feat: added font_family to document.proto (#377)

- [ ] Regenerate this pull request now.

feat: added ImageQualityScores message to document.proto
feat: added PropertyMetadata and EntityTypeMetadata to document_schema.proto

PiperOrigin-RevId: 486975621

Source-Link: https://togithub.com/googleapis/googleapis/commit/398c9f9fccf4dea8d12e0b5a90ed671dec2ee387

Source-Link: https://togithub.com/googleapis/googleapis-gen/commit/7cd1f5f4e435777cb824af268dc8d37134613e6a
Copy-Tag: eyJwIjoiLmdpdGh1Yi8uT3dsQm90LnlhbWwiLCJoIjoiN2NkMWY1ZjRlNDM1Nzc3Y2I4MjRhZjI2OGRjOGQzNzEzNDYxM2U2YSJ9

protos/google/cloud/documentai/v1/barcode.proto

@@ -28,40 +28,42 @@ option ruby_package = "Google::Cloud::DocumentAI::V1";
 message Barcode {
   // Format of a barcode.
   // The supported formats are:
-  //   CODE_128: Code 128 type.
-  //   CODE_39: Code 39 type.
-  //   CODE_93: Code 93 type.
-  //   CODABAR: Codabar type.
-  //   DATA_MATRIX: 2D Data Matrix type.
-  //   ITF: ITF type.
-  //   EAN_13: EAN-13 type.
-  //   EAN_8: EAN-8 type.
-  //   QR_CODE: 2D QR code type.
-  //   UPC_A: UPC-A type.
-  //   UPC_E: UPC-E type.
-  //   PDF417: PDF417 type.
-  //   AZTEC: 2D Aztec code type.
-  //   DATABAR: GS1 DataBar code type.
+  //
+  // - `CODE_128`: Code 128 type.
+  // - `CODE_39`: Code 39 type.
+  // - `CODE_93`: Code 93 type.
+  // - `CODABAR`: Codabar type.
+  // - `DATA_MATRIX`: 2D Data Matrix type.
+  // - `ITF`: ITF type.
+  // - `EAN_13`: EAN-13 type.
+  // - `EAN_8`: EAN-8 type.
+  // - `QR_CODE`: 2D QR code type.
+  // - `UPC_A`: UPC-A type.
+  // - `UPC_E`: UPC-E type.
+  // - `PDF417`: PDF417 type.
+  // - `AZTEC`: 2D Aztec code type.
+  // - `DATABAR`: GS1 DataBar code type.
   string format = 1;
 
   // Value format describes the format of the value that a barcode
   // encodes.
   // The supported formats are:
-  //   CONTACT_INFO: Contact information.
-  //   EMAIL: Email address.
-  //   ISBN: ISBN identifier.
-  //   PHONE: Phone number.
-  //   PRODUCT: Product.
-  //   SMS: SMS message.
-  //   TEXT: Text string.
-  //   URL: URL address.
-  //   WIFI: Wifi information.
-  //   GEO: Geo-localization.
-  //   CALENDAR_EVENT: Calendar event.
-  //   DRIVER_LICENSE: Driver's license.
+  //
+  // - `CONTACT_INFO`: Contact information.
+  // - `EMAIL`: Email address.
+  // - `ISBN`: ISBN identifier.
+  // - `PHONE`: Phone number.
+  // - `PRODUCT`: Product.
+  // - `SMS`: SMS message.
+  // - `TEXT`: Text string.
+  // - `URL`: URL address.
+  // - `WIFI`: Wifi information.
+  // - `GEO`: Geo-localization.
+  // - `CALENDAR_EVENT`: Calendar event.
+  // - `DRIVER_LICENSE`: Driver's license.
   string value_format = 2;
 
   // Raw value encoded in the barcode.
-  // For example, 'MEBKM:TITLE:Google;URL:https://www.google.com;;'.
+  // For example: `'MEBKM:TITLE:Google;URL:https://www.google.com;;'`.
   string raw_value = 3;
 }

protos/google/cloud/documentai/v1/document.proto

@@ -91,6 +91,10 @@ message Document {
 
     // Font size.
     FontSize font_size = 7;
+
+    // Font family such as `Arial`, `Times New Roman`.
+    // https://www.w3schools.com/cssref/pr_font_font-family.asp
+    string font_family = 8;
   }
 
   // A page in a [Document][google.cloud.documentai.v1.Document].
@@ -169,7 +173,7 @@ message Document {
 
       // Confidence of the current [Layout][google.cloud.documentai.v1.Document.Page.Layout] within context of the object this
       // layout is for. e.g. confidence can be for a single token, a table,
-      // a visual element, etc. depending on context. Range [0, 1].
+      // a visual element, etc. depending on context. Range `[0, 1]`.
       float confidence = 2;
 
       // The bounding polygon for the [Layout][google.cloud.documentai.v1.Document.Page.Layout].
@@ -189,7 +193,7 @@ message Document {
       repeated DetectedLanguage detected_languages = 2;
 
       // The history of this annotation.
-      Provenance provenance = 3;
+      Provenance provenance = 3 [deprecated = true];
     }
 
     // A collection of lines that a human would perceive as a paragraph.
@@ -201,7 +205,7 @@ message Document {
       repeated DetectedLanguage detected_languages = 2;
 
       // The  history of this annotation.
-      Provenance provenance = 3;
+      Provenance provenance = 3 [deprecated = true];
     }
 
     // A collection of tokens that a human would perceive as a line.
@@ -214,7 +218,7 @@ message Document {
       repeated DetectedLanguage detected_languages = 2;
 
       // The  history of this annotation.
-      Provenance provenance = 3;
+      Provenance provenance = 3 [deprecated = true];
     }
 
     // A detected token.
@@ -249,8 +253,8 @@ message Document {
       // A list of detected languages together with confidence.
       repeated DetectedLanguage detected_languages = 3;
 
-      // The  history of this annotation.
-      Provenance provenance = 4;
+      // The history of this annotation.
+      Provenance provenance = 4 [deprecated = true];
     }
 
     // A detected symbol.
@@ -309,6 +313,9 @@ message Document {
 
       // A list of detected languages together with confidence.
       repeated DetectedLanguage detected_languages = 4;
+
+      // The history of this table.
+      Provenance provenance = 5;
     }
 
     // A form field detected on the page.
@@ -358,15 +365,43 @@ message Document {
 
     // Detected language for a structural component.
     message DetectedLanguage {
-      // The BCP-47 language code, such as "en-US" or "sr-Latn". For more
+      // The BCP-47 language code, such as `en-US` or `sr-Latn`. For more
       // information, see
       // https://www.unicode.org/reports/tr35/#Unicode_locale_identifier.
       string language_code = 1;
 
-      // Confidence of detected language. Range [0, 1].
+      // Confidence of detected language. Range `[0, 1]`.
       float confidence = 2;
     }
 
+    // Image Quality Scores for the page image
+    message ImageQualityScores {
+      // Image Quality Defects
+      message DetectedDefect {
+        // Name of the defect type. Supported values are:
+        //
+        // - `quality/defect_blurry`
+        // - `quality/defect_noisy`
+        // - `quality/defect_dark`
+        // - `quality/defect_faint`
+        // - `quality/defect_text_too_small`
+        // - `quality/defect_document_cutoff`
+        // - `quality/defect_text_cutoff`
+        // - `quality/defect_glare`
+        string type = 1;
+
+        // Confidence of detected defect. Range `[0, 1]` where 1 indicates
+        // strong confidence of that the defect exists.
+        float confidence = 2;
+      }
+
+      // The overall quality score. Range `[0, 1]` where 1 is perfect quality.
+      float quality_score = 1;
+
+      // A list of detected defects.
+      repeated DetectedDefect detected_defects = 2;
+    }
+
     // 1-based index for current [Page][google.cloud.documentai.v1.Document.Page] in a parent [Document][google.cloud.documentai.v1.Document].
     // Useful when a page is taken out of a [Document][google.cloud.documentai.v1.Document] for individual
     // processing.
@@ -422,8 +457,11 @@ message Document {
     // A list of detected barcodes.
     repeated DetectedBarcode detected_barcodes = 15;
 
+    // Image Quality Scores.
+    ImageQualityScores image_quality_scores = 17;
+
     // The history of this page.
-    Provenance provenance = 16;
+    Provenance provenance = 16 [deprecated = true];
   }
 
   // An entity that could be a phrase in the text or a property that belongs to
@@ -471,6 +509,7 @@ message Document {
       // or int normalized text by default.
       //
       // Below are sample formats mapped to structured values.
+      //
       // - Money/Currency type (`money_value`) is in the ISO 4217 text format.
       // - Date type (`date_value`) is in the ISO 8601 text format.
       // - Datetime type (`datetime_value`) is in the ISO 8601 text format.
@@ -484,14 +523,13 @@ message Document {
     // Required. Entity type from a schema e.g. `Address`.
     string type = 2 [(google.api.field_behavior) = REQUIRED];
 
-    // Optional. Text value in the document e.g. `1600 Amphitheatre Pkwy`. If the entity
-    // is not present in the document, this field will be empty.
+    // Optional. Text value of the entity e.g. `1600 Amphitheatre Pkwy`.
     string mention_text = 3 [(google.api.field_behavior) = OPTIONAL];
 
     // Optional. Deprecated.  Use `id` field instead.
     string mention_id = 4 [(google.api.field_behavior) = OPTIONAL];
 
-    // Optional. Confidence of detected Schema entity. Range [0, 1].
+    // Optional. Confidence of detected Schema entity. Range `[0, 1]`.
     float confidence = 5 [(google.api.field_behavior) = OPTIONAL];
 
     // Optional. Represents the provenance of this entity wrt. the location on the
@@ -605,7 +643,7 @@ message Document {
       // Optional. Identifies the bounding polygon of a layout element on the page.
       BoundingPoly bounding_poly = 4 [(google.api.field_behavior) = OPTIONAL];
 
-      // Optional. Confidence of detected page element, if applicable. Range [0, 1].
+      // Optional. Confidence of detected page element, if applicable. Range `[0, 1]`.
       float confidence = 5 [(google.api.field_behavior) = OPTIONAL];
     }
 
@@ -726,7 +764,7 @@ message Document {
     string changed_text = 2;
 
     // The history of this annotation.
-    repeated Provenance provenance = 3;
+    repeated Provenance provenance = 3 [deprecated = true];
   }
 
   // Original source document from the user.
@@ -765,9 +803,9 @@ message Document {
   // Placeholder.  Relationship among [Document.entities][google.cloud.documentai.v1.Document.entities].
   repeated EntityRelation entity_relations = 8;
 
-  // Placeholder.  A list of text corrections made to [Document.text].  This is
-  // usually used for annotating corrections to OCR mistakes.  Text changes for
-  // a given revision may not overlap with each other.
+  // Placeholder.  A list of text corrections made to [Document.text][google.cloud.documentai.v1.Document.text].  This
+  // is usually used for annotating corrections to OCR mistakes.  Text changes
+  // for a given revision may not overlap with each other.
   repeated TextChange text_changes = 14;
 
   // Information about the sharding if this document is sharded part of a larger

protos/google/cloud/documentai/v1/document_io.proto

@@ -16,6 +16,8 @@ syntax = "proto3";
 
 package google.cloud.documentai.v1;
 
+import "google/protobuf/field_mask.proto";
+
 option csharp_namespace = "Google.Cloud.DocumentAI.V1";
 option go_package = "google.golang.org/genproto/googleapis/cloud/documentai/v1;documentai";
 option java_multiple_files = true;
@@ -30,7 +32,7 @@ message RawDocument {
   bytes content = 1;
 
   // An IANA MIME type (RFC6838) indicating the nature and format of the
-  // [content].
+  // [content][google.cloud.documentai.v1.RawDocument.content].
   string mime_type = 2;
 }
 
@@ -59,7 +61,7 @@ message GcsPrefix {
 message BatchDocumentsInputConfig {
   // The source.
   oneof source {
-    // The set of documents that match the specified Cloud Storage [gcs_prefix].
+    // The set of documents that match the specified Cloud Storage `gcs_prefix`.
     GcsPrefix gcs_prefix = 1;
 
     // The set of documents individually specified on Cloud Storage.
@@ -74,6 +76,11 @@ message DocumentOutputConfig {
   message GcsOutputConfig {
     // The Cloud Storage uri (a directory) of the output.
     string gcs_uri = 1;
+
+    // Specifies which fields to include in the output documents.
+    // Only supports top level document and pages field so it must be in the
+    // form of `{document_field_name}` or `pages.{page_field_name}`.
+    google.protobuf.FieldMask field_mask = 2;
   }
 
   // The destination of the results.

protos/google/cloud/documentai/v1/document_processor_service.proto

@@ -279,6 +279,8 @@ message ProcessRequest {
   bool skip_human_review = 3;
 
   // Specifies which fields to include in ProcessResponse's document.
+  // Only supports top level document and pages field so it must be in the form
+  // of `{document_field_name}` or `pages.{page_field_name}`.
   google.protobuf.FieldMask field_mask = 6;
 }
 

protos/google/cloud/documentai/v1/document_schema.proto

@@ -86,15 +86,16 @@ message DocumentSchema {
     // Name of the type. It must be unique within the schema file and
     // cannot be a 'Common Type'.  Besides that we use the following naming
     // conventions:
-    // - *use snake_casing*
+    //
+    // - *use `snake_casing`*
     // - name matching is case-insensitive
     // - Maximum 64 characters.
     // - Must start with a letter.
     // - Allowed characters: ASCII letters `[a-z0-9_-]`.  (For backward
     //   compatibility internal infrastructure and tooling can handle any ascii
     //   character)
-    // - The '/' is sometimes used to denote a property of a type.  For example
-    //   line_item/amount.  This convention is deprecated, but will still be
+    // - The `/` is sometimes used to denote a property of a type.  For example
+    //   `line_item/amount`.  This convention is deprecated, but will still be
     //   honored for backward compatibility.
     string name = 1;
 

protos/google/cloud/documentai/v1/processor.proto

@@ -18,6 +18,7 @@ package google.cloud.documentai.v1;
 
 import "google/api/field_behavior.proto";
 import "google/api/resource.proto";
+import "google/cloud/documentai/v1/document_schema.proto";
 import "google/protobuf/timestamp.proto";
 
 option csharp_namespace = "Google.Cloud.DocumentAI.V1";
@@ -85,6 +86,9 @@ message ProcessorVersion {
   // The display name of the processor version.
   string display_name = 2;
 
+  // The schema of the processor version. Describes the output.
+  DocumentSchema document_schema = 12;
+
   // The state of the processor version.
   State state = 6;
 
@@ -125,14 +129,14 @@ message Processor {
     // The processor is disabled.
     DISABLED = 2;
 
-    // The processor is being enabled, will become ENABLED if successful.
+    // The processor is being enabled, will become `ENABLED` if successful.
     ENABLING = 3;
 
-    // The processor is being disabled, will become DISABLED if successful.
+    // The processor is being disabled, will become `DISABLED` if successful.
     DISABLING = 4;
 
-    // The processor is being created, will become either ENABLED (for
-    // successful creation) or FAILED (for failed ones).
+    // The processor is being created, will become either `ENABLED` (for
+    // successful creation) or `FAILED` (for failed ones).
     // Once a processor is in this state, it can then be used for document
     // processing, but the feature dependencies of the processor might not be
     // fully created yet.
@@ -154,7 +158,7 @@ message Processor {
     (google.api.field_behavior) = OUTPUT_ONLY
   ];
 
-  // The processor type, e.g., OCR_PROCESSOR, INVOICE_PROCESSOR, etc.
+  // The processor type, e.g., `OCR_PROCESSOR`, `INVOICE_PROCESSOR`, etc.
   // To get a list of processors types, see
   // [FetchProcessorTypes][google.cloud.documentai.v1.DocumentProcessorService.FetchProcessorTypes].
   string type = 2;

protos/google/cloud/documentai/v1/processor_type.proto

@@ -42,10 +42,10 @@ message ProcessorType {
   }
 
   // The resource name of the processor type.
-  // Format: projects/{project}/processorTypes/{processor_type}
+  // Format: `projects/{project}/processorTypes/{processor_type}`
   string name = 1;
 
-  // The type of the processor, e.g., "invoice_parsing".
+  // The processor type, e.g., `OCR_PROCESSOR`, `INVOICE_PROCESSOR`, etc.
   string type = 2;
 
   // The processor category, used by UI to group processor types.