100 lines
5.2 KiB
Python
100 lines
5.2 KiB
Python
 
|
"""The client's reading of Klammertext's character codes (character_codes.js)
|
|||
|
|
against the tables `kdesc -c` prints and the rules of the live Klammertext
|
|||
|
|
tree, K/mac (character.cpp, katom_process.cpp; K/src is deprecated), as
|
|||
|
|
`ktext -s` showed them on 2026-10-13. Needs node (the type checker's).
|
|||
|
|
"""
|
|||
|
|
|
|||
|
|
import json
|
|||
|
|
import os
|
|||
|
|
import subprocess
|
|||
|
|
import unittest
|
|||
|
|
|
|||
|
|
CLIENT = os.path.join(os.path.dirname(os.path.dirname(os.path.abspath(__file__))), "client", "js")
|
|||
|
|
|
|||
|
|
HARNESS = """
|
|||
|
|
import { codeAtEnd } from %s;
|
|||
|
|
const texts = JSON.parse(await new Promise((resolve) => {
|
|||
|
|
let data = '';
|
|||
|
|
process.stdin.on('data', (chunk) => { data += chunk; });
|
|||
|
|
process.stdin.on('end', () => resolve(data));
|
|||
|
|
}));
|
|||
|
|
console.log(JSON.stringify(texts.map((text) => codeAtEnd(text))));
|
|||
|
|
"""
|
|||
|
|
|
|||
|
|
# `kdesc -c`, 2026-10-13: Displayed, Written.
|
|||
|
|
DIACRITICS = [("é", "^e'"), ("à", "^a`"), ("ô", "^oh"), ("ñ", "^n~"), ("ü", '^u"'), ("ç", "^cc"),
|
|||
|
|
("ō", "^o-"), ("ğ", "^gb"), ("å", "^ar"), ("ė", "^ed"), ("š", "^sw"), ("ő", "^oa")]
|
|||
|
|
EXTENDED_LATIN = [("ß", "^s^"), ("ı", "^i^"), ("İ", "^I^"), ("þ", "^t^"), ("Þ", "^T^"), ("ð", "^e^"),
|
|||
|
|
("Ð", "^E^"), ("ø", "^o^"), ("Ø", "^O^"), ("đ", "^d^"), ("Đ", "^D^"), ("æ", "^ae^"),
|
|||
|
|
("Æ", "^AE^"), ("œ", "^oe^"), ("Œ", "^OE^")]
|
|||
|
|
PINYIN = [("ā", "^a1"), ("á", "^a2"), ("ǎ", "^a3"), ("à", "^a4")]
|
|||
|
|
|
|||
|
|
|
|||
|
|
def codes_at_end(texts):
|
|||
|
|
harness = HARNESS % json.dumps(os.path.join(CLIENT, "character_codes.js"))
|
|||
|
|
result = subprocess.run(["node", "--input-type=module", "-e", harness],
|
|||
|
|
input=json.dumps(texts), capture_output=True, text=True, check=True)
|
|||
|
|
return json.loads(result.stdout)
|
|||
|
|
|
|||
|
|
|
|||
|
|
class CharacterCodesTest(unittest.TestCase):
|
|||
|
|
|
|||
|
|
def check(self, cases):
|
|||
|
|
"""[(text, expected)]: expected is the character, or None."""
|
|||
|
|
results = codes_at_end([text for text, _ in cases])
|
|||
|
|
for (text, expected), result in zip(cases, results):
|
|||
|
|
with self.subTest(text=text):
|
|||
|
|
if expected is None:
|
|||
|
|
self.assertIsNone(result)
|
|||
|
|
else:
|
|||
|
|
self.assertEqual(result, {"length": len(text) - len("Vor "), "character": expected})
|
|||
|
|
|
|||
|
|
def test_the_tables(self):
|
|||
|
|
cases = [("Vor " + written, shown) for shown, written in DIACRITICS + EXTENDED_LATIN + PINYIN]
|
|||
|
|
self.check(cases)
|
|||
|
|
# Precomposed, one code point each, as the tables display them.
|
|||
|
|
for shown, _ in DIACRITICS + EXTENDED_LATIN + PINYIN:
|
|||
|
|
self.assertEqual(len(shown), 1)
|
|||
|
|
|
|||
|
|
def test_other_bases_and_capitals(self):
|
|||
|
|
"""The base of a diacritic is a letter of any alphabet, not a
|
|||
|
|
digit, a space or ASCII punctuation."""
|
|||
|
|
self.check([("Vor ^A'", "Á"), ("Vor ^U\"", "Ü"), ("Vor ^Ss", None), ("Vor ^Sw", "Š"),
|
|||
|
|
("Vor ^c'", "ć"), ("Vor ^Ccedilla", None), ("Vor ^e1", "ē"), ("Vor ^u3", "ǔ"),
|
|||
|
|
("Vor ^x1", None), ("Vor ^q~", "q̃"), ("Vor ^1~", None), ("Vor ^.~", None),
|
|||
|
|
("Vor ^ö-", "\u022B"), ("Vor ^α'", "ά")])
|
|||
|
|
|
|||
|
|
def test_code_points_and_the_caret(self):
|
|||
|
|
self.check([("Vor ^00E9^", "é"), ("Vor ^00e9^", "é"), ("Vor ^20AC^", "€"), ("Vor ^^", "^"),
|
|||
|
|
("Vor ^00AE^", "®"), # not ^E^ at its end
|
|||
|
|
("Vor ^130ED^", "𓃭"), ("Vor ^1F600^", "😀"), ("Vor ^10FFFD^", "\U0010FFFD"),
|
|||
|
|
# 1 to 3 digits are not a code point: a control
|
|||
|
|
# character, or a collision with the mnemonics.
|
|||
|
|
("Vor ^a^", None), ("Vor ^7^", None), ("Vor ^ae^", "æ"), ("Vor ^0E9^", None),
|
|||
|
|
# Not a character: beyond the range, a surrogate, a
|
|||
|
|
# noncharacter. Stays as typed (an error in a document).
|
|||
|
|
("Vor ^110000^", None), ("Vor ^D800^", None), ("Vor ^FFFE^", None), ("Vor ^FDD0^", None)])
|
|||
|
|
|
|||
|
|
def test_sequences(self):
|
|||
|
|
"""Code points separated by spaces between one pair of carets."""
|
|||
|
|
self.check([("Vor ^9053 53EF 9053 975E 5A38 9053^", "道可道非娸道"), ("Vor ^9053 53EF^", "道可"),
|
|||
|
|
("Vor ^0041 1F600^", "A😀"),
|
|||
|
|
("Vor ^ 9053^", None), ("Vor ^9053 ^", None), # no space inside the carets
|
|||
|
|
("Vor ^9053 D800^", None), # one bad member spoils it
|
|||
|
|
# A single character stands for itself, as in
|
|||
|
|
# Klammertext, once a code point is among the members
|
|||
|
|
# (Andy, 2026-10-13); without one, x^2 + y^2 stays.
|
|||
|
|
("Vor ^9053 a 53EF^", "道a可"), ("Vor ^a 1F600^", "a😀"), ("Vor ^9053 @^", "道@"),
|
|||
|
|
("Vor ^2 + y^", None), ("Vor ^a b^", None), ("Vor ^ab 9053^", None),
|
|||
|
|
("Vor ^a^", None)])
|
|||
|
|
|
|||
|
|
def test_an_unfinished_or_no_code(self):
|
|||
|
|
"""A keystroke at a time: nothing until the code is complete, and
|
|||
|
|
a caret with something else after it stays."""
|
|||
|
|
self.check([("Vor ^", None), ("Vor ^o", None), ("Vor ^ae", None), ("Vor ^00E", None),
|
|||
|
|
("Vor ^00E9", None), ("Vor ^ a", None), ("Vor ^o\"a", None), ("Vor 2^3", None),
|
|||
|
|
("Vor ^@'", None), ("Vor ", None), ("", None)])
|
|||
|
|
|
|||
|
|
|
|||
|
|
if __name__ == "__main__":
|
|||
|
|
unittest.main()
|