[FR] Expose code generated properties

sdk/formrecognizer/Azure.AI.FormRecognizer/CHANGELOG.md

@@ -3,20 +3,24 @@
 ## 3.1.0-beta.1 (Unreleased)
 
 ### Breaking changes
-- It defaults to the latest supported API version, which currently is `2.1-preview.1`.
+- It defaults to the latest supported API version, which currently is `2.1-preview.2`.
 
 ### New Features
 - Added support for pre-built business card recognition.
 - Added support for pre-built invoices recognition.
 - Added support for providing locale info when recognizing receipts and business cards. Supported locales include support EN-US, EN-AU, EN-CA, EN-GB, EN-IN.
 - Added support to train and recognize custom forms with selection marks such as check boxes and radio buttons. This functionality is only available in train with labels scenarios.
+- Added support to `StartRecognizeContent` to recognize selection marks such as check boxes and radio buttons.
 - Added ability to create a composed model from the `FormTrainingClient` by calling method `StartCreateComposedModel`.
 - Added ability to pass parameter `ModelName` to `StartTraining` methods.
 - Added the properties `ModelName` and `Properties` to types `CustomFormModel` and `CustomFormModelInfo`.
 - Added type `CustomFormModelProperties` that includes information like if a model is a composed model.
 - Added property `ModelId` to `CustomFormSubmodel` and `TrainingDocumentInfo`.
 - Added properties `ModelId` and `FormTypeConfidence` to `RecognizedForm`.
-- Added support to `StartRecognizeContent` to recognize selection marks such as check boxes and radio buttons.
+- Added property `Appearance` to `FormLine` to indicate the style of the extracted text. for example, "handwriting" or "other".
+- Added property `BoundingBox` to `FormTable`.
+- Added support for `ContentType` `image/bmp` in recognize content and prebuilt models.
+- Added property `Pages` to `RecognizeContentOptions` to specify the page numbers to analyze.
 
 ## 3.0.0 (2020-08-20)
 

sdk/formrecognizer/Azure.AI.FormRecognizer/api/Azure.AI.FormRecognizer.netstandard2.0.cs

@@ -1,5 +1,10 @@
 namespace Azure.AI.FormRecognizer
 {
+    public partial class Appearance
+    {
+        internal Appearance() { }
+        public Azure.AI.FormRecognizer.Style Style { get { throw null; } }
+    }
     public enum FormContentType
     {
         Json = 0,
@@ -68,6 +73,7 @@ public partial class RecognizeContentOptions
     {
         public RecognizeContentOptions() { }
         public Azure.AI.FormRecognizer.FormContentType? ContentType { get { throw null; } set { } }
+        public System.Collections.Generic.IEnumerable<string> Pages { get { throw null; } set { } }
     }
     public partial class RecognizeCustomFormsOptions
     {
@@ -89,6 +95,12 @@ public RecognizeReceiptsOptions() { }
         public bool IncludeFieldElements { get { throw null; } set { } }
         public string Locale { get { throw null; } set { } }
     }
+    public partial class Style
+    {
+        internal Style() { }
+        public float Confidence { get { throw null; } }
+        public Azure.AI.FormRecognizer.Models.TextStyle Name { get { throw null; } }
+    }
 }
 namespace Azure.AI.FormRecognizer.Models
 {
@@ -163,6 +175,7 @@ public FormField(Azure.AI.FormRecognizer.Models.FormField field, T value) { }
     public partial class FormLine : Azure.AI.FormRecognizer.Models.FormElement
     {
         internal FormLine() { }
+        public Azure.AI.FormRecognizer.Appearance Appearance { get { throw null; } }
         public System.Collections.Generic.IReadOnlyList<Azure.AI.FormRecognizer.Models.FormWord> Words { get { throw null; } }
     }
     public partial class FormPage
@@ -248,6 +261,7 @@ internal FormSelectionMark() { }
     public partial class FormTable
     {
         internal FormTable() { }
+        public Azure.AI.FormRecognizer.Models.FieldBoundingBox BoundingBox { get { throw null; } }
         public System.Collections.Generic.IReadOnlyList<Azure.AI.FormRecognizer.Models.FormTableCell> Cells { get { throw null; } }
         public int ColumnCount { get { throw null; } }
         public int PageNumber { get { throw null; } }
@@ -356,8 +370,26 @@ public RecognizeReceiptsOperation(string operationId, Azure.AI.FormRecognizer.Fo
     }
     public enum SelectionMarkState
     {
-        Selected = 0,
-        Unselected = 1,
+        Unselected = 0,
+        Selected = 1,
+    }
+    [System.Runtime.InteropServices.StructLayoutAttribute(System.Runtime.InteropServices.LayoutKind.Sequential)]
+    public readonly partial struct TextStyle : System.IEquatable<Azure.AI.FormRecognizer.Models.TextStyle>
+    {
+        private readonly object _dummy;
+        private readonly int _dummyPrimitive;
+        public TextStyle(string value) { throw null; }
+        public static Azure.AI.FormRecognizer.Models.TextStyle Handwriting { get { throw null; } }
+        public static Azure.AI.FormRecognizer.Models.TextStyle Other { get { throw null; } }
+        public bool Equals(Azure.AI.FormRecognizer.Models.TextStyle other) { throw null; }
+        [System.ComponentModel.EditorBrowsableAttribute(System.ComponentModel.EditorBrowsableState.Never)]
+        public override bool Equals(object obj) { throw null; }
+        [System.ComponentModel.EditorBrowsableAttribute(System.ComponentModel.EditorBrowsableState.Never)]
+        public override int GetHashCode() { throw null; }
+        public static bool operator ==(Azure.AI.FormRecognizer.Models.TextStyle left, Azure.AI.FormRecognizer.Models.TextStyle right) { throw null; }
+        public static implicit operator Azure.AI.FormRecognizer.Models.TextStyle (string value) { throw null; }
+        public static bool operator !=(Azure.AI.FormRecognizer.Models.TextStyle left, Azure.AI.FormRecognizer.Models.TextStyle right) { throw null; }
+        public override string ToString() { throw null; }
     }
 }
 namespace Azure.AI.FormRecognizer.Training

sdk/formrecognizer/Azure.AI.FormRecognizer/src/Appearance.cs

@@ -0,0 +1,15 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+using System.Diagnostics.CodeAnalysis;
+using Azure.Core;
+
+namespace Azure.AI.FormRecognizer
+{
+    /// <inheritdoc />
+    [CodeGenModel("Appearance")]
+    [SuppressMessage("Usage", "AZC0012:Avoid single word type names", Justification = "Will fix name in future release")]
+    public partial class Appearance
+    {
+    }
+}

sdk/formrecognizer/Azure.AI.FormRecognizer/src/FormLine.cs

@@ -14,6 +14,7 @@ internal FormLine(TextLine textLine, int pageNumber)
             : base(new FieldBoundingBox(textLine.BoundingBox), pageNumber, textLine.Text)
         {
             Words = ConvertWords(textLine.Words, pageNumber);
+            Appearance = textLine.Appearance;
         }
 
         /// <summary>
@@ -45,5 +46,8 @@ private static IReadOnlyList<FormWord> ConvertWords(IReadOnlyList<TextWord> text
 
             return rawWords;
         }
+
+        /// <summary> Text appearance properties. </summary>
+        public Appearance Appearance { get; }
     }
 }

sdk/formrecognizer/Azure.AI.FormRecognizer/src/FormRecognizerClient.cs

@@ -137,7 +137,12 @@ public virtual RecognizeContentOperation StartRecognizeContent(Stream form, Reco
             {
                 FormContentType formContentType = recognizeContentOptions.ContentType ?? DetectContentType(form, nameof(form));
 
-                Response response = ServiceClient.AnalyzeLayoutAsync(formContentType.ConvertToContentType1(), form, null, null, cancellationToken);
+                Response response = ServiceClient.AnalyzeLayoutAsync(
+                    formContentType.ConvertToContentType1(),
+                    form,
+                    null,
+                    recognizeContentOptions.Pages,
+                    cancellationToken);
                 string location = ClientCommon.GetResponseHeader(response.Headers, Constants.OperationLocationHeader);
 
                 return new RecognizeContentOperation(ServiceClient, Diagnostics, location);
@@ -170,7 +175,12 @@ public virtual async Task<RecognizeContentOperation> StartRecognizeContentAsync(
             {
                 FormContentType formContentType = recognizeContentOptions.ContentType ?? DetectContentType(form, nameof(form));
 
-                Response response = await ServiceClient.AnalyzeLayoutAsyncAsync(formContentType.ConvertToContentType1(), form, null, null, cancellationToken).ConfigureAwait(false);
+                Response response = await ServiceClient.AnalyzeLayoutAsyncAsync(
+                    formContentType.ConvertToContentType1(),
+                    form,
+                    null,
+                    recognizeContentOptions.Pages,
+                    cancellationToken).ConfigureAwait(false);
                 string location = ClientCommon.GetResponseHeader(response.Headers, Constants.OperationLocationHeader);
 
                 return new RecognizeContentOperation(ServiceClient, Diagnostics, location);
@@ -202,7 +212,11 @@ public virtual RecognizeContentOperation StartRecognizeContentFromUri(Uri formUr
             try
             {
                 SourcePath sourcePath = new SourcePath() { Source = formUri.AbsoluteUri };
-                Response response = ServiceClient.AnalyzeLayoutAsync(null, null, sourcePath, cancellationToken);
+                Response response = ServiceClient.AnalyzeLayoutAsync(
+                    null,
+                    recognizeContentOptions.Pages,
+                    sourcePath,
+                    cancellationToken);
                 string location = ClientCommon.GetResponseHeader(response.Headers, Constants.OperationLocationHeader);
 
                 return new RecognizeContentOperation(ServiceClient, Diagnostics, location);
@@ -234,7 +248,11 @@ public virtual async Task<RecognizeContentOperation> StartRecognizeContentFromUr
             try
             {
                 SourcePath sourcePath = new SourcePath() { Source = formUri.AbsoluteUri };
-                Response response = await ServiceClient.AnalyzeLayoutAsyncAsync(null, null, sourcePath, cancellationToken).ConfigureAwait(false);
+                Response response = await ServiceClient.AnalyzeLayoutAsyncAsync(
+                    null,
+                    recognizeContentOptions.Pages,
+                    sourcePath,
+                    cancellationToken).ConfigureAwait(false);
                 string location = ClientCommon.GetResponseHeader(response.Headers, Constants.OperationLocationHeader);
 
                 return new RecognizeContentOperation(ServiceClient, Diagnostics, location);

sdk/formrecognizer/Azure.AI.FormRecognizer/src/FormTable.cs

@@ -19,6 +19,9 @@ internal FormTable(DataTable table, IReadOnlyList<ReadResult> readResults, int p
             ColumnCount = table.Columns;
             RowCount = table.Rows;
             Cells = ConvertCells(table.Cells, readResults, readResult.Page);
+            // Need to verify why Bounding Box is not returned from the service
+            // https://github.com/Azure/azure-sdk-for-net/issues/16827
+            BoundingBox = table.BoundingBox == null ? new FieldBoundingBox(new List<float>()) : new FieldBoundingBox(table.BoundingBox);
         }
 
         /// <summary>
@@ -56,6 +59,9 @@ internal FormTable(int pageNumber, int columnCount, int rowCount, IReadOnlyList<
         /// </summary>
         public int RowCount { get; }
 
+        /// <summary> Bounding box of the table. </summary>
+        public FieldBoundingBox BoundingBox { get; }
+
         // TODO: implement table indexer
         // TODO: Handling column-span?
         // https://github.com/Azure/azure-sdk-for-net/issues/9975

sdk/formrecognizer/Azure.AI.FormRecognizer/src/RecognizeContentOptions.cs

@@ -1,6 +1,8 @@
 // Copyright (c) Microsoft Corporation. All rights reserved.
 // Licensed under the MIT License.
 
+using System.Collections.Generic;
+
 namespace Azure.AI.FormRecognizer
 {
     /// <summary>
@@ -21,5 +23,12 @@ public RecognizeContentOptions()
         /// content type detection.
         /// </summary>
         public FormContentType? ContentType { get; set; } = null;
+
+        /// <summary>
+        /// Custom page numbers for multi-page documents(PDF/TIFF). Input the number of the
+        /// pages you want to get OCR result. For a range of pages, use a hyphen.
+        /// Separate each page or range with a comma.
+        /// </summary>
+        public IEnumerable<string> Pages { get; set; }
     }
 }

sdk/formrecognizer/Azure.AI.FormRecognizer/src/SelectionMarkState.cs

@@ -12,13 +12,13 @@ namespace Azure.AI.FormRecognizer.Models
     public enum SelectionMarkState
     {
         /// <summary>
-        /// Value is selected.
+        /// Value is unselected.
         /// </summary>
-        Selected,
+        Unselected,
 
         /// <summary>
-        /// Value is unselected.
+        /// Value is selected.
         /// </summary>
-        Unselected
+        Selected
     }
 }

sdk/formrecognizer/Azure.AI.FormRecognizer/src/StreamExtensions.cs

@@ -24,6 +24,9 @@ internal static class StreamExtensions
         /// <summary>The set of bytes expected to be present at the start of TIFF (big-endian) files.</summary>
         private static byte[] TiffBeHeader = new byte[] { 0x4D, 0x4D, 0x00, 0x2A };
 
+        /// <summary>The set of bytes expected to be present at the start of BMP files.</summary>
+        private static byte[] BmpHeader = new byte[] { 0x42, 0x4D };
+
         /// <summary>
         /// Attemps to detect the <see cref="FormContentType"/> of a stream of bytes. The algorithm searches through
         /// the first set of bytes in the stream and compares it to well-known file signatures.
@@ -50,6 +53,10 @@ public static bool TryGetContentType(this Stream stream, out FormContentType con
             {
                 contentType = FormContentType.Tiff;
             }
+            else if (stream.BeginsWithHeader(BmpHeader))
+            {
+                contentType = FormContentType.Bmp;
+            }
             else
             {
                 contentType = default;

sdk/formrecognizer/Azure.AI.FormRecognizer/src/Style.cs

@@ -0,0 +1,15 @@
+// Copyright (c) Microsoft Corporation. All rights reserved.
+// Licensed under the MIT License.
+
+using System.Diagnostics.CodeAnalysis;
+using Azure.Core;
+
+namespace Azure.AI.FormRecognizer
+{
+    /// <inheritdoc />
+    [CodeGenModel("Style")]
+    [SuppressMessage("Usage", "AZC0012:Avoid single word type names", Justification = "Will fix name in future release")]
+    public partial class Style
+    {
+    }
+}

sdk/formrecognizer/Azure.AI.FormRecognizer/src/TextStyle.cs

@@ -1,11 +1,13 @@
 // Copyright (c) Microsoft Corporation. All rights reserved.
 // Licensed under the MIT License.
 
+using System.Diagnostics.CodeAnalysis;
 using Azure.Core;
 
 namespace Azure.AI.FormRecognizer.Models
 {
     [CodeGenModel("TextStyle")]
-    internal partial struct TextStyle
+    [SuppressMessage("Usage", "AZC0012:Avoid single word type names", Justification = "Will fix name in future release")]
+    public partial struct TextStyle
     { }
 }