public async Task RecognizeCustomFormsFromUri()
{
string endpoint = TestEnvironment.Endpoint;
string apiKey = TestEnvironment.ApiKey;
FormRecognizerClient client = new FormRecognizerClient(new Uri(endpoint), new AzureKeyCredential(apiKey));
string formUri = FormRecognizerTestEnvironment.CreateUri("Form_1.jpg");
string modelId = "<your model id>";
#region Snippet:FormRecognizerSample3RecognizeCustomFormsFromUri
Response <IReadOnlyList <RecognizedForm> > forms = await client.StartRecognizeCustomFormsFromUri(modelId, new Uri(formUri)).WaitForCompletionAsync();
foreach (RecognizedForm form in forms.Value)
{
Console.WriteLine($"Form of type: {form.FormType}");
foreach (FormField field in form.Fields.Values)
{
Console.WriteLine($"Field '{field.Name}: ");
if (field.LabelText != null)
{
Console.WriteLine($" Label: '{field.LabelText.Text}");
}
Console.WriteLine($" Value: '{field.ValueText.Text}");
Console.WriteLine($" Confidence: '{field.Confidence}");
}
}
#endregion
}
// </snippet_trainlabels_response>
// <snippet_analyze>
// Analyze PDF form data
private static async Task AnalyzePdfForm(
FormRecognizerClient recognizerClient, Guid modelId, string formUrl)
{
Response<IReadOnlyList<RecognizedForm>> forms = await recognizerClient
.StartRecognizeCustomFormsFromUri(modelId.ToString(), new Uri(formUrl))
.WaitForCompletionAsync();
// </snippet_analyze>
// <snippet_analyze_response>
foreach (RecognizedForm form in forms.Value)
{
Console.WriteLine($"Form of type: {form.FormType}");
foreach (FormField field in form.Fields.Values)
{
Console.WriteLine($"Field '{field.Name}: ");
if (field.LabelText != null)
{
Console.WriteLine($" Label: '{field.LabelText.Text}");
}
Console.WriteLine($" Value: '{field.ValueText.Text}");
Console.WriteLine($" Confidence: '{field.Confidence}");
}
}
}
public override void Run(CancellationToken cancellationToken)
{
var op = _client.StartRecognizeCustomFormsFromUri(_modelId, CreateUri(TestFile.Form1), cancellationToken: cancellationToken);
while (!op.HasCompleted)
{
Thread.Sleep(TimeSpan.FromSeconds(1));
op.UpdateStatus(cancellationToken);
}
}
public async Task RecognizeCustomFormsFromUri()
{
string endpoint = TestEnvironment.Endpoint;
string apiKey = TestEnvironment.ApiKey;
string trainingFileUrl = TestEnvironment.BlobContainerSasUrl;
// Firstly, create a trained model we can use to recognize the custom form. Please note that
// models can also be trained using a graphical user interface such as the Form Recognizer
// Labeling Tool found here:
// https://docs.microsoft.com/azure/cognitive-services/form-recognizer/quickstarts/label-tool
FormTrainingClient trainingClient = new FormTrainingClient(new Uri(endpoint), new AzureKeyCredential(apiKey));
CustomFormModel model = await trainingClient.StartTraining(new Uri(trainingFileUrl), useTrainingLabels : false).WaitForCompletionAsync();
// Proceed with the custom form recognition.
FormRecognizerClient client = new FormRecognizerClient(new Uri(endpoint), new AzureKeyCredential(apiKey));
Uri formUri = FormRecognizerTestEnvironment.CreateUri("Form_1.jpg");
string modelId = model.ModelId;
#region Snippet:FormRecognizerSample3RecognizeCustomFormsFromUri
//@@ string modelId = "<modelId>";
RecognizedFormCollection forms = await client.StartRecognizeCustomFormsFromUri(modelId, formUri).WaitForCompletionAsync();
foreach (RecognizedForm form in forms)
{
Console.WriteLine($"Form of type: {form.FormType}");
foreach (FormField field in form.Fields.Values)
{
Console.WriteLine($"Field '{field.Name}: ");
if (field.LabelData != null)
{
Console.WriteLine($" Label: '{field.LabelData.Text}");
}
Console.WriteLine($" Value: '{field.ValueData.Text}");
Console.WriteLine($" Confidence: '{field.Confidence}");
}
}
#endregion
// Delete the model on completion to clean environment.
trainingClient.DeleteModel(model.ModelId);
}
// </snippet_trainlabels_response>
// <snippet_analyze>
// Analyze PDF form data
private static async Task AnalyzePdfForm(
FormRecognizerClient recognizerClient, String modelId, string formUrl)
{
RecognizedFormCollection forms = await recognizerClient
.StartRecognizeCustomFormsFromUri(modelId, new Uri(formUrl))
.WaitForCompletionAsync();
// </snippet_analyze>
// <snippet_analyze_response>
foreach (RecognizedForm form in forms)
{
Console.WriteLine($"Form of type: {form.FormType}");
foreach (FormField field in form.Fields.Values)
{
Console.WriteLine($"Field '{field.Name}: ");
if (field.LabelData != null)
{
Console.WriteLine($" Label: '{field.LabelData.Text}");
}
Console.WriteLine($" Value: '{field.ValueData.Text}");
Console.WriteLine($" Confidence: '{field.Confidence}");
}
Console.WriteLine("Table data:");
foreach (FormPage page in form.Pages)
{
for (int i = 0; i < page.Tables.Count; i++)
{
FormTable table = page.Tables[i];
Console.WriteLine($"Table {i} has {table.RowCount} rows and {table.ColumnCount} columns.");
foreach (FormTableCell cell in table.Cells)
{
Console.WriteLine($" Cell ({cell.RowIndex}, {cell.ColumnIndex}) contains {(cell.IsHeader ? "header" : "text")}: '{cell.Text}'");
}
}
}
}
}
// Analyze PDF form data
private static async Task AnalyzePdfForm(
FormRecognizerClient formClient, string modelId, string pdfFormFile)
{
Response <RecognizedFormCollection> forms = await formClient
.StartRecognizeCustomFormsFromUri(modelId.ToString(), new Uri(pdfFormFile))
.WaitForCompletionAsync();
foreach (RecognizedForm form in forms.Value)
{
Console.WriteLine($"Form of type: {form.FormType}");
foreach (FormField field in form.Fields.Values)
{
Console.WriteLine($"Field '{field.Name}: ");
if (field.LabelData != null)
{
Console.WriteLine($" Label: '{field.LabelData.Text}");
}
Console.WriteLine($" Value: '{field.ValueData.Text}");
Console.WriteLine($" Confidence: '{field.Confidence}");
}
}
}
public static async Task RunAsync([EventGridTrigger] EventGridEvent eventGridEvent, ILogger log)
{
//Extracting content type and url of the blob triggering the function
var jsondata = JsonConvert.SerializeObject(eventGridEvent.Data);
var tmp = new { contentType = "", url = "" };
var data = JsonConvert.DeserializeAnonymousType(jsondata, tmp);
//Checking if the trigger was iniatiated for a PDF File.
if (data.contentType == "application/pdf")
{
var pdfUrl = data.url;
string endpoint = System.Environment.GetEnvironmentVariable("FormsRecognizerEndpoint", EnvironmentVariableTarget.Process);
string apiKey = System.Environment.GetEnvironmentVariable("FormsRecognizerKey", EnvironmentVariableTarget.Process);
string contosoStorageConnectionString = System.Environment.GetEnvironmentVariable("ContosoStorageConnectionString", EnvironmentVariableTarget.Process);
string cosmosEndpointUrl = System.Environment.GetEnvironmentVariable("CosmosDBEndpointUrl", EnvironmentVariableTarget.Process);
string cosmosPrimaryKey = System.Environment.GetEnvironmentVariable("CosmosDBPrimaryKey", EnvironmentVariableTarget.Process);
//Create a SAS Link to give Forms Recognizer read access to the document
BlobServiceClient blobServiceClient = new BlobServiceClient(contosoStorageConnectionString);
BlobContainerClient container = new BlobContainerClient(contosoStorageConnectionString, "claims");
string blobName = pdfUrl.Split('/').Last();
BlobClient blob = container.GetBlobClient(blobName);
BlobSasBuilder sasBuilder = new BlobSasBuilder()
{
BlobContainerName = blob.GetParentBlobContainerClient().Name,
BlobName = blob.Name,
Resource = "b"
};
sasBuilder.ExpiresOn = DateTimeOffset.UtcNow.AddHours(1);
sasBuilder.SetPermissions(BlobSasPermissions.Read);
Uri sasUri = blob.GenerateSasUri(sasBuilder);
var credential = new AzureKeyCredential(apiKey);
//Get the latest trained model
var formTrainingClient = new FormTrainingClient(new Uri(endpoint), credential);
Pageable <CustomFormModelInfo> formsModels = formTrainingClient.GetCustomModels();
var latestModel = (from inc in formsModels orderby inc.TrainingCompletedOn descending select inc).FirstOrDefault();
//Run the document through the model
var formRecognizerClient = new FormRecognizerClient(new Uri(endpoint), credential);
var options = new RecognizeCustomFormsOptions()
{
IncludeFieldElements = true
};
RecognizeCustomFormsOperation operation = formRecognizerClient.StartRecognizeCustomFormsFromUri(latestModel.ModelId, sasUri, options);
Azure.Response <RecognizedFormCollection> operationResponse = await operation.WaitForCompletionAsync();
RecognizedFormCollection forms = operationResponse.Value;
//Insert documents into CosmosDB
var cosmosClient = new CosmosClient(cosmosEndpointUrl, cosmosPrimaryKey);
var cosmosDatabase = (await cosmosClient.CreateDatabaseIfNotExistsAsync("Contoso")).Database;
var cosmosContainer = (await cosmosDatabase.CreateContainerIfNotExistsAsync("Claims", "/InsuredID")).Container;
Model.ClaimsDocument processedDocument = new Model.ClaimsDocument();
processedDocument.DocumentDate = new DateTime(int.Parse(blobName.Substring(0, 4)),
int.Parse(blobName.Substring(4, 2)), int.Parse(blobName.Substring(6, 2)));
processedDocument.PatientName = forms[0].Fields["PatientName"].ValueData?.Text;
processedDocument.InsuredID = forms[0].Fields["InsuredID"].ValueData?.Text;
processedDocument.Diagnosis = forms[0].Fields["Diagnosis"].ValueData?.Text;
decimal.TryParse(forms[0].Fields["TotalCharges"].ValueData?.Text, out decimal totalCharges);
processedDocument.TotalCharges = totalCharges;
DateTime.TryParse(forms[0].Fields["PatientBirthDate"].ValueData?.Text, out DateTime patientBirthDate);
processedDocument.PatientBirthDate = patientBirthDate;
decimal.TryParse(forms[0].Fields["AmountPaid"].ValueData?.Text, out decimal amountPaid);
processedDocument.AmountPaid = amountPaid;
decimal.TryParse(forms[0].Fields["AmountDue"].ValueData?.Text, out decimal amountDue);
processedDocument.AmountDue = amountDue;
processedDocument.FileName = blobName;
if (processedDocument.InsuredID == null)
{
processedDocument.Id = Guid.NewGuid().ToString();
}
else
{
processedDocument.Id = processedDocument.InsuredID;
}
try
{
ItemResponse <Model.ClaimsDocument> cosmosResponse = await
cosmosContainer.CreateItemAsync(processedDocument, new PartitionKey(processedDocument.InsuredID));
}
catch (CosmosException ex) when(ex.StatusCode == System.Net.HttpStatusCode.Conflict)
{
//Conflicting EnsurerID is silently ignored for demo purposes.
}
}
log.LogInformation(eventGridEvent.Data.ToString());
}
