2021/10/27 - Amazon Textract - 2 new api methods
Changes This release adds support for asynchronously analyzing invoice and receipt documents through two new APIs: StartExpenseAnalysis and GetExpenseAnalysis
Starts the asynchronous analysis of invoices or receipts for data like contact information, items purchased, and vendor names.
StartExpenseAnalysis can analyze text in documents that are in JPEG, PNG, and PDF format. The documents must be stored in an Amazon S3 bucket. Use the DocumentLocation parameter to specify the name of your S3 bucket and the name of the document in that bucket.
StartExpenseAnalysis returns a job identifier ( JobId) that you will provide to GetExpenseAnalysis to retrieve the results of the operation. When the analysis of the input invoices/receipts is finished, Amazon Textract publishes a completion status to the Amazon Simple Notification Service (Amazon SNS) topic that you provide to the NotificationChannel. To obtain the results of the invoice and receipt analysis operation, ensure that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetExpenseAnalysis, and pass the job identifier ( JobId) that was returned by your call to StartExpenseAnalysis.
For more information, see Analyzing Invoices and Receipts.
See also: AWS API Documentation
Request Syntax
client.start_expense_analysis(
DocumentLocation={
'S3Object': {
'Bucket': 'string',
'Name': 'string',
'Version': 'string'
}
},
ClientRequestToken='string',
JobTag='string',
NotificationChannel={
'SNSTopicArn': 'string',
'RoleArn': 'string'
},
OutputConfig={
'S3Bucket': 'string',
'S3Prefix': 'string'
},
KMSKeyId='string'
)
dict
[REQUIRED]
The location of the document to be processed.
S3Object (dict) --
The Amazon S3 bucket that contains the input document.
Bucket (string) --
The name of the S3 bucket. Note that the # character is not valid in the file name.
Name (string) --
The file name of the input document. Synchronous operations can use image files that are in JPEG or PNG format. Asynchronous operations also support PDF and TIFF format files.
Version (string) --
If the bucket has versioning enabled, you can specify the object version.
string
The idempotent token that's used to identify the start request. If you use the same token with multiple StartDocumentTextDetection requests, the same JobId is returned. Use ClientRequestToken to prevent the same job from being accidentally started more than once. For more information, see Calling Amazon Textract Asynchronous Operations
string
An identifier you specify that's included in the completion notification published to the Amazon SNS topic. For example, you can use JobTag to identify the type of document that the completion notification corresponds to (such as a tax form or a receipt).
dict
The Amazon SNS topic ARN that you want Amazon Textract to publish the completion status of the operation to.
SNSTopicArn (string) -- [REQUIRED]
The Amazon SNS topic that Amazon Textract posts the completion status to.
RoleArn (string) -- [REQUIRED]
The Amazon Resource Name (ARN) of an IAM role that gives Amazon Textract publishing permissions to the Amazon SNS topic.
dict
Sets if the output will go to a customer defined bucket. By default, Amazon Textract will save the results internally to be accessed by the GetExpenseAnalysis operation.
S3Bucket (string) -- [REQUIRED]
The name of the bucket your output will go to.
S3Prefix (string) --
The prefix of the object key that the output will be saved to. When not enabled, the prefix will be “textract_output".
string
The KMS key used to encrypt the inference results. This can be in either Key ID or Key Alias format. When a KMS key is provided, the KMS key will be used for server-side encryption of the objects in the customer bucket. When this parameter is not enabled, the result will be encrypted server side,using SSE-S3.
dict
Response Syntax
{
'JobId': 'string'
}
Response Structure
(dict) --
JobId (string) --
A unique identifier for the text detection job. The JobId is returned from StartExpenseAnalysis. A JobId value is only valid for 7 days.
Gets the results for an Amazon Textract asynchronous operation that analyzes invoices and receipts. Amazon Textract finds contact information, items purchased, and vendor name, from input invoices and receipts.
You start asynchronous invoice/receipt analysis by calling StartExpenseAnalysis, which returns a job identifier ( JobId). Upon completion of the invoice/receipt analysis, Amazon Textract publishes the completion status to the Amazon Simple Notification Service (Amazon SNS) topic. This topic must be registered in the initial call to StartExpenseAnalysis. To get the results of the invoice/receipt analysis operation, first ensure that the status value published to the Amazon SNS topic is SUCCEEDED. If so, call GetExpenseAnalysis, and pass the job identifier ( JobId) from the initial call to StartExpenseAnalysis.
Use the MaxResults parameter to limit the number of blocks that are returned. If there are more results than specified in MaxResults, the value of NextToken in the operation response contains a pagination token for getting the next set of results. To get the next page of results, call GetExpenseAnalysis, and populate the NextToken request parameter with the token value that's returned from the previous call to GetExpenseAnalysis.
For more information, see Analyzing Invoices and Receipts.
See also: AWS API Documentation
Request Syntax
client.get_expense_analysis(
JobId='string',
MaxResults=123,
NextToken='string'
)
string
[REQUIRED]
A unique identifier for the text detection job. The JobId is returned from StartExpenseAnalysis. A JobId value is only valid for 7 days.
integer
The maximum number of results to return per paginated call. The largest value you can specify is 20. If you specify a value greater than 20, a maximum of 20 results is returned. The default value is 20.
string
If the previous response was incomplete (because there are more blocks to retrieve), Amazon Textract returns a pagination token in the response. You can use this pagination token to retrieve the next set of blocks.
dict
Response Syntax
{
'DocumentMetadata': {
'Pages': 123
},
'JobStatus': 'IN_PROGRESS'|'SUCCEEDED'|'FAILED'|'PARTIAL_SUCCESS',
'NextToken': 'string',
'ExpenseDocuments': [
{
'ExpenseIndex': 123,
'SummaryFields': [
{
'Type': {
'Text': 'string',
'Confidence': ...
},
'LabelDetection': {
'Text': 'string',
'Geometry': {
'BoundingBox': {
'Width': ...,
'Height': ...,
'Left': ...,
'Top': ...
},
'Polygon': [
{
'X': ...,
'Y': ...
},
]
},
'Confidence': ...
},
'ValueDetection': {
'Text': 'string',
'Geometry': {
'BoundingBox': {
'Width': ...,
'Height': ...,
'Left': ...,
'Top': ...
},
'Polygon': [
{
'X': ...,
'Y': ...
},
]
},
'Confidence': ...
},
'PageNumber': 123
},
],
'LineItemGroups': [
{
'LineItemGroupIndex': 123,
'LineItems': [
{
'LineItemExpenseFields': [
{
'Type': {
'Text': 'string',
'Confidence': ...
},
'LabelDetection': {
'Text': 'string',
'Geometry': {
'BoundingBox': {
'Width': ...,
'Height': ...,
'Left': ...,
'Top': ...
},
'Polygon': [
{
'X': ...,
'Y': ...
},
]
},
'Confidence': ...
},
'ValueDetection': {
'Text': 'string',
'Geometry': {
'BoundingBox': {
'Width': ...,
'Height': ...,
'Left': ...,
'Top': ...
},
'Polygon': [
{
'X': ...,
'Y': ...
},
]
},
'Confidence': ...
},
'PageNumber': 123
},
]
},
]
},
]
},
],
'Warnings': [
{
'ErrorCode': 'string',
'Pages': [
123,
]
},
],
'StatusMessage': 'string',
'AnalyzeExpenseModelVersion': 'string'
}
Response Structure
(dict) --
DocumentMetadata (dict) --
Information about a document that Amazon Textract processed. DocumentMetadata is returned in every page of paginated responses from an Amazon Textract operation.
Pages (integer) --
The number of pages that are detected in the document.
JobStatus (string) --
The current status of the text detection job.
NextToken (string) --
If the response is truncated, Amazon Textract returns this token. You can use this token in the subsequent request to retrieve the next set of text-detection results.
ExpenseDocuments (list) --
The expenses detected by Amazon Textract.
(dict) --
The structure holding all the information returned by AnalyzeExpense
ExpenseIndex (integer) --
Denotes which invoice or receipt in the document the information is coming from. First document will be 1, the second 2, and so on.
SummaryFields (list) --
Any information found outside of a table by Amazon Textract.
(dict) --
Breakdown of detected information, seperated into the catagories Type, LableDetection, and ValueDetection
Type (dict) --
The implied label of a detected element. Present alongside LabelDetection for explicit elements.
Text (string) --
The word or line of text detected by Amazon Textract.
Confidence (float) --
The confidence of accuracy, as a percentage.
LabelDetection (dict) --
The explicitly stated label of a detected element.
Text (string) --
The word or line of text recognized by Amazon Textract
Geometry (dict) --
Information about where the following items are located on a document page: detected page, text, key-value pairs, tables, table cells, and selection elements.
BoundingBox (dict) --
An axis-aligned coarse representation of the location of the recognized item on the document page.
Width (float) --
The width of the bounding box as a ratio of the overall document page width.
Height (float) --
The height of the bounding box as a ratio of the overall document page height.
Left (float) --
The left coordinate of the bounding box as a ratio of overall document page width.
Top (float) --
The top coordinate of the bounding box as a ratio of overall document page height.
Polygon (list) --
Within the bounding box, a fine-grained polygon around the recognized item.
(dict) --
The X and Y coordinates of a point on a document page. The X and Y values that are returned are ratios of the overall document page size. For example, if the input document is 700 x 200 and the operation returns X=0.5 and Y=0.25, then the point is at the (350,50) pixel coordinate on the document page.
An array of Point objects, Polygon, is returned by DetectDocumentText. Polygon represents a fine-grained polygon around detected text. For more information, see Geometry in the Amazon Textract Developer Guide.
X (float) --
The value of the X coordinate for a point on a Polygon.
Y (float) --
The value of the Y coordinate for a point on a Polygon.
Confidence (float) --
The confidence in detection, as a percentage
ValueDetection (dict) --
The value of a detected element. Present in explicit and implicit elements.
Text (string) --
The word or line of text recognized by Amazon Textract
Geometry (dict) --
Information about where the following items are located on a document page: detected page, text, key-value pairs, tables, table cells, and selection elements.
BoundingBox (dict) --
An axis-aligned coarse representation of the location of the recognized item on the document page.
Width (float) --
The width of the bounding box as a ratio of the overall document page width.
Height (float) --
The height of the bounding box as a ratio of the overall document page height.
Left (float) --
The left coordinate of the bounding box as a ratio of overall document page width.
Top (float) --
The top coordinate of the bounding box as a ratio of overall document page height.
Polygon (list) --
Within the bounding box, a fine-grained polygon around the recognized item.
(dict) --
The X and Y coordinates of a point on a document page. The X and Y values that are returned are ratios of the overall document page size. For example, if the input document is 700 x 200 and the operation returns X=0.5 and Y=0.25, then the point is at the (350,50) pixel coordinate on the document page.
An array of Point objects, Polygon, is returned by DetectDocumentText. Polygon represents a fine-grained polygon around detected text. For more information, see Geometry in the Amazon Textract Developer Guide.
X (float) --
The value of the X coordinate for a point on a Polygon.
Y (float) --
The value of the Y coordinate for a point on a Polygon.
Confidence (float) --
The confidence in detection, as a percentage
PageNumber (integer) --
The page number the value was detected on.
LineItemGroups (list) --
Information detected on each table of a document, seperated into LineItems.
(dict) --
A grouping of tables which contain LineItems, with each table identified by the table's LineItemGroupIndex.
LineItemGroupIndex (integer) --
The number used to identify a specific table in a document. The first table encountered will have a LineItemGroupIndex of 1, the second 2, etc.
LineItems (list) --
The breakdown of information on a particular line of a table.
(dict) --
A structure that holds information about the different lines found in a document's tables.
LineItemExpenseFields (list) --
ExpenseFields used to show information from detected lines on a table.
(dict) --
Breakdown of detected information, seperated into the catagories Type, LableDetection, and ValueDetection
Type (dict) --
The implied label of a detected element. Present alongside LabelDetection for explicit elements.
Text (string) --
The word or line of text detected by Amazon Textract.
Confidence (float) --
The confidence of accuracy, as a percentage.
LabelDetection (dict) --
The explicitly stated label of a detected element.
Text (string) --
The word or line of text recognized by Amazon Textract
Geometry (dict) --
Information about where the following items are located on a document page: detected page, text, key-value pairs, tables, table cells, and selection elements.
BoundingBox (dict) --
An axis-aligned coarse representation of the location of the recognized item on the document page.
Width (float) --
The width of the bounding box as a ratio of the overall document page width.
Height (float) --
The height of the bounding box as a ratio of the overall document page height.
Left (float) --
The left coordinate of the bounding box as a ratio of overall document page width.
Top (float) --
The top coordinate of the bounding box as a ratio of overall document page height.
Polygon (list) --
Within the bounding box, a fine-grained polygon around the recognized item.
(dict) --
The X and Y coordinates of a point on a document page. The X and Y values that are returned are ratios of the overall document page size. For example, if the input document is 700 x 200 and the operation returns X=0.5 and Y=0.25, then the point is at the (350,50) pixel coordinate on the document page.
An array of Point objects, Polygon, is returned by DetectDocumentText. Polygon represents a fine-grained polygon around detected text. For more information, see Geometry in the Amazon Textract Developer Guide.
X (float) --
The value of the X coordinate for a point on a Polygon.
Y (float) --
The value of the Y coordinate for a point on a Polygon.
Confidence (float) --
The confidence in detection, as a percentage
ValueDetection (dict) --
The value of a detected element. Present in explicit and implicit elements.
Text (string) --
The word or line of text recognized by Amazon Textract
Geometry (dict) --
Information about where the following items are located on a document page: detected page, text, key-value pairs, tables, table cells, and selection elements.
BoundingBox (dict) --
An axis-aligned coarse representation of the location of the recognized item on the document page.
Width (float) --
The width of the bounding box as a ratio of the overall document page width.
Height (float) --
The height of the bounding box as a ratio of the overall document page height.
Left (float) --
The left coordinate of the bounding box as a ratio of overall document page width.
Top (float) --
The top coordinate of the bounding box as a ratio of overall document page height.
Polygon (list) --
Within the bounding box, a fine-grained polygon around the recognized item.
(dict) --
The X and Y coordinates of a point on a document page. The X and Y values that are returned are ratios of the overall document page size. For example, if the input document is 700 x 200 and the operation returns X=0.5 and Y=0.25, then the point is at the (350,50) pixel coordinate on the document page.
An array of Point objects, Polygon, is returned by DetectDocumentText. Polygon represents a fine-grained polygon around detected text. For more information, see Geometry in the Amazon Textract Developer Guide.
X (float) --
The value of the X coordinate for a point on a Polygon.
Y (float) --
The value of the Y coordinate for a point on a Polygon.
Confidence (float) --
The confidence in detection, as a percentage
PageNumber (integer) --
The page number the value was detected on.
Warnings (list) --
A list of warnings that occurred during the text-detection operation for the document.
(dict) --
A warning about an issue that occurred during asynchronous text analysis ( StartDocumentAnalysis) or asynchronous document text detection ( StartDocumentTextDetection).
ErrorCode (string) --
The error code for the warning.
Pages (list) --
A list of the pages that the warning applies to.
(integer) --
StatusMessage (string) --
Returns if the detection job could not be completed. Contains explanation for what error occured.
AnalyzeExpenseModelVersion (string) --
The current model version of AnalyzeExpense.