2017-03-24 14:09:11

When trying to use OCR I get this error.

ERROR - scriptHandler.executeScript (06:08:03):
error executing script: <bound method GlobalPlugin.script_OCR of <globalPlugins.cloudOCR.GlobalPlugin object at 0x04DEEC50>> with gesture u'shift+NVDA+r'
Traceback (most recent call last):
  File "scriptHandler.pyo", line 187, in executeScript
  File "C:\Users\Ahmad\AppData\Roaming\nvda\addons\cloudOCR\globalPlugins\cloudOCR\__init__.py", line 77, in script_OCR
  File "C:\Users\Ahmad\AppData\Roaming\nvda\addons\cloudOCR\globalPlugins\cloudOCR\ocrspace.py", line 31, in OCR_file
TypeError: string indices must be integers
I'm hoping someone can help me fix this.

Silly Gohan, animals don't eat people. People eat animals.

2017-03-24 19:38:47

that don't look good haus, Maybe you could find a newer version of the addon? I know it works because I used it through someone else's computer via NVDA remote. I don't really know what the issue is though, gonna need someone who knows python to help with that.

Facts with Tom MacDonald, Adam Calhoun, and Dax
End racism
End division
Become united

2017-03-24 19:41:31

Is there a newer version of the addon? I just see 0.1.

Silly Gohan, animals don't eat people. People eat animals.

2017-03-24 21:16:41

This is by no means an official version of the add-in, but after experiencing difficulties installing the add-in on my girlfriend's laptop, I took the liberty of recompiling the add-in with a fresh set of files from the repository. Here is the Cloud OCR 0.2 add-in with the required extra files and updated manifest.ini to reflect that it is version 0.2, as per the github page.

I assume no responsibilities if you elect to use this version. I can only offer it with the guarantee that it worked for me, so it should work for you. It is, however, not my project, and am only offering this recompiled version out of a desire to assist others who are having difficulties with version 0.1. In short, if it reformats your hard drive or unleashes the demon hordes upon your grandmother, don't come cryin' to me about it.

Spill chuck you spots!

2017-03-25 02:22:01 (edited by death 2017-03-25 02:22:45)

Still getting tsame error for some reason. If someone can figure this out I'd really appreciate it. Thanks for the fix though either way.

Silly Gohan, animals don't eat people. People eat animals.

2017-03-25 04:56:58

So apparently it started working again? Kind of? I have no idea what is going on.

Silly Gohan, animals don't eat people. People eat animals.

2017-03-25 17:28:05

I get this error sometimes too. Things that I suspect  are happening here are:
certain images break the screen capture, causing the entire thing to fall flat on it's face, or the image is successfully captured, uploaded and converted to text. When this text is passed back to the cloud OCR script, certain parts of the image that have been turned into garbage text break the script.  I suspect this is the case because using windows magnifier to zoom in on bits of text, then OCRing the zoomed image will get results where zoomed out captures of a window have failed. What windows magnifier is effectively doing here is cropping out parts of the window that aren't text, but still get turned into text anyway.

Every record has been destroyed or falsified, every book rewritten, every picture has been repainted, every statue and street building has been renamed, every date has been altered. And the process is continuing day by day and minute by minute. History has stopped. Nothing exists except an endless present in which the Party is always right.

2017-03-25 22:17:52 (edited by death 2017-03-27 01:14:09)

Yeah now it's... half working? idk I'll take what I can get until it gets updated, I suppose. Which isn't to say I don't appreciate the plugin, just that I'm not complaining even though it's not perfect.
Does the error come because of the sight only having a certain number of OCR uses? Because it works at such random times.

Silly Gohan, animals don't eat people. People eat animals.

2017-03-29 23:30:44 (edited by Sean-Terry01 2017-03-29 23:38:38)

Hi all, I just got the unofficial 0.2 addon. thanks a lot for this. Is there some kind of manual to read so I can understand how to use it? Thanks a lot in advance.

2017-03-30 19:30:51

Read the github page, and it says what you need to do. NVDA+shift+r to ocr, use object and screen review commands to navigate around the ocred text.