cmos dev
4cad38ef30
Initial phase-1 baseline of the karpathy/autoresearch-style loop. The formatter module is the inner-loop artifact; parser and linter are infra. The linter carries a LINTER_VERSION hash (v0.2.0) that will force a re-baseline on any rule change. Components: - harness/diff.py: case-sensitive field-level substring diff - harness/score.py: three-axis scoring (field, linter, canary exact) - src/cmos/linter.py: 9 CMOS 18 structural rules, each Purdue/CMOS cited - src/cmos/parser.py: locate ## Bibliography section, split entries - src/cmos/formatter.py: prompt + OpenAI call with caller injection - src/cmos/cli.py: cmos format path/to/draft.md - scripts/run_loop.py: loop runner with --fake mode for no-API runs - exemplars/: 3 canary seed exemplars (book, journal w/DOI, web), sourced from chicagomanualofstyle.org quick guide Tests: 48 passing. Fake-mode baseline scalar = 0.000 on the 3 seed exemplars (identity caller fails the linter on every rule). This is the floor the real GPT-5 formatter needs to improve from.
267 lines
9.1 KiB
Python
267 lines
9.1 KiB
Python
"""Tests for src/cmos/linter.py — deterministic CMOS 18 structural rules.
|
||
|
||
Each rule has a positive and a negative unit test. Doc comments cite the
|
||
source that justifies the rule (Purdue OWL page or CMOS quick guide URL).
|
||
|
||
The linter only checks structural properties of the RAW output string
|
||
(italic markers as emitted, punctuation as emitted). Fact survival
|
||
(author/title/year/etc. matching canonical) is the diff's job, not the
|
||
linter's, so we deliberately don't duplicate it here.
|
||
"""
|
||
|
||
from cmos.linter import (
|
||
LINTER_VERSION,
|
||
check_passes,
|
||
lint,
|
||
rule_article_title_quoted,
|
||
rule_book_title_italicized,
|
||
rule_doi_is_https_url,
|
||
rule_ends_with_period,
|
||
rule_journal_title_italicized,
|
||
rule_no_ibid,
|
||
rule_no_place_of_publication_for_books,
|
||
rule_page_range_uses_en_dash,
|
||
rule_web_page_title_quoted,
|
||
)
|
||
from harness.score import Exemplar
|
||
|
||
BOOK = Exemplar(
|
||
name="book",
|
||
source="test",
|
||
type="book",
|
||
tags=["book"],
|
||
canary=False,
|
||
messy_input="",
|
||
canonical={
|
||
"author": "Yu, Charles",
|
||
"title": "Interior Chinatown",
|
||
"publisher": "Pantheon Books",
|
||
"year": 2020,
|
||
},
|
||
expected_bibliography="Yu, Charles. *Interior Chinatown*. Pantheon Books, 2020.",
|
||
)
|
||
|
||
JOURNAL = Exemplar(
|
||
name="journal",
|
||
source="test",
|
||
type="journal",
|
||
tags=["journal", "doi"],
|
||
canary=False,
|
||
messy_input="",
|
||
canonical={
|
||
"author": "Kwon, Hyeyoung",
|
||
"article_title": "Inclusion Work: Children of Immigrants Claiming Membership in Everyday Life",
|
||
"journal": "American Journal of Sociology",
|
||
"volume": 127,
|
||
"issue": 6,
|
||
"year": 2022,
|
||
"pages": "1818–59",
|
||
"doi": "https://doi.org/10.1086/720277",
|
||
},
|
||
expected_bibliography=(
|
||
'Kwon, Hyeyoung. "Inclusion Work: Children of Immigrants Claiming Membership in Everyday Life." '
|
||
"*American Journal of Sociology* 127, no. 6 (2022): 1818–59. https://doi.org/10.1086/720277."
|
||
),
|
||
)
|
||
|
||
WEB = Exemplar(
|
||
name="web",
|
||
source="test",
|
||
type="web",
|
||
tags=["web"],
|
||
canary=False,
|
||
messy_input="",
|
||
canonical={
|
||
"author": "Google",
|
||
"title": "Privacy Policy",
|
||
"site": "Privacy & Terms",
|
||
"date": "November 15, 2023",
|
||
"url": "https://policies.google.com/privacy",
|
||
},
|
||
expected_bibliography='Google. "Privacy Policy." Privacy & Terms. Effective November 15, 2023. https://policies.google.com/privacy.',
|
||
)
|
||
|
||
|
||
def test_linter_version_is_set():
|
||
assert LINTER_VERSION
|
||
assert isinstance(LINTER_VERSION, str)
|
||
|
||
|
||
def test_ends_with_period_passes_on_well_formed_entry():
|
||
# CMOS 18: bibliography entries separate major elements by periods and
|
||
# end with a period. Source: chicagomanualofstyle.org quick guide.
|
||
result = rule_ends_with_period("Yu, Charles. *Interior Chinatown*. Pantheon Books, 2020.", BOOK)
|
||
assert result.applicable is True
|
||
assert result.passed is True
|
||
|
||
|
||
def test_ends_with_period_fails_on_missing_period():
|
||
result = rule_ends_with_period("Yu, Charles. *Interior Chinatown*. Pantheon Books, 2020", BOOK)
|
||
assert result.applicable is True
|
||
assert result.passed is False
|
||
|
||
|
||
# --- No place of publication for books (CMOS 18 change from 17, CMOS 14.30) ----
|
||
|
||
|
||
def test_no_place_of_publication_passes_when_absent():
|
||
# CMOS 18: place of publication no longer required for books.
|
||
# Source: CMOS 14.30 (cited on chicagomanualofstyle.org quick guide).
|
||
result = rule_no_place_of_publication_for_books(
|
||
"Yu, Charles. *Interior Chinatown*. Pantheon Books, 2020.", BOOK
|
||
)
|
||
assert result.applicable is True
|
||
assert result.passed is True
|
||
|
||
|
||
def test_no_place_of_publication_fails_when_city_prefix_present():
|
||
result = rule_no_place_of_publication_for_books(
|
||
"Yu, Charles. *Interior Chinatown*. New York: Pantheon Books, 2020.", BOOK
|
||
)
|
||
assert result.passed is False
|
||
|
||
|
||
def test_no_place_of_publication_not_applicable_to_journals():
|
||
# The rule should not fire on non-book types even if their output happens
|
||
# to contain a colon (which journals routinely do, e.g., article title).
|
||
result = rule_no_place_of_publication_for_books("anything: anything.", JOURNAL)
|
||
assert result.applicable is False
|
||
|
||
|
||
# --- Book title italicized ----------------------------------------------------
|
||
|
||
|
||
def test_book_title_italicized_passes_when_wrapped_in_asterisks():
|
||
# CMOS 18 bibliography basics: titles of books and journals are italicized.
|
||
# Source: chicagomanualofstyle.org quick guide; Purdue OWL CMOS 18 NB poster
|
||
# page 3 "Bibliography: Basics".
|
||
result = rule_book_title_italicized(
|
||
"Yu, Charles. *Interior Chinatown*. Pantheon Books, 2020.", BOOK
|
||
)
|
||
assert result.passed is True
|
||
|
||
|
||
def test_book_title_italicized_fails_when_title_plain():
|
||
result = rule_book_title_italicized(
|
||
"Yu, Charles. Interior Chinatown. Pantheon Books, 2020.", BOOK
|
||
)
|
||
assert result.passed is False
|
||
|
||
|
||
def test_book_title_italicized_not_applicable_to_web():
|
||
result = rule_book_title_italicized("whatever", WEB)
|
||
assert result.applicable is False
|
||
|
||
|
||
# --- No "Ibid." (CMOS 18 deprecation) -----------------------------------------
|
||
|
||
|
||
def test_no_ibid_passes_on_clean_entry():
|
||
# CMOS 18 deprecates "Ibid." in favor of shortened notes.
|
||
# Source: Widener "New in 18th" LibGuide.
|
||
result = rule_no_ibid("Yu, Charles. *Interior Chinatown*. Pantheon Books, 2020.", BOOK)
|
||
assert result.passed is True
|
||
|
||
|
||
def test_no_ibid_fails_when_ibid_present():
|
||
result = rule_no_ibid("Ibid., 42.", BOOK)
|
||
assert result.passed is False
|
||
|
||
|
||
# --- Journal title italicized --------------------------------------------------
|
||
|
||
|
||
def test_journal_title_italicized_passes_when_wrapped():
|
||
result = rule_journal_title_italicized(JOURNAL.expected_bibliography, JOURNAL)
|
||
assert result.passed is True
|
||
|
||
|
||
def test_journal_title_italicized_fails_when_plain():
|
||
bad = JOURNAL.expected_bibliography.replace(
|
||
"*American Journal of Sociology*", "American Journal of Sociology"
|
||
)
|
||
result = rule_journal_title_italicized(bad, JOURNAL)
|
||
assert result.passed is False
|
||
|
||
|
||
# --- Article title in quotes ---------------------------------------------------
|
||
|
||
|
||
def test_article_title_quoted_passes_when_wrapped_in_double_quotes():
|
||
# CMOS 18: titles of articles, chapters, poems are in quotation marks.
|
||
# Source: Purdue OWL CMOS 18 NB poster page 3 Bibliography Basics.
|
||
result = rule_article_title_quoted(JOURNAL.expected_bibliography, JOURNAL)
|
||
assert result.passed is True
|
||
|
||
|
||
def test_article_title_quoted_fails_when_unquoted():
|
||
bad = JOURNAL.expected_bibliography.replace(
|
||
'"Inclusion Work: Children of Immigrants Claiming Membership in Everyday Life."',
|
||
"Inclusion Work: Children of Immigrants Claiming Membership in Everyday Life.",
|
||
)
|
||
result = rule_article_title_quoted(bad, JOURNAL)
|
||
assert result.passed is False
|
||
|
||
|
||
# --- Page range uses en-dash ---------------------------------------------------
|
||
|
||
|
||
def test_page_range_uses_en_dash_passes():
|
||
# CMOS convention: page ranges use en-dash (–), not ASCII hyphen (-).
|
||
result = rule_page_range_uses_en_dash(JOURNAL.expected_bibliography, JOURNAL)
|
||
assert result.passed is True
|
||
|
||
|
||
def test_page_range_uses_en_dash_fails_on_ascii_hyphen():
|
||
bad = JOURNAL.expected_bibliography.replace("1818–59", "1818-59")
|
||
result = rule_page_range_uses_en_dash(bad, JOURNAL)
|
||
assert result.passed is False
|
||
|
||
|
||
# --- DOI as https://doi.org/ URL -----------------------------------------------
|
||
|
||
|
||
def test_doi_is_https_url_passes_when_canonical_prefix_present():
|
||
# CMOS 18 prefers DOIs formatted as https://doi.org/... rather than bare
|
||
# DOIs or "DOI:" prefixes. Source: CMOS 18 quick guide.
|
||
result = rule_doi_is_https_url(JOURNAL.expected_bibliography, JOURNAL)
|
||
assert result.passed is True
|
||
|
||
|
||
def test_doi_is_https_url_fails_on_bare_doi():
|
||
bad = JOURNAL.expected_bibliography.replace("https://doi.org/10.1086/720277", "10.1086/720277")
|
||
result = rule_doi_is_https_url(bad, JOURNAL)
|
||
assert result.passed is False
|
||
|
||
|
||
def test_doi_rule_not_applicable_without_doi():
|
||
# Web exemplar has no DOI field.
|
||
result = rule_doi_is_https_url(WEB.expected_bibliography, WEB)
|
||
assert result.applicable is False
|
||
|
||
|
||
# --- Web page title in quotes --------------------------------------------------
|
||
|
||
|
||
def test_web_page_title_quoted_passes():
|
||
result = rule_web_page_title_quoted(WEB.expected_bibliography, WEB)
|
||
assert result.passed is True
|
||
|
||
|
||
def test_web_page_title_quoted_fails_when_unquoted():
|
||
bad = WEB.expected_bibliography.replace('"Privacy Policy."', "Privacy Policy.")
|
||
result = rule_web_page_title_quoted(bad, WEB)
|
||
assert result.passed is False
|
||
|
||
|
||
# --- Aggregate check_passes ---------------------------------------------------
|
||
|
||
|
||
def test_check_passes_on_canonical_outputs():
|
||
# All three seed exemplars' expected_bibliography strings must lint clean.
|
||
# If any of them fail, either the linter is too strict or the exemplar is
|
||
# wrong — either way, we want to know immediately.
|
||
assert check_passes(BOOK.expected_bibliography, BOOK)
|
||
assert check_passes(JOURNAL.expected_bibliography, JOURNAL)
|
||
assert check_passes(WEB.expected_bibliography, WEB)
|