There appear to PDF417 barcodes that contain multiple records that are not processed by the iOS 7 barcode scanner:

http://www.idautomation.com/barcode-faq/2d/pdf417/

BridgeIt 1.0.3 now uses unicode-friendly URL encoding.

It appears that occasionally null objects will be returned to the scan callback. These are detected by the above fix.

Crash only occurred once, so it may be a rare condition with failed face detection rather than barcode detection. The computer generated equivalent barcode is easily recognized, but it is higher contrast and less blurry. As an initial fix, the above crash issue will be addressed.

Crash report from Rob's "UPC code of death" (attached):

Dec 4 16:49:17 BridgeIt[376] <Error>: -[AVMetadataFaceObject corners]: unrecognized selector sent to instance 0x14596aa0
Dec 4 16:49:17 BridgeIt[376] <Error>: *** Terminating app due to uncaught exception 'NSInvalidArgumentException', reason: '-[AVMetadataFaceObject corners]: unrecognized selector sent to instance 0x14596aa0'

• • • First throw call stack:

Fix for iOS BridgeIt is checked in, but complete fix will require the following for bridgeit.js

+++ b/src/bridgeit.js
@@ -369,7 +369,7 @@ if (!window.console) {
if (useBase64 && (data.indexOf("!") < 0))

var params = data.split("&");
var len = params.length;

http://labs.icesoft.com/mobileshowcase/javax.faces.resource/qrcode:%25C3%2585ngstr%25C3%25B6m%2Bis%2B10%255E-10m.jsf

QR code with internationalized character caused BridgeIt to crash:

http://labs.icesoft.com/mobileshowcase/javax.faces.resource/qrcode:Ångström.jsf

If a metal band is encoded as "M%C3%B6tley%20Cr%C3%BCe", unescape() in JavaScript is not sufficient for decoding. This results in individual characters for each UTF multi-character sequence.

decodeURIComponent() is sufficient, however.