fix(preprocess): capture shared-prefix ticker lists

[bozarnr]

Summary

• Extend exchange-qualified ticker hint extraction to capture comma-separated symbols that share the same supported exchange prefix.
• Stop the shared-prefix scan at semicolons or closing parentheses so separate exchange groups are not merged.
• Add deterministic regressions for the public examples listed in bug: exchange-qualified multi-symbol lists are only partially captured #108.

Fixes #108

Testing

• python -m pytest tests/preprocess/test_news.py -k exchange_ticker_hints -o addopts=
• python -m pytest tests/preprocess/test_news.py -o addopts=
• python -m ruff check quantmind/preprocess/news.py tests/preprocess/test_news.py
• git diff --check

Note: running the targeted pytest command with the repository default addopts produced 3 passing selected tests, then failed the global 75% coverage threshold because only this one test slice was executed.

[Copilot]

Pull request overview

This PR improves quantmind.preprocess.news.extract_exchange_ticker_hints to capture additional comma-separated ticker symbols that share the same supported exchange prefix (e.g., (NYSE: EVEX, EVEXW; ...)), and adds deterministic regressions based on the public examples in issue #108.

Changes:

• Extend exchange-qualified ticker hint extraction to include comma-separated list members following a supported exchange prefix.
• Stop the shared-prefix list scan at ; or ) to avoid merging separate groups.
• Add regression tests covering the five example shapes from #108.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
quantmind/preprocess/news.py	Adds shared-prefix list-member extraction for exchange-qualified ticker hints.
tests/preprocess/test_news.py	Adds regression cases for shared-prefix lists and related multi-exchange examples.

Suppressed comments (1)

quantmind/preprocess/news.py:452

• The shared-prefix scan currently runs for every exchange match and starts scanning from match.end(). If the match already consumed a closing ), this will scan beyond the parenthetical (e.g. "(NASDAQ: ABC), a leading..."), and the comma after the ) can cause unrelated words to be captured as extra tickers. Consider only scanning list members when the match started a parenthetical group and did not already close it.

        tail = re.split(r"[;)]", scan_text[match.end() :], maxsplit=1)[0]
        for list_match in _EXCHANGE_TICKER_LIST_MEMBER_RE.finditer(tail):
            append_hint(
                symbol=list_match.group(1).upper(),
                exchange=exchange,
                raw=f"{raw_exchange}: {list_match.group(1).upper()}",
            )

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.