use of com.google.cloud.vision.v1p3beta1.TextAnnotation in project java-vision by googleapis.
the class Detect method detectDocumentTextGcs.
// [END vision_fulltext_detection]
/**
* Performs document text detection on a remote image on Google Cloud Storage.
*
* @param gcsPath The path to the remote file on Google Cloud Storage to detect document text on.
* @throws Exception on errors while closing the client.
* @throws IOException on Input/Output errors.
*/
// [START vision_fulltext_detection_gcs]
public static void detectDocumentTextGcs(String gcsPath) throws IOException {
List<AnnotateImageRequest> requests = new ArrayList<>();
ImageSource imgSource = ImageSource.newBuilder().setGcsImageUri(gcsPath).build();
Image img = Image.newBuilder().setSource(imgSource).build();
Feature feat = Feature.newBuilder().setType(Type.DOCUMENT_TEXT_DETECTION).build();
AnnotateImageRequest request = AnnotateImageRequest.newBuilder().addFeatures(feat).setImage(img).build();
requests.add(request);
// the "close" method on the client to safely clean up any remaining background resources.
try (ImageAnnotatorClient client = ImageAnnotatorClient.create()) {
BatchAnnotateImagesResponse response = client.batchAnnotateImages(requests);
List<AnnotateImageResponse> responses = response.getResponsesList();
client.close();
for (AnnotateImageResponse res : responses) {
if (res.hasError()) {
System.out.format("Error: %s%n", res.getError().getMessage());
return;
}
// For full list of available annotations, see http://g.co/cloud/vision/docs
TextAnnotation annotation = res.getFullTextAnnotation();
for (Page page : annotation.getPagesList()) {
String pageText = "";
for (Block block : page.getBlocksList()) {
String blockText = "";
for (Paragraph para : block.getParagraphsList()) {
String paraText = "";
for (Word word : para.getWordsList()) {
String wordText = "";
for (Symbol symbol : word.getSymbolsList()) {
wordText = wordText + symbol.getText();
System.out.format("Symbol text: %s (confidence: %f)%n", symbol.getText(), symbol.getConfidence());
}
System.out.format("Word text: %s (confidence: %f)%n%n", wordText, word.getConfidence());
paraText = String.format("%s %s", paraText, wordText);
}
// Output Example using Paragraph:
System.out.println("%nParagraph: %n" + paraText);
System.out.format("Paragraph Confidence: %f%n", para.getConfidence());
blockText = blockText + paraText;
}
pageText = pageText + blockText;
}
}
System.out.println("%nComplete annotation:");
System.out.println(annotation.getText());
}
}
}
use of com.google.cloud.vision.v1p3beta1.TextAnnotation in project java-vision by googleapis.
the class ITSystemTest method detectDocumentTextGcs.
@Test
public void detectDocumentTextGcs() throws IOException {
List<AnnotateImageResponse> responses = getResponsesList("text/screen.jpg", Type.DOCUMENT_TEXT_DETECTION, true);
String actual = "";
for (AnnotateImageResponse imgResponse : responses) {
TextAnnotation annotation = imgResponse.getFullTextAnnotation();
actual = annotation.getText();
}
assertThat(actual).contains("After preparation is complete");
}
use of com.google.cloud.vision.v1p3beta1.TextAnnotation in project spring-cloud-gcp by GoogleCloudPlatform.
the class DocumentOcrTemplateIntegrationTests method testDocumentOcrTemplate.
@Test
void testDocumentOcrTemplate() throws ExecutionException, InterruptedException, InvalidProtocolBufferException, TimeoutException {
GoogleStorageLocation document = GoogleStorageLocation.forFile("vision-integration-test-bucket", "test.pdf");
GoogleStorageLocation outputLocationPrefix = GoogleStorageLocation.forFile("vision-integration-test-bucket", "it_output/test-");
ListenableFuture<DocumentOcrResultSet> result = this.documentOcrTemplate.runOcrForDocument(document, outputLocationPrefix);
DocumentOcrResultSet ocrPages = result.get(5, TimeUnit.MINUTES);
String page1Text = ocrPages.getPage(1).getText();
assertThat(page1Text).contains("Hello World. Is mayonnaise an instrument?");
String page2Text = ocrPages.getPage(2).getText();
assertThat(page2Text).contains("Page 2 stuff");
ArrayList<String> pageContent = new ArrayList<>();
Iterator<TextAnnotation> pageIterator = ocrPages.getAllPages();
while (pageIterator.hasNext()) {
pageContent.add(pageIterator.next().getText());
}
assertThat(pageContent).containsExactly("Hello World. Is mayonnaise an instrument?\n", "Page 2 stuff\n", "Page 3 stuff\n", "Page 4 stuff\n");
}
use of com.google.cloud.vision.v1p3beta1.TextAnnotation in project spring-cloud-gcp by GoogleCloudPlatform.
the class WebController method renderViewDocumentPage.
@GetMapping("/viewDocument")
public ModelAndView renderViewDocumentPage(@RequestParam("gcsDocumentUrl") String gcsDocumentUrl, @RequestParam("pageNumber") int pageNumber, ModelMap map) throws ExecutionException, InterruptedException, InvalidProtocolBufferException {
TextAnnotation textAnnotation = ocrStatusReporter.getDocumentOcrStatuses().get(gcsDocumentUrl).getResultSet().getPage(pageNumber);
String[] firstWordsTokens = textAnnotation.getText().split(" ", 50);
map.put("pageNumber", pageNumber);
map.put("gcsDocumentUrl", gcsDocumentUrl);
map.put("text", String.join(" ", firstWordsTokens));
return new ModelAndView("viewDocument", map);
}
use of com.google.cloud.vision.v1p3beta1.TextAnnotation in project kie-wb-common by kiegroup.
the class Bpmn2JsonUnmarshaller method unmarshallItem.
public BaseElement unmarshallItem(JsonParser parser, String preProcessingData) throws JsonParseException, IOException {
String resourceId = null;
Map<String, String> properties = null;
String stencil = null;
List<BaseElement> childElements = new ArrayList<BaseElement>();
List<String> outgoing = new ArrayList<String>();
while (parser.nextToken() != JsonToken.END_OBJECT) {
String fieldname = parser.getCurrentName();
parser.nextToken();
if ("resourceId".equals(fieldname)) {
resourceId = parser.getText();
} else if ("properties".equals(fieldname)) {
properties = unmarshallProperties(parser);
} else if ("stencil".equals(fieldname)) {
// "stencil":{"id":"Task"},
parser.nextToken();
parser.nextToken();
stencil = parser.getText();
parser.nextToken();
} else if ("childShapes".equals(fieldname)) {
while (parser.nextToken() != JsonToken.END_ARRAY) {
// open the
// object
// the childShapes element is a json array. We opened the
// array.
childElements.add(unmarshallItem(parser, preProcessingData));
}
} else if ("bounds".equals(fieldname)) {
// bounds: {"lowerRight":{"x":484.0,"y":198.0},"upperLeft":{"x":454.0,"y":168.0}}
parser.nextToken();
parser.nextToken();
parser.nextToken();
parser.nextToken();
Integer x2 = parser.getIntValue();
parser.nextToken();
parser.nextToken();
Integer y2 = parser.getIntValue();
parser.nextToken();
parser.nextToken();
parser.nextToken();
parser.nextToken();
parser.nextToken();
Integer x1 = parser.getIntValue();
parser.nextToken();
parser.nextToken();
Integer y1 = parser.getIntValue();
parser.nextToken();
parser.nextToken();
Bounds b = DcFactory.eINSTANCE.createBounds();
b.setX(x1);
b.setY(y1);
b.setWidth(x2 - x1);
b.setHeight(y2 - y1);
this._bounds.put(resourceId, b);
} else if ("dockers".equals(fieldname)) {
// "dockers":[{"x":50,"y":40},{"x":353.5,"y":115},{"x":353.5,"y":152},{"x":50,"y":40}],
List<Point> dockers = new ArrayList<Point>();
JsonToken nextToken = parser.nextToken();
boolean end = JsonToken.END_ARRAY.equals(nextToken);
while (!end) {
nextToken = parser.nextToken();
nextToken = parser.nextToken();
Integer x = parser.getIntValue();
parser.nextToken();
parser.nextToken();
Integer y = parser.getIntValue();
Point point = DcFactory.eINSTANCE.createPoint();
point.setX(x);
point.setY(y);
dockers.add(point);
parser.nextToken();
nextToken = parser.nextToken();
end = JsonToken.END_ARRAY.equals(nextToken);
}
this._dockers.put(resourceId, dockers);
} else if ("outgoing".equals(fieldname)) {
while (parser.nextToken() != JsonToken.END_ARRAY) {
// {resourceId: oryx_1AAA8C9A-39A5-42FC-8ED1-507A7F3728EA}
parser.nextToken();
parser.nextToken();
outgoing.add(parser.getText());
parser.nextToken();
}
// pass on the array
parser.skipChildren();
} else if ("target".equals(fieldname)) {
// we already collected that info with the outgoing field.
parser.skipChildren();
// "target": {
// "resourceId": "oryx_A75E7546-DF71-48EA-84D3-2A8FD4A47568"
// }
// add to the map:
// parser.nextToken(); // resourceId:
// parser.nextToken(); // the value we want to save
// targetId = parser.getText();
// parser.nextToken(); // }, closing the object
}
}
properties.put("resourceId", resourceId);
boolean customElement = isCustomElement(properties.get("tasktype"), preProcessingData);
BaseElement baseElt = this.createBaseElement(stencil, properties.get("tasktype"), customElement);
// register the sequence flow targets.
if (baseElt instanceof SequenceFlow) {
_sequenceFlowTargets.addAll(outgoing);
}
_outgoingFlows.put(baseElt, outgoing);
_objMap.put(baseElt, // keep the object around to do connections
resourceId);
_idMap.put(resourceId, baseElt);
// baseElt.setId(resourceId); commented out as bpmn2 seems to create
// duplicate ids right now.
applyProperties(baseElt, properties, preProcessingData);
if (baseElt instanceof Definitions) {
Process rootLevelProcess = null;
if (childElements == null || childElements.size() < 1) {
if (rootLevelProcess == null) {
rootLevelProcess = Bpmn2Factory.eINSTANCE.createProcess();
// set the properties and item definitions first
if (properties.get("vardefs") != null && properties.get("vardefs").length() > 0) {
String[] vardefs = properties.get("vardefs").split(",\\s*");
for (String vardef : vardefs) {
Property prop = Bpmn2Factory.eINSTANCE.createProperty();
ItemDefinition itemdef = Bpmn2Factory.eINSTANCE.createItemDefinition();
// check if we define a structure ref in the definition
if (vardef.contains(":")) {
String[] vardefParts = vardef.split(":\\s*");
prop.setId(vardefParts[0]);
itemdef.setId("_" + prop.getId() + "Item");
boolean haveKPI = false;
String kpiValue = "";
if (vardefParts.length == 3) {
itemdef.setStructureRef(vardefParts[1]);
if (vardefParts[2].equals("true")) {
haveKPI = true;
kpiValue = vardefParts[2];
}
}
if (vardefParts.length == 2) {
if (vardefParts[1].equals("true") || vardefParts[1].equals("false")) {
if (vardefParts[1].equals("true")) {
haveKPI = true;
kpiValue = vardefParts[1];
}
} else {
itemdef.setStructureRef(vardefParts[1]);
}
}
if (haveKPI) {
Utils.setMetaDataExtensionValue(prop, "customKPI", wrapInCDATABlock(kpiValue));
}
} else {
prop.setId(vardef);
itemdef.setId("_" + prop.getId() + "Item");
}
prop.setItemSubjectRef(itemdef);
rootLevelProcess.getProperties().add(prop);
((Definitions) baseElt).getRootElements().add(itemdef);
}
}
if (properties.get("adhocprocess") != null && properties.get("adhocprocess").equals("true")) {
ExtendedMetaData metadata = ExtendedMetaData.INSTANCE;
EAttributeImpl extensionAttribute = (EAttributeImpl) metadata.demandFeature("http://www.jboss.org/drools", "adHoc", false, false);
SimpleFeatureMapEntry extensionEntry = new SimpleFeatureMapEntry(extensionAttribute, properties.get("adhocprocess"));
rootLevelProcess.getAnyAttribute().add(extensionEntry);
}
if (properties.get("customdescription") != null && properties.get("customdescription").length() > 0) {
Utils.setMetaDataExtensionValue(rootLevelProcess, "customDescription", wrapInCDATABlock(properties.get("customdescription")));
}
rootLevelProcess.setId(properties.get("id"));
applyProcessProperties(rootLevelProcess, properties);
((Definitions) baseElt).getRootElements().add(rootLevelProcess);
}
} else {
for (BaseElement child : childElements) {
// } else {
if (child instanceof SequenceFlow) {
// for some reason sequence flows are placed as root elements.
// find if the target has a container, and if we can use it:
List<String> ids = _outgoingFlows.get(child);
FlowElementsContainer container = null;
for (String id : ids) {
// yes, we iterate, but we'll take the first in the list that will work.
Object obj = _idMap.get(id);
if (obj instanceof EObject && ((EObject) obj).eContainer() instanceof FlowElementsContainer) {
container = (FlowElementsContainer) ((EObject) obj).eContainer();
break;
}
}
if (container != null) {
container.getFlowElements().add((SequenceFlow) child);
continue;
}
}
if (child instanceof Task || child instanceof SequenceFlow || child instanceof Gateway || child instanceof Event || child instanceof Artifact || child instanceof DataObject || child instanceof SubProcess || child instanceof Lane || child instanceof CallActivity || child instanceof TextAnnotation) {
if (rootLevelProcess == null) {
rootLevelProcess = Bpmn2Factory.eINSTANCE.createProcess();
// set the properties and item definitions first
if (properties.get("vardefs") != null && properties.get("vardefs").length() > 0) {
String[] vardefs = properties.get("vardefs").split(",\\s*");
for (String vardef : vardefs) {
Property prop = Bpmn2Factory.eINSTANCE.createProperty();
ItemDefinition itemdef = Bpmn2Factory.eINSTANCE.createItemDefinition();
// check if we define a structure ref in the definition
if (vardef.contains(":")) {
String[] vardefParts = vardef.split(":\\s*");
prop.setId(vardefParts[0]);
itemdef.setId("_" + prop.getId() + "Item");
boolean haveKPI = false;
String kpiValue = "";
if (vardefParts.length == 3) {
itemdef.setStructureRef(vardefParts[1]);
if (vardefParts[2].equals("true")) {
haveKPI = true;
kpiValue = vardefParts[2];
}
}
if (vardefParts.length == 2) {
if (vardefParts[1].equals("true") || vardefParts[1].equals("false")) {
if (vardefParts[1].equals("true")) {
haveKPI = true;
kpiValue = vardefParts[1];
}
} else {
itemdef.setStructureRef(vardefParts[1]);
}
}
if (haveKPI) {
Utils.setMetaDataExtensionValue(prop, "customKPI", wrapInCDATABlock(kpiValue));
}
} else {
prop.setId(vardef);
itemdef.setId("_" + prop.getId() + "Item");
}
prop.setItemSubjectRef(itemdef);
rootLevelProcess.getProperties().add(prop);
((Definitions) baseElt).getRootElements().add(itemdef);
}
}
if (properties.get("adhocprocess") != null && properties.get("adhocprocess").equals("true")) {
ExtendedMetaData metadata = ExtendedMetaData.INSTANCE;
EAttributeImpl extensionAttribute = (EAttributeImpl) metadata.demandFeature("http://www.jboss.org/drools", "adHoc", false, false);
SimpleFeatureMapEntry extensionEntry = new SimpleFeatureMapEntry(extensionAttribute, properties.get("adhocprocess"));
rootLevelProcess.getAnyAttribute().add(extensionEntry);
}
if (properties.get("customdescription") != null && properties.get("customdescription").length() > 0) {
Utils.setMetaDataExtensionValue(rootLevelProcess, "customDescription", wrapInCDATABlock(properties.get("customdescription")));
}
rootLevelProcess.setId(properties.get("id"));
applyProcessProperties(rootLevelProcess, properties);
((Definitions) baseElt).getRootElements().add(rootLevelProcess);
}
}
if (child instanceof Task) {
rootLevelProcess.getFlowElements().add((Task) child);
} else if (child instanceof CallActivity) {
rootLevelProcess.getFlowElements().add((CallActivity) child);
} else if (child instanceof RootElement) {
((Definitions) baseElt).getRootElements().add((RootElement) child);
} else if (child instanceof SequenceFlow) {
rootLevelProcess.getFlowElements().add((SequenceFlow) child);
} else if (child instanceof Gateway) {
rootLevelProcess.getFlowElements().add((Gateway) child);
} else if (child instanceof Event) {
rootLevelProcess.getFlowElements().add((Event) child);
} else if (child instanceof TextAnnotation) {
rootLevelProcess.getFlowElements().add((TextAnnotation) child);
} else if (child instanceof Artifact) {
rootLevelProcess.getArtifacts().add((Artifact) child);
} else if (child instanceof DataObject) {
// bubble up data objects
// rootLevelProcess.getFlowElements().add(0, (DataObject) child);
rootLevelProcess.getFlowElements().add((DataObject) child);
// ItemDefinition def = ((DataObject) child).getItemSubjectRef();
// if (def != null) {
// if (def.eResource() == null) {
// ((Definitions) rootLevelProcess.eContainer()).getRootElements().add(0, def);
// }
// Import imported = def.getImport();
// if (imported != null && imported.eResource() == null) {
// ((Definitions) rootLevelProcess.eContainer()).getImports().add(0, imported);
// }
// }
} else if (child instanceof SubProcess) {
rootLevelProcess.getFlowElements().add((SubProcess) child);
} else if (child instanceof Lane) {
// lanes handled later
} else {
_logger.error("Don't know what to do of " + child);
}
// }
}
}
} else if (baseElt instanceof Process) {
for (BaseElement child : childElements) {
if (child instanceof Lane) {
if (((Process) baseElt).getLaneSets().isEmpty()) {
((Process) baseElt).getLaneSets().add(Bpmn2Factory.eINSTANCE.createLaneSet());
}
((Process) baseElt).getLaneSets().get(0).getLanes().add((Lane) child);
addLaneFlowNodes((Process) baseElt, (Lane) child);
} else if (child instanceof Artifact) {
((Process) baseElt).getArtifacts().add((Artifact) child);
} else {
_logger.error("Don't know what to do of " + child);
}
}
} else if (baseElt instanceof SubProcess) {
for (BaseElement child : childElements) {
if (child instanceof FlowElement) {
((SubProcess) baseElt).getFlowElements().add((FlowElement) child);
} else if (child instanceof Artifact) {
((SubProcess) baseElt).getArtifacts().add((Artifact) child);
} else {
_logger.error("Subprocess - don't know what to do of " + child);
}
}
} else if (baseElt instanceof Message) {
// we do not support base-element messages from the json. They are created dynamically for events that use them.
} else if (baseElt instanceof Lane) {
for (BaseElement child : childElements) {
if (child instanceof FlowNode) {
((Lane) baseElt).getFlowNodeRefs().add((FlowNode) child);
} else // }
if (child instanceof Artifact) {
_artifacts.add((Artifact) child);
} else {
_logger.error("Don't know what to do of " + childElements);
}
}
_lanes.add((Lane) baseElt);
} else {
if (!childElements.isEmpty()) {
_logger.error("Don't know what to do of " + childElements + " with " + baseElt);
}
}
return baseElt;
}
Aggregations