I've been testing some features in the web client, and I've noticed an issue with the redaction and highlight tools. If I draw the annotations directly on the image, the annotation matching annotation in the text pane will be offset by one character.

See the following screen shot.

If I use the tool in the text pane, both annotations match.

This issue is specific to the web client. It works fine on the desktop client.

Here is the version we are on.

I've tested this in Chrome 65.0.3325.181 (Official Build) (64-bit) and Edge 41.16299.15.0.

Update:

I just tried it on our test server which is running

On 10.3 it's worse. It's offset by several characters instead of just one.

Update:

There seems to be a difference between various documents. Older documents that have OCR'd text seem to work fine. The above document was imported via Snapshot from Word. Other documents that I'm having trouble with are tiff's imported with the SDK.

I also took an older document and re-reran OCR from the Desktop Client, and it still works fine in the Web Client.

I took the above document that had been snapshotted, and re-ran OCR from the Desktop Client, and it now works fine in the Web Client.

Update:

I took one of the documents that I had imported using the SDK, and re-ran OCR from the Desktop Client, and it also works fine in the Web Client.

I also changed my snapshot settings from "Obtain text from the print job" to "Perform OCR on the images created for the print job" and then the Web Client was able to correctly apply the annotations.

Here is my import code for reference:

var importer = new DocumentImporter();
importer.Document = doc;
importer.OcrImages = true;
importer.OcrOptions.OptimizationMode = OcrOptimizationMode.Accuracy;
importer.ImportImages(memoryStream);

My client version is 10.2 (10.2.1.998).