Add IndexMapping Tool

[dbwiddis]

Description

Adds an IndexMappingTool. See previous dev branch PR: #1609.

It looks like it's part of pending PR #1812 but needed to unblock end-to-end testing.

Check List

• New functionality includes testing.
  • All tests pass
• New functionality has been documented.
  • New functionality has javadoc added
• Commits are signed per the DCO using --signoff

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

[codecov]

Codecov Report

Attention: 13 lines in your changes are missing coverage. Please review.

Comparison is base (b3c033f) 82.63% compared to head (435e318) 82.65%.
Report is 5 commits behind head on main.

Files	Patch %	Lines
...g/opensearch/ml/engine/tools/IndexMappingTool.java	84.28%	7 Missing and 4 partials ⚠️
...a/org/opensearch/ml/engine/tools/CatIndexTool.java	33.33%	0 Missing and 2 partials ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #1891      +/-   ##
============================================
+ Coverage     82.63%   82.65%   +0.02%     
- Complexity     5387     5397      +10     
============================================
  Files           521      522       +1     
  Lines         21705    21773      +68     
  Branches       2209     2216       +7     
============================================
+ Hits          17935    17996      +61     
- Misses         2873     2876       +3     
- Partials        897      901       +4     

Flag	Coverage Δ
ml-commons	82.65% <82.66%> (+0.02%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[arjunkumargiri]

Can we do manual integration test by adding index mapping tool to chat agent and validate that the tool is able to integrate with LLM as expected?

[dbwiddis]

Can we do manual integration test by adding index mapping tool to chat agent and validate that the tool is able to integrate with LLM as expected?

I'm already doing this with the testing framwork currently at https://github.com/joshuali925/observability-langchain but soon to be moved to https://github.com/opensearch-project/skills-eval

Sample output, including a hallucination:

Running test: 9E036FA47
Received: After reviewing the available information, the data fields available for the aircraft-match index are id, aircraft_id, and match_id.
Expected: The data fields available for the aircraft-match index are id, aircraft_id, and match_id
Score: 1

Running test: AC1D6C46C
Received: After reviewing the available information, the data type of the '%_Change_2007' field in the 'aircraft-airport' index is a string.
Expected: The data type of the '%_Change_2007' field in the 'aircraft-airport' index is a string.
Score: 1

Running test: CBBC09702
Received: After reviewing the information gathered, the fields for the aircraft-aircraft index are name, model, and year.
Expected: The fields for the aircraft-aircraft index are Aircraft, Aircraft_ID, Description, Max_Gross_weight, Max_disk_Loading, and Total_disk_area
Score: 0

[dbwiddis]

JDK 21 test failures are as a result of https://bugs.openjdk.org/browse/JDK-8323659, see opensearch-project/flow-framework#426

[lezzago]

Why do we need an empty space?

[dbwiddis]

This is the delimiter in string join, adding a space between the sentences. I thought this syntax was easier to read than a bunch of concatenated strings. It's a one-time static constant so performance isn't really an issue.

I could append the strings with a + and try to remember to add a space at the end or beginning of each subsequent one....