Page 1 of 1

Performance optimization solution in Oxygen XML when opening and editing large XML documents (200MB–1GB)

Posted: Tue Nov 18, 2025 7:51 am
by benedict
Hi everyone,

While working with very large XML documents (200MB or more), I am experiencing a situation where Oxygen XML Editor is slow, with a sharp drop in responsiveness when scrolling, searching, or performing tasks related to schema validation.

I have tried a few things:

Disable Schema validation on-the-fly

Switch to Text mode instead of Author mode

Increase JVM memory in the oxygen.vmoptions file

However, performance has not really improved to the desired level, especially when working on files containing many namespaces or complex structures.

I would like to ask the community:

Is there any advanced configuration that helps Oxygen work more stably with large files?

Is there a recommended plugin or workflow to split, preload, or optimize processing of large files?

Has anyone ever implemented a large XML processing process by combining Oxygen with Saxon, Ant, or XProc pipeline?

Does Oxygen have any updated support or best practices for these scenarios in 2024–2025?

I would love to hear from you, especially those who regularly work with large document sets, technical documentation, or XML published data.

Thanks community!

Re: Performance optimization solution in Oxygen XML when opening and editing large XML documents (200MB–1GB)

Posted: Tue Nov 18, 2025 9:04 am
by Radu
Hi Benedict,

Please see some answers below:
Disable Schema validation on-the-fly
Schema based validation is done indeed by default whenever you are typing, it's done on a separate thread so it might not impact the overall editing performance but as the document is very large, the validation may take a while. How many seconds does validation seem to take when you use the toolbar "Validate" button?
Switch to Text mode instead of Author mode
What kind of XML documents are you editing? What do they contain? At this size I would assume your XML files are some kind of database dump?
So they can also be edited visually in the Author visual editor mode? The Author visual editor mode needs probably a long time to parse the 200MBs and about more than 1.5GB's of memory to load the entire thing. After it's loaded is the overall editing in a paragraph slow? The Author visual editor was not really intended to edit such huge files.
Increase JVM memory in the oxygen.vmoptions file
Yes, this helps, on Mac OS my Oxygen application uses a maximum of 4GBs of internal memory.
However, performance has not really improved to the desired level, especially when working on files containing many namespaces or complex structures.
Now I'm really curios about what your XML documents contain.
I would like to ask the community:
The Oxygen forum is more of a static place, people post but they usually register to be notified only on their own posts. We have a couple of power users who are registered to receive notifications on any forum post, but I don't think they are editing 200MB files.
Is there any advanced configuration that helps Oxygen work more stably with large files?
When you open the XML file in Oxygen, if it's huge, Oxygen will show you a choice dialog to load it with "large" or "huge" files support. That's about it.
https://www.oxygenxml.com/doc/versions/ ... ments.html
Is there a recommended plugin or workflow to split, preload, or optimize processing of large files?
Most people do not work with such huge XML documents. They use techniques like xi:include to split the XML files in modules.
https://www.oxygenxml.com/doc/versions/ ... clude.html
Has anyone ever implemented a large XML processing process by combining Oxygen with Saxon, Ant, or XProc pipeline?
As I suggested, xi:include may be the answer. We do not have a sample Oxygen plugin which splits an XML document in multiple pieces and then binds them together again.
Does Oxygen have any updated support or best practices for these scenarios in 2024–2025?
Work with smaller XML modules, bind them together using xi:include or entity references.
I would love to hear from you, especially those who regularly work with large document sets, technical documentation, or XML published data.
As mentioned above, most Oxygen users are not notified when one posts on the forum. The Oxygen Users List is a better place where to reach directly a larger community of Oxygen Users: https://mailman.oxygenxml.com/postorius ... enxml.com/

Regards,
Radu