docs: Regenerated REST and RPC Document AI documentation to include missing changes from September 2024 to November 2025

Google APIs · copybara-github · commit 64f78809d70e · 2026-01-21T16:25:02.000-08:00
docs: Fixed language issues in multiple message field descriptions

PiperOrigin-RevId: 859310712
diff --git a/google/cloud/documentai/v1/document.proto b/google/cloud/documentai/v1/document.proto
@@ -1195,10 +1195,10 @@ message Document {
   // entity extraction results at different stages of the processing.
   repeated EntitiesRevision entities_revisions = 22;
 
-  // The entity revision id that `document.entities` field is based on.
+  // The entity revision ID that `document.entities` field is based on.
   // If this field is set and `entities_revisions` is not empty, the entities in
   // `document.entities` field are the entities in the entity revision with this
-  // id and `document.entity_validation_output` field is the
+  // ID and `document.entity_validation_output` field is the
   // `entity_validation_output` field in this entity revision.
   string entities_revision_id = 23;
 }
diff --git a/google/cloud/documentai/v1beta3/document.proto b/google/cloud/documentai/v1beta3/document.proto
@@ -180,9 +180,9 @@ message Document {
 
       // Confidence of the current
       // [Layout][google.cloud.documentai.v1beta3.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]`.
+      // context of the object this layout is for. For example, confidence can
+      // be for a single token, a table, a visual element, etc. depending on
+      // context. Range `[0, 1]`.
       float confidence = 2;
 
       // The bounding polygon for the
@@ -340,8 +340,8 @@ message Document {
       repeated DetectedLanguage detected_languages = 2;
     }
 
-    // Detected non-text visual elements e.g. checkbox, signature etc. on the
-    // page.
+    // Detected non-text visual elements, for example, checkbox, signature, etc.
+    // on the page.
     message VisualElement {
       // [Layout][google.cloud.documentai.v1beta3.Document.Page.Layout] for
       // [VisualElement][google.cloud.documentai.v1beta3.Document.Page.VisualElement].
@@ -400,7 +400,8 @@ message Document {
     message FormField {
       // [Layout][google.cloud.documentai.v1beta3.Document.Page.Layout] for the
       // [FormField][google.cloud.documentai.v1beta3.Document.Page.FormField]
-      // name. e.g. `Address`, `Email`, `Grand total`, `Phone number`, etc.
+      // name. For example, `Address`, `Email`, `Grand total`, `Phone number`,
+      // etc.
       Layout field_name = 1;
 
       // [Layout][google.cloud.documentai.v1beta3.Document.Page.Layout] for the
@@ -529,7 +530,7 @@ message Document {
     // A list of visually detected tokens on the page.
     repeated Token tokens = 8;
 
-    // A list of detected non-text visual elements e.g. checkbox,
+    // A list of detected non-text visual elements for example, checkbox,
     // signature etc. on the page.
     repeated VisualElement visual_elements = 9;
 
@@ -627,10 +628,10 @@ message Document {
     // [Document.text][google.cloud.documentai.v1beta3.Document.text].
     TextAnchor text_anchor = 1 [(google.api.field_behavior) = OPTIONAL];
 
-    // Required. Entity type from a schema e.g. `Address`.
+    // Required. Entity type from a schema for example, `Address`.
     string type = 2 [(google.api.field_behavior) = REQUIRED];
 
-    // Optional. Text value of the entity e.g. `1600 Amphitheatre Pkwy`.
+    // Optional. Text value of the entity for example, `1600 Amphitheatre Pkwy`.
     string mention_text = 3 [(google.api.field_behavior) = OPTIONAL];
 
     // Optional. Deprecated.  Use `id` field instead.
@@ -643,14 +644,14 @@ message Document {
     // the page where it was found.
     PageAnchor page_anchor = 6 [(google.api.field_behavior) = OPTIONAL];
 
-    // Optional. Canonical id. This will be a unique value in the entity list
+    // Optional. Canonical ID. This will be a unique value in the entity list
     // for this document.
     string id = 7 [(google.api.field_behavior) = OPTIONAL];
 
     // Optional. Normalized entity value. Absent if the extracted value could
-    // not be converted or the type (e.g. address) is not supported for certain
-    // parsers. This field is also only populated for certain supported document
-    // types.
+    // not be converted or the type (for example, address) is not supported for
+    // certain parsers. This field is also only populated for certain supported
+    // document types.
     NormalizedValue normalized_value = 9
         [(google.api.field_behavior) = OPTIONAL];
 
@@ -672,10 +673,10 @@ message Document {
   // Relationship between
   // [Entities][google.cloud.documentai.v1beta3.Document.Entity].
   message EntityRelation {
-    // Subject entity id.
+    // Subject entity ID.
     string subject_id = 1;
 
-    // Object entity id.
+    // Object entity ID.
     string object_id = 2;
 
     // Relationship description.
@@ -805,7 +806,7 @@ message Document {
       // of entities, properties within entities, etc.) in the parent revision.
       int32 index = 3;
 
-      // The id of the parent provenance.
+      // The ID of the parent provenance.
       int32 id = 2 [deprecated = true];
     }
 
@@ -845,7 +846,7 @@ message Document {
     // The index of the revision that produced this element.
     int32 revision = 1 [deprecated = true];
 
-    // The Id of this operation.  Needs to be unique within the scope of the
+    // The ID of this operation.  Needs to be unique within the scope of the
     // revision.
     int32 id = 2 [deprecated = true];
 
@@ -860,7 +861,7 @@ message Document {
   message Revision {
     // Human Review information of the document.
     message HumanReview {
-      // Human review state. e.g. `requested`, `succeeded`, `rejected`.
+      // Human review state. For example, `requested`, `succeeded`, `rejected`.
       string state = 1;
 
       // A message providing more details about the current state of processing.
@@ -870,7 +871,7 @@ message Document {
 
     // Who/what made the change
     oneof source {
-      // If the change was made by a person specify the name or id of that
+      // If the change was made by a person specify the name or ID of that
       // person.
       string agent = 4;
 
@@ -879,7 +880,7 @@ message Document {
       string processor = 5;
     }
 
-    // Id of the revision, internally generated by doc proto storage.
+    // ID of the revision, internally generated by doc proto storage.
     // Unique within the context of the document.
     string id = 1;
 
@@ -888,7 +889,7 @@ message Document {
     // index into the `revisions` field.
     repeated int32 parent = 2 [deprecated = true];
 
-    // The revisions that this revision is based on. Must include all the ids
+    // The revisions that this revision is based on. Must include all the IDs
     // that have anything to do with this revision - eg. there are
     // `provenance.parent.revision` fields that index into this field.
     repeated string parent_ids = 7;
@@ -1012,18 +1013,18 @@ message Document {
       message LayoutImageBlock {
         // Source of the image.
         oneof image_source {
-          // Optional. Asset id of the inline image. If set, find the image
+          // Optional. Asset ID of the inline image. If set, find the image
           // content in the blob_assets field.
           string blob_asset_id = 4 [(google.api.field_behavior) = OPTIONAL];
 
-          // Optional. Google Cloud Storage uri of the image.
+          // Optional. Google Cloud Storage URI of the image.
           string gcs_uri = 5 [(google.api.field_behavior) = OPTIONAL];
 
-          // Optional. Data uri of the image.
+          // Optional. Data URI of the image.
           // It is composed of four parts: a prefix (data:), a MIME type
           // indicating the type of data, an optional base64 token if
           // non-textual, and the data itself:
-          // data:[<mediatype>][;base64],<data>
+          // data:[<mediatype>][;base64],<data>.
           string data_uri = 6 [(google.api.field_behavior) = OPTIONAL];
         }
 
@@ -1102,18 +1103,18 @@ message Document {
       message ImageChunkField {
         // Source of the image.
         oneof image_source {
-          // Optional. Asset id of the inline image. If set, find the image
+          // Optional. Asset ID of the inline image. If set, find the image
           // content in the blob_assets field.
           string blob_asset_id = 1 [(google.api.field_behavior) = OPTIONAL];
 
-          // Optional. Google Cloud Storage uri of the image.
+          // Optional. Google Cloud Storage URI of the image.
           string gcs_uri = 2 [(google.api.field_behavior) = OPTIONAL];
 
-          // Optional. Data uri of the image.
+          // Optional. Data URI of the image.
           // It is composed of four parts: a prefix (data:), a MIME type
           // indicating the type of data, an optional base64 token if
           // non-textual, and the data itself:
-          // data:[<mediatype>][;base64],<data>
+          // data:[<mediatype>][;base64],<data>.
           string data_uri = 3 [(google.api.field_behavior) = OPTIONAL];
         }
 
@@ -1128,7 +1129,7 @@ message Document {
       }
 
       // The chunk field in the chunk. A chunk field could be one of the various
-      // types (e.g. image, table) supported.
+      // types (for example, image, table) supported.
       message ChunkField {
         // The type of the chunk field.
         oneof field_type {
@@ -1167,13 +1168,13 @@ message Document {
   }
 
   // Represents a blob asset. It's used to store the content of the inline blob
-  // in this document, e.g. image bytes, such that it can be referenced by
-  // other fields in the document via asset id.
+  // in this document, for example, image bytes, such that it can be referenced
+  // by other fields in the document via asset ID.
   message BlobAsset {
-    // Optional. The id of the blob asset.
+    // Optional. The ID of the blob asset.
     string asset_id = 1 [(google.api.field_behavior) = OPTIONAL];
 
-    // Optional. The content of the blob asset, e.g. image bytes.
+    // Optional. The content of the blob asset, for example, image bytes.
     bytes content = 2 [(google.api.field_behavior) = OPTIONAL];
 
     // The mime type of the blob asset.
@@ -1228,7 +1229,7 @@ message Document {
 
   // Entity revision.
   message EntitiesRevision {
-    // The revision id.
+    // The revision ID.
     string revision_id = 1;
 
     // The entities in this revision.
@@ -1302,8 +1303,9 @@ message Document {
   ChunkedDocument chunked_document = 18;
 
   // Optional. The blob assets in this document. This is used to store the
-  // content of the inline blobs in this document, e.g. image bytes, such that
-  // it can be referenced by other fields in the document via asset id.
+  // content of the inline blobs in this document, for example, image bytes,
+  // such that it can be referenced by other fields in the document via asset
+  // id.
   repeated BlobAsset blob_assets = 19 [(google.api.field_behavior) = OPTIONAL];
 
   // The entity validation output for the document. This is the validation
@@ -1315,10 +1317,10 @@ message Document {
   // entity extraction results at different stages of the processing.
   repeated EntitiesRevision entities_revisions = 22;
 
-  // The entity revision id that `document.entities` field is based on.
+  // The entity revision ID that `document.entities` field is based on.
   // If this field is set and `entities_revisions` is not empty, the entities in
   // `document.entities` field are the entities in the entity revision with this
-  // id and `document.entity_validation_output` field is the
+  // ID and `document.entity_validation_output` field is the
   // `entity_validation_output` field in this entity revision.
   string entities_revision_id = 23;
 }
@@ -1345,7 +1347,7 @@ message RevisionRef {
     // Reads the revision by the predefined case.
     RevisionCase revision_case = 1;
 
-    // Reads the revision given by the id.
+    // Reads the revision given by the ID.
     string revision_id = 2;
 
     // Reads the revision generated by the processor version.
diff --git a/google/cloud/documentai/v1beta3/document_processor_service.proto b/google/cloud/documentai/v1beta3/document_processor_service.proto
@@ -359,13 +359,15 @@ message ProcessOptions {
           [(google.api.field_behavior) = OPTIONAL];
 
       // Optional. The number of tokens to group together when evaluating
-      // semantic similarity. THIS FIELD IS NOT YET USED.
+      // semantic similarity.
+      // **Note:** This field is not yet used.
       bool semantic_chunking_group_size = 3
           [(google.api.field_behavior) = OPTIONAL];
 
       // Optional. The percentile of cosine dissimilarity that must be exceeded
       // between a group of tokens and the next. The smaller this number is, the
-      // more chunks will be generated. THIS FIELD IS NOT YET USED.
+      // more chunks will be generated.
+      // **Note:** This field is not yet used.
       int32 breakpoint_percentile_threshold = 4
           [(google.api.field_behavior) = OPTIONAL];
     }
