fix(translation): accept SSE data fields with no space after the colon

[pucedoteth]

Closes #399.

Problem

parse_json_sse_frame matched the data field with a literal prefix that includes the space:

.filter_map(|line| line.strip_prefix("data: ").map(|l| l.to_string()))

The SSE spec treats that space as optional framing, not part of the delimiter:

Collect the characters on the line before the first U+003A COLON character (:), and let field be that string.
Collect the characters on the line after the first U+003A COLON character (:), and let value be that string. If value starts with a U+0020 SPACE character, remove it from value.

So data:{"delta":"hi"} is a well-formed frame that this parser dropped. The frame then folded to an empty string and decoded to SseFrame::Empty, silently discarding the event — including a data:[DONE] terminator, which stopped being recognised as the end of the stream.

Fix

Split on the first colon and match the field name, stripping at most one leading space from the value:

let value = match line.split_once(':') {
    Some(("data", value)) => value,
    None if line == "data" => "",
    _ => return None,
};
Some(value.strip_prefix(' ').unwrap_or(value).to_string())

The None arm covers the spec's other case — a field line with no colon is the field name with an empty value.

Matching the field name rather than a prefix also keeps a line like database: {...} from being read as a data field. The old strip_prefix("data: ") already got that right by accident; it is now covered by a test so the new parsing can't regress it.

Comment lines are unaffected: they are filtered before this point, and :comment splits to a field name of "", which does not match data either way.

Tests

Three tests added to the existing module in sse.rs:

• parses_a_data_line_without_a_space_after_the_colon — the reported case
• done_marker_is_recognised_without_a_space — data:[DONE] still terminates the stream
• field_names_are_matched_exactly — database: is not a data field

The first two fail on main:

test sse::tests::done_marker_is_recognised_without_a_space ... FAILED
test sse::tests::parses_a_data_line_without_a_space_after_the_colon ... FAILED
Error: "expected a payload"

I also wrote a fourth test asserting that only one leading space is framing, then deleted it: this parser hands the value to serde_json, which ignores leading whitespace, so the distinction is not observable here. The behaviour is still spec-correct, it just isn't something this function can demonstrate, and I would rather not leave a test that passes for the wrong reason.

Gates

gate	result
cargo fmt --all --check	clean
cargo clippy --workspace --all-targets -- -D warnings	clean
cargo test --workspace --exclude switchyard-py	501 passed, 0 failed
uv run ruff check .	All checks passed!
uv run mypy switchyard	Success: no issues found in 21 source files

Two pre-existing failures in my environment, both verified identical on a clean checkout of main and unrelated to this change:

• cargo test --workspace fails to link switchyard-py (pyo3 symbol(s) not found for architecture arm64 — missing Python symbols in my local toolchain). Hence the --exclude above; I did not touch that crate.
• uv run pytest tests/ fails tests/e2e/test_closed_book_proxy_integration.py, an e2e test that needs live services.

Commit is signed off per DCO.

Summary by CodeRabbit

• Bug Fixes
  • Improved handling of streamed event data with or without a space after the separator.
  • Correctly processes empty data fields and terminal markers.
  • Prevents similarly named fields from being interpreted as valid event data.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: a86ca5cc-a414-4911-80e7-5465a8e7bd18

📥 Commits

Reviewing files that changed from the base of the PR and between 9ad6744 and 89b0bed.

📒 Files selected for processing (1)

• crates/switchyard-translation/src/sse.rs

Included review availability: Your plan includes up to 12 reviews per rolling hour; 11 remain after this review.

---

Walkthrough

The SSE parser now accepts data fields with or without a space after the colon, accepts colonless data fields as empty values, and matches the field name exactly. Tests cover payloads, terminal markers, and similarly prefixed fields.

Changes

SSE parsing

Layer / File(s)	Summary
Parse SSE data fields crates/switchyard-translation/src/sse.rs	The parser recognizes exact data fields, supports optional separators and one leading space, and preserves empty colonless values. Tests cover no-space JSON payloads, no-space [DONE] markers, and rejection of fields such as database:.

Estimated code review effort: 2 (Simple) | ~10 minutes

Merge Risk: ⚪ Minimal · up to 89b0b

This change broadens SSE parsing to accept valid data fields without a space while preserving exact field matching and terminator handling. No actionable merge-blocking risk remains after normal checks and review.

Poem

I’m a rabbit with ears held high,
Parsing each data stream nearby.
Spaces may come, or spaces may flee,
[DONE] now ends the stream properly.
No false database hops by me!

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly describes the primary change: accepting SSE data fields without a space after the colon.
Linked Issues check	✅ Passed	The parser changes and regression tests satisfy issue #399 by supporting both data formats, bare data lines, and exact field-name matching.
Out of Scope Changes check	✅ Passed	The changes are limited to SSE parsing behavior and focused regression tests required by issue #399.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

---

_{Comment @coderabbitai help to get the list of available commands.}