[dvd233]

Summary

Reject document statements that run into the next assignment or table header without a newline, such as a = 1 b = 2 and [table] a = 1.

A private boundary check runs after successful document-level key/value pairs and table or array-table headers, before existing whitespace/comment skipping can lose the newline evidence. It permits LF, CRLF, EOF, and trailing comments after optional spaces or tabs. Comments remain on the existing scanner path. Reusable pair/value parsers, generic repetition, the scanner, inline tables, and multiline value parsing are unchanged.

The tests cover missing separators, valid comments and line endings, EOF, multiline values, and error positions. Removing four ignore entries enables six existing fixtures under invalid/key/: after-array, after-table, after-value, no-eol-01, no-eol-03, and no-eol-04.

This addresses one document-statement separation defect tracked by #6663. It is partial progress on that issue and does not depend on #7347 or #7348.

Validation

• Focused repository test run, including the added regressions and enabled fixtures: 650 passed / 9 failed before the parser change; 659 passed / 0 failed afterward, with 105 existing unrelated ignores.
• Statement-only TOML tests plus docs, rerun after committing: 697 passed, 0 failed, 105 ignored.
• A 5,012-input corpus matches Python tomllib acceptance and normalized values after the fix. Its repeated cases exercise the same separator defect and preserved behavior.
• Repository formatting, changed-file configured lint, and offline browser-compatible entry-point typechecking passed. Two independent source reviews found no actionable defects.
• Exact deno task ok passed in the isolated fork validation run, using the unchanged reviewed source plus a separate validation workflow: 7,504 passed (3,288 steps), 0 failed, 116 ignored.

AI assistance was used for implementation, tests, review, and this PR description.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 95.04%. Comparing base (f834d02) to head (cd2920b).

Additional details and impacted files

@@           Coverage Diff            @@
##             main    #7349    +/-   ##
========================================
  Coverage   95.04%   95.04%            
========================================
  Files         619      618     -1     
  Lines       52012    51771   -241     
  Branches     9450     9405    -45     
========================================
- Hits        49434    49206   -228     
+ Misses       2031     2022     -9     
+ Partials      547      543     -4     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.