[BUG]Regarding automatic file truncation

[chenxluo]

Problem (one or two sentences)

The automatic truncation feature for overly long content doesn't seem to work for JSON files. Even though I've set a limit of 50 lines, whenever the model requests to read a JSON file, the context almost always ends up being too long.

Context (who is affected and when)

Reproduction steps

1.On Windows 11

2.There is a multi-line JSON file in the project directory larger than 2MB

3.The model's request to read the file was approved

Expected result

The file should be properly truncated, and the added context should remain under 10k tokens.

Actual result

The context rapidly exceeds the maximum limit.

Variations tried (optional)

No response

App Version

v3.28.3

API Provider (optional)

Google Gemini

Model Used (optional)

No response

Roo Code Task Links (optional)

No response

Relevant logs or errors (optional)

[roomote]

I'm exploring a potential fix for the JSON file truncation issue and will open a PR to share progress.

[roomote]

I've investigated this issue and created PR #8151 with comprehensive tests to verify the JSON file truncation behavior.

The good news is that the tests confirm the truncation feature is working correctly for JSON files. When a JSON file exceeds the configured maxReadFileLine limit, it is properly truncated as expected.

The PR adds test coverage for various scenarios including:

• Large JSON files (>50 lines) being truncated at the limit
• Small JSON files within the limit not being truncated
• Very large JSON files (simulating >2MB) being handled correctly
• Complex nested JSON structures being truncated at the correct line

All tests pass successfully. The issue you're experiencing might be related to how the setting is being applied in the UI or could have been fixed in a recent update. Please check the PR for more details: #8151