InspectConfig

{
  "infoType": {
    object (InfoType)
  },
  "minLikelihood": enum (Likelihood)
}

object (InfoType)

Type of information the likelihood threshold applies to. Only one likelihood per infoType should be provided. If InfoTypeLikelihood does not have an infoType, the configuration fails.

enum (Likelihood)

Only returns findings equal to or above this threshold. This field is required or else the configuration fails.

{
  "maxFindingsPerItem": integer,
  "maxFindingsPerRequest": integer,
  "maxFindingsPerInfoType": [
    {
      object (InfoTypeLimit)
    }
  ]
}

integer

Max number of findings that are returned for each item scanned.

When set within an InspectContentRequest, this field is ignored.

This value isn't a hard limit. If the number of findings for an item reaches this limit, the inspection of that item ends gradually, not abruptly. Therefore, the actual number of findings that Cloud DLP returns for the item can be multiple times higher than this value.

integer

Max number of findings that are returned per request or job.

If you set this field in an InspectContentRequest, the resulting maximum value is the value that you set or 3,000, whichever is lower.

This value isn't a hard limit. If an inspection reaches this limit, the inspection ends gradually, not abruptly. Therefore, the actual number of findings that Cloud DLP returns can be multiple times higher than this value.

object (InfoTypeLimit)

Configuration of findings limit given for specified infoTypes.

{
  "infoType": {
    object (InfoType)
  },
  "maxFindings": integer
}

object (InfoType)

Type of information the findings limit applies to. Only one limit per infoType should be provided. If InfoTypeLimit does not have an infoType, the DLP API applies the limit against all infoTypes that are found but not specified in another InfoTypeLimit.

integer

Max findings limit for the given infoType.

{
  "infoType": {
    object (InfoType)
  },
  "likelihood": enum (Likelihood),
  "detectionRules": [
    {
      object (DetectionRule)
    }
  ],
  "exclusionType": enum (ExclusionType),
  "sensitivityScore": {
    object (SensitivityScore)
  },

  // Union field type can be only one of the following:
  "dictionary": {
    object (Dictionary)
  },
  "regex": {
    object (Regex)
  },
  "surrogateType": {
    object (SurrogateType)
  },
  "storedType": {
    object (StoredType)
  }
  // End of list of possible types for union field type.
}

object (InfoType)

CustomInfoType can either be a new infoType, or an extension of built-in infoType, when the name matches one of existing infoTypes and that infoType is specified in content.inspect.info_types field. Specifying the latter adds findings to the one detected by the system. If built-in info type is not specified in content.inspect.info_types list then the name is treated as a custom info type.

enum (Likelihood)

Likelihood to return for this CustomInfoType. This base value can be altered by a detection rule if the finding meets the criteria specified by the rule. Defaults to VERY_LIKELY if not specified.

object (DetectionRule)

Set of detection rules to apply to all findings of this CustomInfoType. Rules are applied in order that they are specified. Not supported for the surrogateType CustomInfoType.

enum (ExclusionType)

If set to EXCLUSION_TYPE_EXCLUDE this infoType will not cause a finding to be returned. It still can be used for rules matching.

object (SensitivityScore)

Sensitivity for this CustomInfoType. If this CustomInfoType extends an existing InfoType, the sensitivity here will take precedence over that of the original InfoType. If unset for a CustomInfoType, it will default to HIGH. This only applies to data profiling.

object (Dictionary)

A list of phrases to detect as a CustomInfoType.

object (Regex)

Regular expression based CustomInfoType.

object (SurrogateType)

Message for detecting output from deidentification transformations that support reversing.

object (StoredType)

Load an existing StoredInfoType resource for use in InspectDataSource. Not currently supported in content.inspect.

{
  "name": string,
  "createTime": string
}

string

Resource name of the requested StoredInfoType, for example organizations/433245324/storedInfoTypes/432452342 or projects/project-id/storedInfoTypes/432452342.

string (Timestamp format)

Timestamp indicating when the version of the StoredInfoType used for inspection was created. Output-only field, populated by the system.

A timestamp in RFC3339 UTC "Zulu" format, with nanosecond resolution and up to nine fractional digits. Examples: "2014-10-02T15:01:23Z" and "2014-10-02T15:01:23.045123456Z".

{

  // Union field type can be only one of the following:
  "hotwordRule": {
    object (HotwordRule)
  }
  // End of list of possible types for union field type.
}

object (HotwordRule)

Hotword-based detection rule.

{
  "hotwordRegex": {
    object (Regex)
  },
  "proximity": {
    object (Proximity)
  },
  "likelihoodAdjustment": {
    object (LikelihoodAdjustment)
  }
}

object (Regex)

Regular expression pattern defining what qualifies as a hotword.

object (Proximity)

Range of characters within which the entire hotword must reside. The total length of the window cannot exceed 1000 characters. The finding itself will be included in the window, so that hotwords can be used to match substrings of the finding itself. Suppose you want Cloud DLP to promote the likelihood of the phone number regex "(\d{3}) \d{3}-\d{4}" if the area code is known to be the area code of a company's office. In this case, use the hotword regex "(xxx)", where "xxx" is the area code in question.

For tabular data, if you want to modify the likelihood of an entire column of findngs, see Hotword example: Set the match likelihood of a table column.

object (LikelihoodAdjustment)

Likelihood adjustment to apply to all matching findings.

{
  "windowBefore": integer,
  "windowAfter": integer
}

integer

Number of characters before the finding to consider. For tabular data, if you want to modify the likelihood of an entire column of findngs, set this to 1. For more information, see Hotword example: Set the match likelihood of a table column.

integer

Number of characters after the finding to consider.

{

  // Union field adjustment can be only one of the following:
  "fixedLikelihood": enum (Likelihood),
  "relativeLikelihood": integer
  // End of list of possible types for union field adjustment.
}

enum (Likelihood)

Set the likelihood of a finding to a fixed value.

integer

Increase or decrease the likelihood by the specified number of levels. For example, if a finding would be POSSIBLE without the detection rule and relativeLikelihood is 1, then it is upgraded to LIKELY, while a value of -1 would downgrade it to UNLIKELY. Likelihood may never drop below VERY_UNLIKELY or exceed VERY_LIKELY, so applying an adjustment of 1 followed by an adjustment of -1 when base likelihood is VERY_LIKELY will result in a final likelihood of LIKELY.

{
  "infoTypes": [
    {
      object (InfoType)
    }
  ],
  "rules": [
    {
      object (InspectionRule)
    }
  ]
}

object (InfoType)

List of infoTypes this rule set is applied to.

object (InspectionRule)

Set of rules to be applied to infoTypes. The rules are applied in order.

{

  // Union field type can be only one of the following:
  "hotwordRule": {
    object (HotwordRule)
  },
  "exclusionRule": {
    object (ExclusionRule)
  }
  // End of list of possible types for union field type.
}

object (HotwordRule)

Hotword-based detection rule.

object (ExclusionRule)

Exclusion rule.

{
  "matchingType": enum (MatchingType),

  // Union field type can be only one of the following:
  "dictionary": {
    object (Dictionary)
  },
  "regex": {
    object (Regex)
  },
  "excludeInfoTypes": {
    object (ExcludeInfoTypes)
  },
  "excludeByHotword": {
    object (ExcludeByHotword)
  }
  // End of list of possible types for union field type.
}

enum (MatchingType)

How the rule is applied, see MatchingType documentation for details.

object (Dictionary)

Dictionary which defines the rule.

object (Regex)

Regular expression which defines the rule.

object (ExcludeInfoTypes)

Set of infoTypes for which findings would affect this rule.

object (ExcludeByHotword)

Drop if the hotword rule is contained in the proximate context. For tabular data, the context includes the column name.

{
  "infoTypes": [
    {
      object (InfoType)
    }
  ]
}

object (InfoType)

InfoType list in ExclusionRule rule drops a finding when it overlaps or contained within with a finding of an infoType from this list. For example, for InspectionRuleSet.info_types containing "PHONE_NUMBER"and exclusionRulecontainingexcludeInfoTypes.info_types` with "EMAIL_ADDRESS" the phone number findings are dropped if they overlap with EMAIL_ADDRESS finding. That leads to "[email protected]" to generate only a single finding, namely email address.

{
  "hotwordRegex": {
    object (Regex)
  },
  "proximity": {
    object (Proximity)
  }
}

object (Regex)

Regular expression pattern defining what qualifies as a hotword.

object (Proximity)

Range of characters within which the entire hotword must reside. The total length of the window cannot exceed 1000 characters. The windowBefore property in proximity should be set to 1 if the hotword needs to be included in a column header.