Topics

Requesting Help with Add-on Developement #add-on


jngasto2@...
 

So I've been working on this addon for year now. Initially I was making good progress testing functionality on notepad++. Currently I'm trying to expand compatibility to internet browsers. In short, my goal is to "capture" all text/characters present within a window/tab (even text that's off-screen) and print it to a file. To do this, I've been traversing the hierarchy of NVDAObjects that make up said window/tab and I've been looking through their coinciding TextInfo objects to see if they provide meaningful text. If so, I write this text to a file. I had no problems with this method when testing it on Notepad++, but as I've moved on to internet browsers, I've found traversing objects to be much more difficult given the number of objects that may make up one window. I've even tried using sayAllHandler._ObjectsReader's "walk" generator function to no avail. The problem originates from an inconsistency in traversing the object hierarchy due to a "core.CallCancelled" error when simpleNext/simpleFirstChild is called. Any insight to this problem or perhaps a more efficient solution to this problem would be greatly appreciated. (please note, it is important for me to also have the coinciding TextInfo, ultimately this addon is meant to search through text for a given phrase/regex pattern)


Adriani Botez
 

The addon dtk (developer toolkit) could help you further here. You can find it on addons.nvda-project.org

Best
Adriani
 

Von meinem iPhone gesendet

Am 12.07.2020 um 00:34 schrieb jngasto2@...:

So I've been working on this addon for year now. Initially I was making good progress testing functionality on notepad++. Currently I'm trying to expand compatibility to internet browsers. In short, my goal is to "capture" all text/characters present within a window/tab (even text that's off-screen) and print it to a file. To do this, I've been traversing the hierarchy of NVDAObjects that make up said window/tab and I've been looking through their coinciding TextInfo objects to see if they provide meaningful text. If so, I write this text to a file. I had no problems with this method when testing it on Notepad++, but as I've moved on to internet browsers, I've found traversing objects to be much more difficult given the number of objects that may make up one window. I've even tried using sayAllHandler._ObjectsReader's "walk" generator function to no avail. The problem originates from an inconsistency in traversing the object hierarchy due to a "core.CallCancelled" error when simpleNext/simpleFirstChild is called. Any insight to this problem or perhaps a more efficient solution to this problem would be greatly appreciated. (please note, it is important for me to also have the coinciding TextInfo, ultimately this addon is meant to search through text for a given phrase/regex pattern)


 

Hi,

If I understood the original question correctly, the actual place to start with would be NVDA’s own find command and how it works (cursor manager is the place to look at).

Cheers,

Jsoeph

 

From: nvda-addons@nvda-addons.groups.io <nvda-addons@nvda-addons.groups.io> On Behalf Of Adriani Botez
Sent: Saturday, July 11, 2020 10:15 PM
To: nvda-addons@nvda-addons.groups.io
Subject: Re: [nvda-addons] Requesting Help with Add-on Developement #add-on

 

The addon dtk (developer toolkit) could help you further here. You can find it on addons.nvda-project.org

 

Best

Adriani

 

Von meinem iPhone gesendet



Am 12.07.2020 um 00:34 schrieb jngasto2@...:

So I've been working on this addon for year now. Initially I was making good progress testing functionality on notepad++. Currently I'm trying to expand compatibility to internet browsers. In short, my goal is to "capture" all text/characters present within a window/tab (even text that's off-screen) and print it to a file. To do this, I've been traversing the hierarchy of NVDAObjects that make up said window/tab and I've been looking through their coinciding TextInfo objects to see if they provide meaningful text. If so, I write this text to a file. I had no problems with this method when testing it on Notepad++, but as I've moved on to internet browsers, I've found traversing objects to be much more difficult given the number of objects that may make up one window. I've even tried using sayAllHandler._ObjectsReader's "walk" generator function to no avail. The problem originates from an inconsistency in traversing the object hierarchy due to a "core.CallCancelled" error when simpleNext/simpleFirstChild is called. Any insight to this problem or perhaps a more efficient solution to this problem would be greatly appreciated. (please note, it is important for me to also have the coinciding TextInfo, ultimately this addon is meant to search through text for a given phrase/regex pattern)