting the error to STDERR −

package T;

require Exporter;
@ISA = qw/Exporter/;
@EXPORT = qw/function/;
use Carp;

sub function {
   croak "Error in module!";
}
1;

当从一个像下面这样的脚本调用时 -

When called from a script like below −

use T;
function();

它将产生以下结果 −

It will produce the following result −

Error in module! at test.pl line 4

与 carp 一样，关于包含行和文件信息也适用于 warn 和 die 函数的基本规则。

As with carp, the same basic rules apply regarding the including of line and file information according to the warn and die functions.

The confess Function

功能 confess 类似于 cluck ；它调用 die，然后打印一个堆栈跟踪，一直到原始脚本。

The confess function is like cluck; it calls die and then prints a stack trace all the way up to the origination script.

package T;

require Exporter;
@ISA = qw/Exporter/;
@EXPORT = qw/function/;
use Carp;

sub function {
   confess "Error in module!";
}
1;

当从一个像下面这样的脚本调用时 -

When called from a script like below −

use T;
function();

它将产生以下结果 −

It will produce the following result −

Error in module! at T.pm line 9
   T::function() called at test.pl line 4