arborist/tests/test_wikitext.py
russell@unturf.com 22a8071936
verify: strip wikitext from context before substring matching
Adds aborist/wikitext.py with to_base() — deterministic wikitext →
prose conversion via mwparserfromhell. Drops <ref>...</ref>,
[[File:...]], [[Image:...]], [[Category:...]] entirely; resolves
piped wikilinks to display text; collapses templates, formatting,
and HTML markup. extract_wikilinks() preserves the link graph for
the definition-cloud artifact (recoverable from any page on demand
without re-parsing wikitext at query time).

Wires to_base() into verify_quotes(). Without the strip, the
verifier compared the model's clean prose against [[Cloud Strife]],
[[Shinra Electric Power Company|Shinra]], etc. and falsely flagged
real source quotes as VISUAL. Concrete case from a make-query run:
'Cloud Strife, an unsociable mercenary who claims to be a former
1st Class member of Shinra's SOLDIER unit;' is verbatim in the
Final_Fantasy_VII article wikitext (modulo markup). With the strip
that quote now verifies; the genuine model hallucinations in the
same answer still flag honestly. Side-effect: 43% smaller context
size on average so less LLM token waste.

BASE_VERSION = 'wikitext-base-v1' is the algorithm pin. Bump when
the strip rules change. Soft-imported in verify.py so environments
without mwparserfromhell installed degrade gracefully (no strip,
same behavior as before this commit).

Build:
- pyproject.toml: new [wikitext] extras (mwparserfromhell>=0.6),
  pulled in by [dev]
- Makefile: chain-check / chain-check-shards targets — fast
  audit-chain integrity probe, counts dangling prev_event_hash
  references; 0 = chain intact

Tests:
- tests/test_wikitext.py: 31 tests (rules, idempotence, real-corpus
  fixture)
- tests/test_verify.py: 2 regression tests pinning the FF7 flip
  (Cloud Strife quote: VISUAL → STRICT after strip; genuine
  hallucination: stays VISUAL)
- tests/fixtures/ff7_characters_chunk0.wikitext: real chunk from a
  shard, used to validate the strip on actual Wikipedia content
2026-04-28 15:48:07 -04:00

256 lines
7.9 KiB
Python

"""Unit tests for aborist.wikitext — wikitext-base-v1.
Most tests are tiny synthetic inputs that pin one rule each. The
``test_real_chunk_*`` tests use a real Wikipedia chunk (FF7 characters)
pulled from a live shard so we catch realistic interactions: nested
templates, mixed namespaces, refs inside paragraphs.
The fixture file ``tests/fixtures/ff7_characters_chunk0.wikitext`` is
chunk 0 of ``Characters_of_the_Final_Fantasy_VII_series`` and was the
input that triggered the wikitext-base-v1 design (verifier flagged 6
"quotes" as VISUAL because the model paraphrased the wikitext form).
"""
from __future__ import annotations
from pathlib import Path
import pytest
from aborist.wikitext import BASE_VERSION, extract_wikilinks, to_base
FIXTURE = Path(__file__).parent / "fixtures" / "ff7_characters_chunk0.wikitext"
def test_base_version_pinned():
"""If you change BASE_VERSION you stale every cached answer. The pin
here is a tripwire: bumping requires intent."""
assert BASE_VERSION == "wikitext-base-v1"
def test_empty_input():
assert to_base("") == ""
assert to_base(" \n\t ") == ""
def test_plain_prose_round_trips():
src = "This is plain prose with no markup."
assert to_base(src) == src
def test_simple_wikilink():
assert to_base("Hello [[Cloud Strife]] world.") == "Hello Cloud Strife world."
def test_piped_wikilink_uses_display_text():
assert to_base("Hello [[Cloud Strife|Cloud]] world.") == "Hello Cloud world."
def test_sectioned_wikilink_uses_display_text():
src = "See [[Cloud Strife#weapons|the buster sword]]."
assert to_base(src) == "See the buster sword."
def test_file_link_dropped_entirely():
src = "[[File:foo.jpg|thumb|250px|A caption with [[Cloud]] inside.]] body text"
out = to_base(src)
assert "thumb" not in out
assert "250px" not in out
assert "foo.jpg" not in out
assert "body text" in out
def test_image_link_dropped_entirely():
src = "[[Image:bar.png|left|Some image]] body text"
out = to_base(src)
assert "left" not in out
assert "bar.png" not in out
assert "body text" in out
def test_category_link_dropped():
src = "Body. [[Category:Final Fantasy VII characters]]"
out = to_base(src)
assert "Category:" not in out
assert "Body." in out
def test_ref_tag_inline_dropped():
src = "Cloud is great.<ref>Smith 2010, p. 5.</ref> The end."
assert to_base(src) == "Cloud is great. The end."
def test_ref_tag_named_dropped():
src = 'Cloud is great.<ref name="x">Smith 2010</ref> Reuse.<ref name="x"/> End.'
out = to_base(src)
assert "Smith" not in out
assert "Cloud is great." in out
assert "Reuse." in out
assert "End." in out
def test_template_cite_dropped():
src = "Plain prose. {{cite book|title=Foo|author=Bar}} more."
out = to_base(src)
assert "Foo" not in out
assert "Plain prose." in out
assert "more." in out
def test_template_infobox_dropped():
src = "{{Infobox character|name=Cloud|hair=blond}} Body."
assert to_base(src) == "Body."
def test_bold_and_italic_preserve_text():
assert to_base("This is '''bold''' text.") == "This is bold text."
assert to_base("This is ''italic'' text.") == "This is italic text."
def test_header_keeps_text():
out = to_base("== Section Title ==\nBody under it.")
assert "Section Title" in out
assert "Body under it." in out
assert "==" not in out
def test_html_entities_decoded():
out = to_base("Tom &amp; Jerry, 50&nbsp;words.")
assert "Tom & Jerry" in out
assert "&amp;" not in out
def test_html_tags_removed_text_kept():
out = to_base("Inline <small>caveat</small> text. <br/>Newline.")
assert "caveat" in out
assert "<" not in out
assert ">" not in out
def test_external_link_with_text():
assert to_base("Visit [https://example.com our site].") == "Visit our site."
def test_external_link_bare_dropped():
out = to_base("Visit [https://example.com] now.")
assert "https://" not in out
assert "Visit" in out
assert "now." in out
def test_html_comment_dropped():
assert to_base("Visible <!-- hidden --> rest.") == "Visible rest."
def test_idempotence():
"""to_base(to_base(x)) == to_base(x). Critical for the verifier:
if context is already-clean prose (e.g. an HTML source), running
to_base on it must not corrupt the prose."""
raw = (
"Hello [[Cloud Strife|Cloud]]<ref>cite</ref>, see [[File:foo.jpg]] "
"and {{Infobox|x=1}} the end."
)
once = to_base(raw)
twice = to_base(once)
assert once == twice
def test_nested_templates_dropped():
src = "{{cite |title={{lang|en|Foo}}}} body."
assert to_base(src) == "body."
# ---------------------------------------------------------------------------
# Real-corpus fixture: chunk 0 of Characters_of_the_Final_Fantasy_VII_series
# ---------------------------------------------------------------------------
@pytest.fixture
def ff7_raw() -> str:
return FIXTURE.read_text()
def test_real_chunk_strips_markup(ff7_raw):
"""The chunk that triggered this design. After to_base, the named
characters should appear as bare prose, not wikitext."""
base = to_base(ff7_raw)
# Concrete topical content survives.
assert "Cloud Strife" in base
assert "Sephiroth" in base
assert "Final Fantasy VII" in base
# Wikitext markers are gone.
assert "[[" not in base
assert "]]" not in base
assert "{{" not in base
assert "}}" not in base
assert "<ref" not in base
assert "File:" not in base
assert "Category:" not in base
def test_real_chunk_protagonist_phrase_recoverable(ff7_raw):
"""The wikitext source has:
...follows [[protagonist]] [[Cloud Strife]], a troubled mercenary...
After to_base this should read as continuous prose. We verify the
canonical FF7 pitch surfaces verbatim — the verifier can substring-test
against this exact form."""
base = to_base(ff7_raw)
assert "follows protagonist Cloud Strife, a troubled mercenary" in base
def test_real_chunk_substantial_compression(ff7_raw):
"""Sanity check: the base form should be smaller than raw (markup
removed) but not so small that we've lost the prose."""
base = to_base(ff7_raw)
assert len(base) < len(ff7_raw)
# FF7 chunk 0 is 3722 chars raw; base form is roughly 75% of that.
assert len(base) > len(ff7_raw) * 0.5
# ---------------------------------------------------------------------------
# extract_wikilinks — the link graph artifact
# ---------------------------------------------------------------------------
def test_extract_wikilinks_simple():
out = extract_wikilinks("Hello [[Cloud Strife]] world.")
assert out == [("Cloud Strife", None)]
def test_extract_wikilinks_with_display():
out = extract_wikilinks("Hello [[Cloud Strife|Cloud]] world.")
assert out == [("Cloud Strife", "Cloud")]
def test_extract_wikilinks_section_collapsed():
"""[[Foo#bar]] and [[Foo#baz]] both collapse to ('Foo', None)."""
out = extract_wikilinks("See [[Foo#bar]] and [[Foo#baz|qux]].")
assert out == [("Foo", None), ("Foo", "qux")]
def test_extract_wikilinks_includes_namespaces():
"""File: / Image: / Category: are dropped from PROSE but kept in the
link graph — they are exactly the structural signal the graph wants."""
src = "[[File:x.jpg]] [[Category:Y]] [[Image:z.png|caption]] [[Cloud]]"
out = extract_wikilinks(src)
targets = [t for t, _ in out]
assert "File:x.jpg" in targets
assert "Category:Y" in targets
assert "Image:z.png" in targets
assert "Cloud" in targets
def test_extract_wikilinks_empty_input():
assert extract_wikilinks("") == []
assert extract_wikilinks(" ") == []
assert extract_wikilinks("plain prose, no links.") == []
def test_extract_wikilinks_real_chunk(ff7_raw):
out = extract_wikilinks(ff7_raw)
assert len(out) > 10
targets = {t for t, _ in out}
# A few links we expect from chunk 0.
assert "Cloud Strife" in targets
assert "Final Fantasy VII" in targets