Cisco Cisco Email Security Appliance C170 사용자 가이드

The primary case for using regular expressions in content matching classifiers is to detect specific 
account, patient, or student identification numbers. These are usually simple regular expressions that 
describe patterns of numbers and letters. For example: 

An 8-digit number: 

\d{8}

Identification code with hyphens between sets of numbers: 

\d{3}-\d{4}-\d

Identification code that begins with a single letter that can be upper or lower case: 

[a-zA-Z]\d{7}

Identification code that begins with three digits and is followed by nine uppercase letters: 

\d{3}[A-Z]{9}

Using 

|

 to define two different number patterns to search for: 

\d{3}[A-Z]{9}|\d{2}[A-Z]{9}-\d

Regular expressions are case sensitive, so they should include upper and lower case, such as 

[a-zA-Z]

. 

If only certain letters are used, you can define the regular expression accordingly.

The less specific the pattern, such as an 8-digit number, the more likely you will want the policy to search 
for additional words and phrases to distinguish a random 8-digit number from an actual customer 
number. 

If the available RSA Email DLP policy templates do not meet the unique requirements of your 
organization, a number of options are available for creating your own DLP policies from scratch. These 
options include:

Creating your own DLP policy using the Custom Policy Template

Creating your own classifiers to use in a custom policy

Creating and importing your own DLP dictionaries to use in a custom policy

These options are advanced and should only be used in cases where predefined settings do not meet your 
organization’s needs.

The regular expression notation that indicates the number of 
repetitions of the previous token is supported. 

For example, the expression “

\d{8}

” matches 

12345678

 and 

11223344

 but not 

8

.

Alternation, or the “or” operator. If A and B are regular expressions, 
the expression “

A|B

” will match any string that matches either “A” or 

“B.” Can be used to combine number patterns in a regular expression.

For example, the expression “

foo|bar

” will match either 

foo

 or 

bar

, 

but not 

foobar

.