Ignoring Patterns
Describes how Philter can ignore terms that match patterns.
Unlike the Ignoring Terms functionality, the ignoring patterns functionality is able to exclude text from being identified as sensitive based on one or more regular expression patterns.
An example use of this feature is to ignore terms that are present in your text but dynamic, such as logged timestamps. When using the date filter these timestamps may be identified as being sensitive even though they are not. Using an ignore pattern we can exclude the timestamps from being identified as sensitive information.
Ignore patterns can be specified at the filter profile level and/or at the level of each type of filter. When set at the filter profile level, the list of ignored patterns will be applied to all filter types. When set for an individual filter, the list of ignored patterns will be applied only to that filter.
In the filter profile shown below, ignore patterns are set at the level of the filter profile. The patterns specified in the list will be ignored for all filter types enabled in the filter profile.
{
"name": "example-filter-profile",
"ignoredPatterns": [
{
"name": "ignore-room-numbers",
"pattern": "Room [A-Z0-4]{4}"
}
],
"identifiers": {
"emailAddress": {
"emailAddressFilterStrategies": [
{
"strategy": "REDACT",
"redactionFormat": "{{{REDACTED-%t}}}"
}
]
}
}
}
In the filter profile shown below, ignore patterns are set at the level of a filter. The patterns specified in the list will be ignored only for that filter type. Each filter in a filter profile can have its own list of ignored patterns.
{
"name": "example-filter-profile",
"identifiers": {
"emailAddress": {
"ignoredPatterns": [
{
"name": "ignore-room-numbers",
"pattern": "Room [A-Z0-4]{4}"
}
],
"emailAddressFilterStrategies": [
{
"strategy": "REDACT",
"redactionFormat": "{{{REDACTED-%t}}}"
}
]
}
}
}
Last modified 4mo ago