Hi everyone!
I’ve been spending some of my personal time lately exploring AI Agents (specifically OpenClaw) as a hobby. To see how these agents handle specific technical tasks, I decided to point one at a problem many of us face: DPL (Dynatrace Pattern Language).
Let’s be honest—DPL is powerful, but building the exact parsing pattern can be confusing and a bit tiresome. We've all spent too much time staring at log lines trying to get the syntax just right.
A small step for DPL-kind
I’ve put together a small CLI utility called dynatrace-dpl-helper. It’s definitely a "baby step" and a trial run, but if it solves even one person's parsing headache, that’s a big win in my book.
You can find it on PyPI:
pip install dynatrace-dpl-helper
How it works (with an example)
Say you have a file called sample.json with this content:
{ "content":"Ai Webservice took: 1234ms, AgentReadyCount: 32" }
You can run this command to generate your pattern:
dynatrace-dpl-helper --file sample.json --literal "Webservice took:" --value 1234 --type INT --alias responseTime
Breakdown of the command:
--file: The source file containing your log sample.
--literal: A "hook" or static text that appears right before the data you want. (Note: This doesn't have to be long! If your log says "The webservice took 1234ms", you could just use --literal "took").
--value: A sample of the data you want to extract (e.g., 1234). This just helps the CLI identify exactly which part of the string you're targeting.
--type: The data format you want (INT, DOUBLE, STRING, etc.).
--alias: The name you want to give this field in Dynatrace.
Help me make it better!
This is an ongoing experiment, and I’m really looking for active feedback to improve it. If you give it a try, please let me know:
What did you use it for?
How did it help?
Specifically, where did it fail? Knowing exactly where the logic breaks is the best way for me to develop this further and refine how the AI agent builds these patterns. Looking forward to hearing your thoughts!
I’m excited to see if this helps make your log-parsing life a little bit easier. Drop a comment below with your experience or any questions!
Happy parsing!
@Maheedhar_T
Maheedhar
