Add paperspec correct, check math forms and datasets, and give the tools public names

The work the other paper tools planned for paperspec (paperdesk PLAN M5.4,
paperform PLAN 4.2):

- paperspec correct and paperspec.revision apply proposed corrections to a
  package, folder or zip, and write it only once it validates.
- doc.math.form is checked: label and style commands always, the form's own
  spelling when papermath is installed or forms= is passed.
- paperspec.datasets and seven doc.dataset.* rules check a package's datasets.
- fixtures/invalid gains suspect-witness and math-form-label (33 in all).
- Canonical JSON reads numbers as doubles, so 1e20 round-trips.
- The registries and the JSON-LD context ship in the wheel.
- Public names for what paperdesk and paperform imported as private ones,
  with the old names kept as aliases.
- The text and lossy-projection functions are about three times faster
  through the gate, tested equal to the old code on every code point.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013XwDhn4LkrKQfw4ArDEr2E
This commit is contained in:
alexmandClaude committed 2026-10-20 22:10:17 -04:00
1 parent 8c07dd7c15
commit 4fb8b4f5a7
70 files changed
+4470 -270

No files matched your search

+63
View File
@@ -692,3 +692,66 @@ class TestMetaHash:
other = dict(self.DOCUMENT, policy="withheld", key="another")
assert meta_hash(other) == meta_hash(self.DOCUMENT)
assert meta_hash(dict(self.DOCUMENT, language="fr")) != meta_hash(self.DOCUMENT)
def _lossy_by_hand(s):
"""The lossy projection of a strict projection, character by character (spec/05 §5.5)."""
out, stack, i, n = [], [], 0, len(s)
while i < n:
ch = s[i]
if ch in "_^" and i + 1 < n and s[i + 1] == "{":
stack.append("s")
i += 2
continue
if ch == "{":
stack.append("t")
out.append(ch)
elif ch == "}":
if not stack or stack.pop() == "t":
out.append(ch)
elif ch == hashing.MARK_OPEN:
j = s.find(hashing.MARK_CLOSE, i)
i = n if j < 0 else j + 1
continue
else:
out.append(ch)
i += 1
t = unicodedata.normalize("NFKC", "".join(out).casefold())
return "".join(
c for c in t if not is_ws(c) and unicodedata.category(c) != "Pd" and c != "­"
)
LOSSY_ALPHABET = [
"_",
"^",
"{",
"}",
hashing.MARK_OPEN,
hashing.MARK_CLOSE,
hashing.OBJ,
"a",
"B",
"\u00df", # sharp s, which case folding expands
"\u00ad", # soft hyphen
"-",
"\u2013", # en dash
"\u2212", # minus sign
" ",
"\u00a0", # no-break space
"\u2009", # thin space
"\ufb01", # the fi ligature, which NFKC expands
"\u00b2", # superscript two
"1",
]
@given(st.text(alphabet=st.sampled_from(LOSSY_ALPHABET), max_size=40))
@settings(derandomize=True, max_examples=500, database=None, deadline=None)
def test_the_lossy_projection_matches_its_character_by_character_definition(s):
assert hashing.lossy_from_strict(s) == _lossy_by_hand(s)
def test_the_lossy_projection_drops_every_dash_and_white_space_character():
every = "".join(chr(c) for c in range(0x110000) if not 0xD800 <= c <= 0xDFFF)
assert hashing.lossy_from_strict(every) == _lossy_by_hand(every)