Skip to main content

when i use the action : 

App integration: Capture area

it retrieves window title -the running app

iam trying to capture  text from chat  gpt 
and this is the result !!

i used run from here » so» focus on just the action and mesage box » not all actions 

i tried with another file from notbad and word file but all have same ouput » window title-running app 

 

any soluotions please ? 

Hi ​@a000hmedrpa,

 

Any specific reason for not using the Recorder → Capture or OCR → Capture area action instead of this?

 

Some apps don’t expose text elements to Automation Anywhere’s object cloning engine. By using the App Integrate: Capture area action, I also found the same behavior while extracting text from Notepad & Word. However, I was able to extract text from the Browser window successfully. The primary target of this package is to retrieve text from an interface like CLI.
 

 

So, if you haven't tried the above 2 packages for this, please try once and let me know.


@a000hmedrpa The “App Integration” package is really not designed for what you’re doing.

https://docs.automationanywhere.com/bundle/enterprise-v2019/page/enterprise-cloud/topics/aae-client/bot-creator/commands/cloud-app-integration-package.html

It’s more for getting text from command prompts.

If you’re trying to get data from ChatGPT, use the OpenAI API. No, it’s not free. 

Another way is using Recorder: Capture. If you carefully align the capture, you can capture the entire response.

And yes, I knew that using App Integration for what you’re doing isn’t a good idea before I asked ChatGPT. 😂

Even better is to use the Enterprise version of our software (again, not free) that can integrate with ALL the large language models out there.


@Padmakumar my task is in word file not gpt window

recoder: capture >>captures the options of the word options menu not the document text 

ocr captures text in screen but i need to  capture some text from begining then scroll down and capture other things down at the last page 


@Aaron.Gleason 

my task is in word file not gpt window

recoder: capture >>captures the options of the word options menu not the document text 

ocr captures text in screen but i need to  capture some text from begining then scroll down and capture other things down at the last page 


@Padmakumar my task is in word file not gpt window

recoder: capture >>captures the options of the word options menu not the document text 

ocr captures text in screen but i need to  capture some text from begining then scroll down and capture other things down at the last page 

 

There are two packages in Bot store that you can use to interract with Word documents. 

 

  1. MS Word-Apache POI - Bot Store (Word installation is not required) 
  2. Microsoft Word Package - Bot Store (Word installation is required)

@a000hmedrpa In addition to ​@Padmakumar’s great suggestions (which unfortunately won’t work in Community Edition), you have the ability to use keystrokes with Word to extract information.

Use Simulate Keystrokes and pass a which will select all the text and copy it to the clipboard. 

Once on the clipboard, you can access the contents with $Clipboard:Clipboard$.


@Padmakumar 

@Aaron.Gleason 

ok i will try ur solutions > hanks alot ❤️


@a000hmedrpa In addition to ​@Padmakumar’s great suggestions (which unfortunately won’t work in Community Edition), you have the ability to use keystrokes with Word to extract information.

Use Simulate Keystrokes and pass a which will select all the text and copy it to the clipboard. 

Once on the clipboard, you can access the contents with $Clipboard:Clipboard$.

These will work in Community Edition as well

 

 


Reply