feat: InfoType categories were added to built-in infoTypes (#730)

* feat: InfoType categories were added to built-in infoTypes

PiperOrigin-RevId: 460542545

Source-Link: googleapis/googleapis@898fcea

Source-Link: https://github.com/googleapis/googleapis-gen/commit/e97b0c9527bb7172a3f8f3bc411de83468fd6a77
Copy-Tag: eyJwIjoiLmdpdGh1Yi8uT3dsQm90LnlhbWwiLCJoIjoiZTk3YjBjOTUyN2JiNzE3MmEzZjhmM2JjNDExZGU4MzQ2OGZkNmE3NyJ9

* 🦉 Updates from OwlBot post-processor

See https://github.com/googleapis/repo-automation-bots/blob/main/packages/owl-bot/README.md

Co-authored-by: Owl Bot <gcf-owl-bot[bot]@users.noreply.github.com>

protos/google/privacy/dlp/v2/dlp.proto

@@ -773,7 +773,7 @@ message InspectConfig {
   FindingLimits limits = 3;
 
   // When true, a contextual quote from the data that triggered a finding is
-  // included in the response; see Finding.quote.
+  // included in the response; see [Finding.quote][google.privacy.dlp.v2.Finding.quote].
   // This is not used for data profiling.
   bool include_quote = 4;
 
@@ -1554,6 +1554,201 @@ message InfoTypeDescription {
   // Description of the infotype. Translated when language is provided in the
   // request.
   string description = 4;
+
+  // The category of the infoType.
+  repeated InfoTypeCategory categories = 10;
+}
+
+// Classification of infoTypes to organize them according to geographic
+// location, industry, and data type.
+message InfoTypeCategory {
+  // Enum of the current locations.
+  // We might add more locations in the future.
+  enum LocationCategory {
+    // Unused location
+    LOCATION_UNSPECIFIED = 0;
+
+    // The infoType is not issued by or tied to a specific region, but is used
+    // almost everywhere.
+    GLOBAL = 1;
+
+    // The infoType is typically used in Argentina.
+    ARGENTINA = 2;
+
+    // The infoType is typically used in Australia.
+    AUSTRALIA = 3;
+
+    // The infoType is typically used in Belgium.
+    BELGIUM = 4;
+
+    // The infoType is typically used in Brazil.
+    BRAZIL = 5;
+
+    // The infoType is typically used in Canada.
+    CANADA = 6;
+
+    // The infoType is typically used in Chile.
+    CHILE = 7;
+
+    // The infoType is typically used in China.
+    CHINA = 8;
+
+    // The infoType is typically used in Colombia.
+    COLOMBIA = 9;
+
+    // The infoType is typically used in Denmark.
+    DENMARK = 10;
+
+    // The infoType is typically used in France.
+    FRANCE = 11;
+
+    // The infoType is typically used in Finland.
+    FINLAND = 12;
+
+    // The infoType is typically used in Germany.
+    GERMANY = 13;
+
+    // The infoType is typically used in Hong Kong.
+    HONG_KONG = 14;
+
+    // The infoType is typically used in India.
+    INDIA = 15;
+
+    // The infoType is typically used in Indonesia.
+    INDONESIA = 16;
+
+    // The infoType is typically used in Ireland.
+    IRELAND = 17;
+
+    // The infoType is typically used in Israel.
+    ISRAEL = 18;
+
+    // The infoType is typically used in Italy.
+    ITALY = 19;
+
+    // The infoType is typically used in Japan.
+    JAPAN = 20;
+
+    // The infoType is typically used in Korea.
+    KOREA = 21;
+
+    // The infoType is typically used in Mexico.
+    MEXICO = 22;
+
+    // The infoType is typically used in the Netherlands.
+    THE_NETHERLANDS = 23;
+
+    // The infoType is typically used in Norway.
+    NORWAY = 24;
+
+    // The infoType is typically used in Paraguay.
+    PARAGUAY = 25;
+
+    // The infoType is typically used in Peru.
+    PERU = 26;
+
+    // The infoType is typically used in Poland.
+    POLAND = 27;
+
+    // The infoType is typically used in Portugal.
+    PORTUGAL = 28;
+
+    // The infoType is typically used in Singapore.
+    SINGAPORE = 29;
+
+    // The infoType is typically used in South Africa.
+    SOUTH_AFRICA = 30;
+
+    // The infoType is typically used in Spain.
+    SPAIN = 31;
+
+    // The infoType is typically used in Sweden.
+    SWEDEN = 32;
+
+    // The infoType is typically used in Taiwan.
+    TAIWAN = 33;
+
+    // The infoType is typically used in Thailand.
+    THAILAND = 34;
+
+    // The infoType is typically used in Turkey.
+    TURKEY = 35;
+
+    // The infoType is typically used in the United Kingdom.
+    UNITED_KINGDOM = 36;
+
+    // The infoType is typically used in the United States.
+    UNITED_STATES = 37;
+
+    // The infoType is typically used in Uruguay.
+    URUGUAY = 38;
+
+    // The infoType is typically used in Venezuela.
+    VENEZUELA = 39;
+
+    // The infoType is typically used in Google internally.
+    INTERNAL = 40;
+  }
+
+  // Enum of the current industries in the category.
+  // We might add more industries in the future.
+  enum IndustryCategory {
+    // Unused industry
+    INDUSTRY_UNSPECIFIED = 0;
+
+    // The infoType is typically used in the finance industry.
+    FINANCE = 1;
+
+    // The infoType is typically used in the health industry.
+    HEALTH = 2;
+
+    // The infoType is typically used in the telecommunications industry.
+    TELECOMMUNICATIONS = 3;
+  }
+
+  // Enum of the current types in the category.
+  // We might add more types in the future.
+  enum TypeCategory {
+    // Unused type
+    TYPE_UNSPECIFIED = 0;
+
+    // Personally identifiable information, for example, a
+    // name or phone number
+    PII = 1;
+
+    // Personally identifiable information that is especially sensitive, for
+    // example, a passport number.
+    SPII = 2;
+
+    // Attributes that can partially identify someone, especially in
+    // combination with other attributes, like age, height, and gender.
+    DEMOGRAPHIC = 3;
+
+    // Confidential or secret information, for example, a password.
+    CREDENTIAL = 4;
+
+    // An identification document issued by a government.
+    GOVERNMENT_ID = 5;
+
+    // A document, for example, a resume or source code.
+    DOCUMENT = 6;
+
+    // Information that is not sensitive on its own, but provides details about
+    // the circumstances surrounding an entity or an event.
+    CONTEXTUAL_INFORMATION = 7;
+  }
+
+  oneof category {
+    // The region or country that issued the ID or document represented by the
+    // infoType.
+    LocationCategory location_category = 1;
+
+    // The group of relevant businesses where this infoType is commonly used
+    IndustryCategory industry_category = 2;
+
+    // The class of identifiers where this infoType belongs
+    TypeCategory type_category = 3;
+  }
 }
 
 // Request for the list of infoTypes.

protos/google/privacy/dlp/v2/storage.proto

@@ -149,7 +149,11 @@ message CustomInfoType {
     // Message for specifying a window around a finding to apply a detection
     // rule.
     message Proximity {
-      // Number of characters before the finding to consider.
+      // 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]
+      // (https://cloud.google.com/dlp/docs/creating-custom-infotypes-likelihood#match-column-values).
       int32 window_before = 1;
 
       // Number of characters after the finding to consider.
@@ -181,14 +185,19 @@ message CustomInfoType {
       // Regular expression pattern defining what qualifies as a hotword.
       Regex hotword_regex = 1;
 
-      // Proximity of the finding within which the entire hotword must reside.
-      // The total length of the window cannot exceed 1000 characters. Note that
-      // the finding itself will be included in the window, so that hotwords may
-      // be used to match substrings of the finding itself. For example, the
-      // certainty of a phone number regex "\(\d{3}\) \d{3}-\d{4}" could be
-      // adjusted upwards if the area code is known to be the local area code of
-      // a company office using the hotword regex "\(xxx\)", where "xxx"
-      // is the area code in question.
+      // 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]
+      // (https://cloud.google.com/dlp/docs/creating-custom-infotypes-likelihood#match-column-values).
       Proximity proximity = 2;
 
       // Likelihood adjustment to apply to all matching findings.