cmos/tests/test_cli.py

"""Tests for src/cmos/cli.py — end-to-end draft reformatting.

The CLI orchestrates parser → formatter → reassemble. Tests inject a fake
formatter so no API calls happen; the goal is to pin the glue logic, not the
LLM behavior.

v1: ``reformat_draft`` rewrites the ``## Bibliography`` section.
v2: ``reformat_notes`` rewrites pandoc-style markdown footnote definitions
    in place by line number, preserving everything else byte-for-byte.
"""

import subprocess
import sys
from pathlib import Path

import pytest

from cmos.cli import reformat_draft, reformat_notes
from cmos.parser import NoBibliographyError


def _fake_formatter(messy: str) -> str:
    # Deterministic fake: return a cleaned-up marker string so the output is
    # recognizable.
    return f"FORMATTED({messy.strip()})"


def test_reformat_draft_rewrites_only_bibliography_section():
    draft = """\
# My Paper

Some prose with citations.

## Bibliography

yu, charles. interior chinatown. 2020.
kwon, hyeyoung. inclusion work. 2022.
"""
    output = reformat_draft(draft, formatter=_fake_formatter)
    assert "# My Paper" in output
    assert "Some prose with citations." in output
    assert "FORMATTED(yu, charles. interior chinatown. 2020.)" in output
    assert "FORMATTED(kwon, hyeyoung. inclusion work. 2022.)" in output
    # Original messy lines must not also survive.
    assert "yu, charles. interior chinatown. 2020." not in output.replace(
        "FORMATTED(yu, charles. interior chinatown. 2020.)", ""
    )


def test_reformat_draft_preserves_sections_after_bibliography():
    draft = """\
## Bibliography

Yu, Charles. *Interior Chinatown* (New York: Pantheon, 2020).

## Appendix

appendix text.
"""
    output = reformat_draft(draft, formatter=_fake_formatter)
    assert "FORMATTED(Yu, Charles. *Interior Chinatown* (New York: Pantheon, 2020).)" in output
    assert "## Appendix" in output
    assert "appendix text." in output


def test_reformat_draft_raises_when_no_bibliography():
    with pytest.raises(NoBibliographyError):
        reformat_draft("## Intro\n\nno bib here.\n", formatter=_fake_formatter)


def test_reformat_draft_bibliography_heading_preserved():
    draft = '## Bibliography\n\nYu, Charles. *Interior Chinatown* (Pantheon, 2020).\n'
    output = reformat_draft(draft, formatter=_fake_formatter)
    assert "## Bibliography" in output


def test_reformat_draft_preserves_works_cited_heading():
    """When the input heading is ``## Works Cited``, the output must
    keep that heading — not replace it with ``## Bibliography``."""
    draft = '## Works Cited\n\nDavidson, Donald. "On the Very Idea." (1974).\n'
    output = reformat_draft(draft, formatter=_fake_formatter)
    assert "## Works Cited" in output
    assert "## Bibliography" not in output


def test_reformat_draft_preserves_bold_wrapped_heading():
    """Bold markers in the heading should be preserved in the output."""
    draft = '## **Works Cited**\n\nDavidson, Donald. "On the Very Idea." (1974).\n'
    output = reformat_draft(draft, formatter=_fake_formatter)
    assert "## **Works Cited**" in output


def test_python_dash_m_invocation_actually_runs_main():
    """Regression test: `python -m cmos.cli` must actually invoke main().

    Without an `if __name__ == "__main__"` guard at the bottom of cli.py,
    `python -m cmos.cli` imports the module body but never calls main(),
    so the process silently exits 0 with empty stdout. That looks
    indistinguishable from a successful run that produced no output —
    the worst kind of bug, since callers assume the pipeline ran. This
    test forces the guard to exist by invoking the CLI as a subprocess
    with --help and asserting argparse actually fired.

    Also acts as a smoke test for both subcommands appearing in the
    top-level help: catches the case where a future refactor removes
    a subcommand registration without anyone noticing.
    """
    result = subprocess.run(
        [sys.executable, "-m", "cmos.cli", "--help"],
        capture_output=True,
        text=True,
        timeout=10,
    )
    # argparse --help exits 0 whether or not main() ran, so the
    # discriminator is the stdout content. Without main(), stdout is empty.
    assert result.returncode == 0, (
        f"expected exit 0, got {result.returncode}; stderr={result.stderr!r}"
    )
    assert result.stdout, (
        "stdout was empty — `python -m cmos.cli` likely silently exited "
        "without invoking main(). Check that cli.py has an "
        '`if __name__ == "__main__": sys.exit(main())` guard at the bottom.'
    )
    assert "format" in result.stdout
    assert "format-notes" in result.stdout
    assert "usage" in result.stdout.lower()


# --- reformat_notes (v2 CLI plumbing) ----------------------------------------


def _fake_note_formatter(messy: str) -> str:
    # Deterministic fake mirroring the v1 fake_formatter pattern.
    return f"NOTE_FORMATTED({messy.strip()})"


def test_reformat_notes_substitutes_definition_in_place():
    draft = """\
Some prose with a citation.[^1]

[^1]: yu, charles. interior chinatown. 2020. p 45.
"""
    output = reformat_notes(draft, formatter=_fake_note_formatter)
    assert "Some prose with a citation.[^1]" in output
    assert "[^1]: NOTE_FORMATTED(yu, charles. interior chinatown. 2020. p 45.)" in output
    # The original messy definition line must be gone.
    assert "[^1]: yu, charles" not in output


def test_reformat_notes_rewrites_multiple_definitions_in_order():
    draft = """\
Prose.[^1] More.[^2] Again.[^3]

[^1]: first messy.
[^2]: second messy.
[^3]: third messy.
"""
    output = reformat_notes(draft, formatter=_fake_note_formatter)
    assert "[^1]: NOTE_FORMATTED(first messy.)" in output
    assert "[^2]: NOTE_FORMATTED(second messy.)" in output
    assert "[^3]: NOTE_FORMATTED(third messy.)" in output


def test_reformat_notes_preserves_non_definition_lines():
    draft = """\
# Title

Some prose with a citation.[^1]

More prose, no citation here.

[^1]: messy definition.

Conclusion paragraph.
"""
    output = reformat_notes(draft, formatter=_fake_note_formatter)
    assert "# Title" in output
    assert "Some prose with a citation.[^1]" in output
    assert "More prose, no citation here." in output
    assert "Conclusion paragraph." in output
    assert "[^1]: NOTE_FORMATTED(messy definition.)" in output


def test_reformat_notes_preserves_reference_markers_in_prose():
    """The [^1] reference inside the prose must NOT be touched — only
    the [^1]: definition line should be reformatted."""
    draft = """\
Some prose with a citation.[^1] And another.[^2]

[^1]: first.
[^2]: second.
"""
    output = reformat_notes(draft, formatter=_fake_note_formatter)
    # References in prose stay verbatim
    assert "Some prose with a citation.[^1] And another.[^2]" in output
    # Definitions are reformatted
    assert "[^1]: NOTE_FORMATTED(first.)" in output
    assert "[^2]: NOTE_FORMATTED(second.)" in output


def test_reformat_notes_returns_unchanged_when_no_definitions():
    draft = "Just prose with no footnote definitions.\n"
    output = reformat_notes(draft, formatter=_fake_note_formatter)
    assert output == draft


def test_reformat_notes_preserves_trailing_newline():
    with_trailing = "[^1]: messy.\n"
    without_trailing = "[^1]: messy."
    assert reformat_notes(with_trailing, formatter=_fake_note_formatter).endswith("\n")
    assert not reformat_notes(without_trailing, formatter=_fake_note_formatter).endswith("\n")


def test_reformat_notes_auto_detects_numbered_format():
    """Real drafts converted from docx use [N] text format, not pandoc.
    reformat_notes must auto-detect: try pandoc first, fall back to
    numbered. Reassembly must use the original prefix (e.g. "[1] ") so
    the output round-trips in the source's marker syntax."""
    draft = """\
## Notes

[1] yu, charles. interior chinatown. 2020. p 45.
[2] kwon, hyeyoung. inclusion work. 2022, p 1830.
"""
    output = reformat_notes(draft, formatter=_fake_note_formatter)
    # Numbered prefix preserved, NOT rewritten as pandoc.
    assert "[1] NOTE_FORMATTED(yu, charles. interior chinatown. 2020. p 45.)" in output
    assert "[2] NOTE_FORMATTED(kwon, hyeyoung. inclusion work. 2022, p 1830.)" in output
    # No pandoc-style markers should appear in the output (we did not
    # auto-convert numbered to pandoc).
    assert "[^1]" not in output
    assert "[^2]" not in output


def test_reformat_notes_auto_detects_blockquote_format():
    """PDF-to-markdown drafts use ``> N text`` blockquote footnotes.
    reformat_notes must auto-detect after pandoc and numbered fail."""
    draft = """\
Some prose.

> 1 Davidson, "On the Very Idea of a Conceptual Scheme."
> 2 Frankenberry 1999, 526.
"""
    output = reformat_notes(draft, formatter=_fake_note_formatter)
    assert '> 1 NOTE_FORMATTED(Davidson, "On the Very Idea of a Conceptual Scheme.")' in output
    assert "> 2 NOTE_FORMATTED(Frankenberry 1999, 526.)" in output
    # No pandoc or numbered markers should appear.
    assert "[^1]" not in output
    assert "[1] " not in output


def test_reformat_notes_pandoc_input_still_round_trips_as_pandoc():
    """Regression: after the auto-detect change, pandoc-format input
    must still produce pandoc-format output. The original_prefix path
    must work for both formats."""
    draft = "[^1]: messy.\n"
    output = reformat_notes(draft, formatter=_fake_note_formatter)
    assert "[^1]: NOTE_FORMATTED(messy.)" in output
    assert "[1] " not in output  # numbered prefix must NOT appear


def test_reformat_notes_preserves_order_under_concurrency():
    """With concurrent execution the formatter is called on all definitions
    in parallel; the output must reassemble them in input order regardless
    of which call finishes first. Mirrors the v1 order-preservation test."""
    import time

    def slow_fake(messy: str) -> str:
        # Earlier definitions sleep longer so they finish last under naive
        # completion-order tracking.
        n = int(messy.split()[-1])
        time.sleep(0.05 * (5 - n))
        return f"NOTE_FORMATTED({messy})"

    draft = """\
Prose.

[^1]: definition 0
[^2]: definition 1
[^3]: definition 2
[^4]: definition 3
[^5]: definition 4
"""
    output = reformat_notes(draft, formatter=slow_fake, concurrency=4)
    # Check order: each marker should be paired with its correct definition.
    assert "[^1]: NOTE_FORMATTED(definition 0)" in output
    assert "[^2]: NOTE_FORMATTED(definition 1)" in output
    assert "[^3]: NOTE_FORMATTED(definition 2)" in output
    assert "[^4]: NOTE_FORMATTED(definition 3)" in output
    assert "[^5]: NOTE_FORMATTED(definition 4)" in output


def test_reformat_draft_preserves_order_under_concurrency():
    # With concurrent execution the formatter is called on all entries in
    # parallel; the CLI must reassemble them in input order regardless of
    # which call finishes first. This test uses a fake formatter that
    # sleeps based on the entry content so earlier entries finish AFTER
    # later ones if order were naively tied to completion.
    import time

    # Map each entry to a sleep duration so earlier entries finish last.
    order = {"2020": 4, "2021": 3, "2022": 2, "2023": 1, "2024": 0}

    def slow_fake(messy: str) -> str:
        year = messy.strip()[-5:-1]  # extract "2020" etc.
        time.sleep(0.05 * order.get(year, 0))
        return f"FORMATTED({messy})"

    draft = """\
## Bibliography

Author, A. *Title Zero* (Publisher, 2020).
Author, B. *Title One* (Publisher, 2021).
Author, C. *Title Two* (Publisher, 2022).
Author, D. *Title Three* (Publisher, 2023).
Author, E. *Title Four* (Publisher, 2024).
"""
    output = reformat_draft(draft, formatter=slow_fake, concurrency=4)
    # Entries must appear in input order.
    lines = [l for l in output.splitlines() if l.startswith("FORMATTED(")]
    assert lines[0].startswith("FORMATTED(Author, A.")
    assert lines[1].startswith("FORMATTED(Author, B.")
    assert lines[2].startswith("FORMATTED(Author, C.")
    assert lines[3].startswith("FORMATTED(Author, D.")
    assert lines[4].startswith("FORMATTED(Author, E.")