Fix adobe reader int/str to NoneType comparisons as well as other identified cases

[LeonarddeR]

Link to issue number:

Fixes #9859

Summary of the issue:

Reading documents in Adobe Reader raises a TypeError

Description of how this pull request fixes the issue:

Fixed this int to NoneType comparison, identified some others while at it. The following commands where used:

• git grep "ID > 0"
• git grep "ID < 0"
• git grep "ID>0"
• git grep "ID<0"

I did not touch comparisons for IAccessibleChildID, as I'm 99% sure that this will never be None.

Testing performed:

T.b.d. @lukaszgo1 would you be able to test your case with Acrobat?

Known issues with pull request:

one

Change log entry:

None

[zstanecic]

I can test it too.

[lukaszgo1]

This specific bug has been fixed. However after pressing CTRL+Page Down error sound is played with each key press, NVDA needs to be killed, and the following is in the log multiple times:

ERROR - eventHandler.executeEvent (21:53:43.938):
error executing event: gainFocus on <NVDAObjects.IAccessible.adobeAcrobat.Document object at 0x06FA8D30> with extra args of {}
Traceback (most recent call last):
  File "eventHandler.pyc", line 150, in executeEvent
  File "eventHandler.pyc", line 163, in doPreGainFocus
  File "api.pyc", line 117, in setFocusObject
  File "treeInterceptorHandler.pyc", line 57, in cleanup
  File "baseObject.pyc", line 30, in __get__
  File "virtualBuffers\adobeAcrobat.pyc", line 99, in _get_isAlive
  File "baseObject.pyc", line 30, in __get__
  File "NVDAObjects\IAccessible\adobeAcrobat.pyc", line 86, in _get_role
  File "NVDAObjects\IAccessible\adobeAcrobat.pyc", line 42, in normalizeStdName
TypeError: '<=' not supported between instances of 'str' and 'NoneType'

[LeonarddeR]

Ugh, another one. Makes sense to fix this as part of this one as well.

[LeonarddeR]

Thanks for reporting again @lukaszgo1. Could you have another try?

[lukaszgo1]

This time error is as follows:

ERROR - eventHandler.executeEvent (12:14:52.955):
error executing event: gainFocus on <NVDAObjects.IAccessible.adobeAcrobat.Document object at 0x04E46DF0> with extra args of {}
Traceback (most recent call last):
  File "eventHandler.pyc", line 150, in executeEvent
  File "eventHandler.pyc", line 163, in doPreGainFocus
  File "api.pyc", line 89, in setFocusObject
  File "NVDAObjects\__init__.pyc", line 318, in __eq__
  File "NVDAObjects\IAccessible\adobeAcrobat.pyc", line 107, in _isEqual
  File "NVDAObjects\IAccessible\__init__.pyc", line 738, in _isEqual
  File "baseObject.pyc", line 46, in __get__
  File "baseObject.pyc", line 149, in _getPropertyViaCache
  File "NVDAObjects\IAccessible\__init__.pyc", line 744, in _get_name
  File "baseObject.pyc", line 30, in __get__
  File "NVDAObjects\IAccessible\adobeAcrobat.pyc", line 85, in _get_role
AttributeError: 'NoneType' object has no attribute 'GetStdName'

[LeonarddeR]

Ugh, my bad. Just did another push.

[lukaszgo1]

Now it works. Thanks @LeonarddeR

[feerrenrut]

All looks good, but a question about logging the caught exception.

[feerrenrut]

This means that it will hide any other TypeErrors. I think this should instead log at Debug / Debug warning.

[feerrenrut]

Will this be noisy? Are regular LookupError or AttributeError exceptions expected?

[LeonarddeR]

Hmm, good point. Do you want me to only log this for type errors? I'd say that this could still be noisy though. May be I should avoid raising the typeerror altogether.