[Hardik K (hardikkaurani)]

Summary of Changes

This pull request resolves a substring bounds calculation bug and index-out-of-bounds edge cases in TextParseNode, and adds test coverage for TextSerializationWriter:

1. Quote stripping bounds:
   • In TextParseNode, when rawText starts and ends with double quotes, rawText.substring(1, rawText.length() - 2) previously evaluated with an off-by-one upper bound. In Java, substring(begin, end) is exclusive of end, meaning length() - 2 dropped the trailing character of the inner content (e.g. "\"hello world\"" became "hello worl"). The end index is corrected to rawText.length() - 1.
2. Bounds safety:
   • Added rawText.length() >= 2 guard before attempting quote stripping. Previously:
     • An empty quoted string ("\"\"", length 2) evaluated rawText.length() - 2 == 0, throwing StringIndexOutOfBoundsException on substring(1, 0).
     • A single quote string ("\"", length 1) matched both startsWith("\"") and endsWith("\""), throwing StringIndexOutOfBoundsException on substring(1, -1).
3. Test coverage:
   • Added test cases in TextParseNodeTest for quoted strings, empty quoted strings, single quote strings, and unquoted strings.
   • Added TextSerializationWriterTest covering string, boolean, integer, and byte array serialization, single-value restrictions, and structured data rejection.

Verification

• Spotless formatting checks pass: ./gradlew :components:serialization:text:spotlessCheck.
• All tests in components/serialization/text pass: 15 passing tests.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota.

[Hardik K (hardikkaurani)]

Hi Vincent Biret (@baywet),

This PR addresses an index calculation issue and edge cases in TextParseNode:

1. Quote stripping bounds: In Java, String.substring(begin, end) has an exclusive upper bound. Calling substring(1, rawText.length() - 2) was stripping the last character of any quoted string (e.g. "\"hello world\"" parsed as "hello worl"). This is corrected to rawText.length() - 1.
2. Length check and index safety: Added rawText.length() >= 2 guard. Previously, empty quotes ("\"\"") and single quotes ("\"") resulted in StringIndexOutOfBoundsException because length() - 2 evaluated to 0 or -1.
3. Test coverage:
   • Added unit tests in TextParseNodeTest for quoted strings, empty quoted strings, single quote strings, and unquoted strings.
   • Added TextSerializationWriterTest to validate primitives, single-value enforcement, and structured data handling.

All tests pass (15 passed) and Spotless formatting is verified locally.

[Copilot]

🟢 Approval recommended

The narrow bounds correction handles the reported edge cases correctly, and the added tests cover the affected behavior with no unresolved findings.

No open issues 🎉

_{Review effort: Balanced}

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota.

[Hardik K (hardikkaurani)]

Hi Vincent Biret (@baywet),

I have updated the branch to address the SpotBugs \OS_OPEN_STREAM\ failure reported in :components:serialization:text:spotbugsTest. Specifically, the stream reading helper in \TextSerializationWriterTest\ now uses try-with-resources to ensure both the underlying \InputStream\ and readers are safely closed on all execution paths.

Local verification:

• ./gradlew :components:serialization:text:spotlessCheck\ passes.
• ./gradlew :components:serialization:text:spotbugsTest\ passes cleanly with 0 violations.
• ./gradlew :components:serialization:text:test\ passes all test suites.

All 4 commits are clean and signed-off. Ready for review when you have a moment.

[Copilot]

🟢 Approval recommended

The narrow bounds correction handles the identified edge cases, and the added regression tests cover them with no unresolved issues.

No open issues 🎉

_{Review effort: Balanced}