Link to issue number:

Reported by @derekriemer in #7476 (comment).

Summary of the issue:

If you attempt to recognize an object where the location property returns None for some reason (e.g. the object died), you get this unhandled exception:

ERROR - scriptHandler.executeScript (03:14:01.536):
error executing script: <bound method GlobalCommands.script_recognizeWithCaptionBot of <globalCommands.GlobalCommands object at 0x0595E2D0>> with gesture u'alt+NVDA+d'
Traceback (most recent call last):
File "scriptHandler.py", line 187, in executeScript
script(gesture)
File "globalCommands.py", line 2064, in script_recognizeWithCaptionBot
recogUi.recognizeNavigatorObject(recog)
File "contentRecog\recogUi.py", line 129, in recognizeNavigatorObject
left, top, width, height = nav.location
TypeError: 'NoneType' object is not iterable

Description of how this pull request fixes the issue:

We catch the TypeError exception, report "Content is not visible" to the user and log a debugWarning.

Testing performed:

1. Opened google.com in Firefox.
2. Turned off caret moves review cursor (NVDA+6) and focus moves navigator object (NVDA+7).
3. Opened nvaccess.org.
4. Pressed NVDA+r.

Known issues with pull request:

None.

Change log entry:

Bug fixes:

- When using a content recognition command (e.g. NVDA+r), NVDA now reports an error message instead of nothing if the navigator object has disappeared. (#7567)