Hi @Jack Bao ,

This issue is pretty common when using TEXTractor with PDFs in OutSystems. 

This means: 

• TEXTractor is trying to write or execute native binaries (PDF parsing libraries) 
• These binaries are being extracted at runtime into the Platform Server installation folder 
• The OutSystems service account does NOT have permission to access or execute files in that directory

This is not a problem with your app logic or the PDF file itself.

There are two solutions I can think of presently.

Solution 1: 

On the application server: 

1. Identify the OutSystems service account (usually one of these): 
   • NT AUTHORITY\NETWORK SERVICE 
   • A custom domain/service account running the Platform Server 
2. Grant Read / Write / Execute permissions to: C:\Program Files\OutSystems\Platform Server\running\ (or at minimum the specific bin2\x64 folder) 

This is the recommended fix and works in almost all environments.

Solution 2:

Move TEXTractor execution to a writable temp folder. 

If you have control over configuration: 

• Ensure TEXTractor extracts binaries to: C:\Windows\Temp or C:\Temp 

These folders already allow execution by service accounts.

I hope this helps.

Cheers,

Saugat

Hi @Jack Bao,

Please try this built:

https://drive.google.com/file/d/1yr0dFg5UC3q8HsdjgZ0DbK1XPXVoHnZf/view?usp=sharing

You might need to run it through IPP to remove the intellectual property:

http://www.outsystems.com/ipp

Best,

Bruno

Hi @Jack Bao,

Did you had the chance to test the TEXTractor package I shared earlier?

Best,

Bruno

Hi @Jack Bao,

Just published a new version of TEXTractor (2.3.2) that should solve this issue.

Best,

Bruno