122 lines
4.9 KiB
Python
122 lines
4.9 KiB
Python
import pytest
|
|
from core.tokenizer import Tokenizer, Token, TokenKind, LexerError, Keywords
|
|
|
|
|
|
def test_i_can_tokenize():
|
|
source = "+*-/{}[]() ,;:.?\n\n\r\r\r\nidentifier_0\t \t10.15 10 'string\n' \"another string\"=|&<>"
|
|
tokens = list(Tokenizer(source))
|
|
assert tokens[0] == Token(TokenKind.PLUS, "+", 0, 1, 1)
|
|
assert tokens[1] == Token(TokenKind.STAR, "*", 1, 1, 2)
|
|
assert tokens[2] == Token(TokenKind.MINUS, "-", 2, 1, 3)
|
|
assert tokens[3] == Token(TokenKind.SLASH, "/", 3, 1, 4)
|
|
assert tokens[4] == Token(TokenKind.LBRACE, "{", 4, 1, 5)
|
|
assert tokens[5] == Token(TokenKind.RBRACE, "}", 5, 1, 6)
|
|
assert tokens[6] == Token(TokenKind.LBRACKET, "[", 6, 1, 7)
|
|
assert tokens[7] == Token(TokenKind.RBRACKET, "]", 7, 1, 8)
|
|
assert tokens[8] == Token(TokenKind.LPAR, "(", 8, 1, 9)
|
|
assert tokens[9] == Token(TokenKind.RPAR, ")", 9, 1, 10)
|
|
assert tokens[10] == Token(TokenKind.WHITESPACE, " ", 10, 1, 11)
|
|
assert tokens[11] == Token(TokenKind.COMMA, ",", 14, 1, 15)
|
|
assert tokens[12] == Token(TokenKind.SEMICOLON, ";", 15, 1, 16)
|
|
assert tokens[13] == Token(TokenKind.COLON, ":", 16, 1, 17)
|
|
assert tokens[14] == Token(TokenKind.DOT, ".", 17, 1, 18)
|
|
assert tokens[15] == Token(TokenKind.QMARK, "?", 18, 1, 19)
|
|
assert tokens[16] == Token(TokenKind.NEWLINE, "\n", 19, 1, 20)
|
|
assert tokens[17] == Token(TokenKind.NEWLINE, "\n\r", 20, 2, 1)
|
|
assert tokens[18] == Token(TokenKind.NEWLINE, "\r", 22, 3, 1)
|
|
assert tokens[19] == Token(TokenKind.NEWLINE, "\r\n", 23, 4, 1)
|
|
assert tokens[20] == Token(TokenKind.IDENTIFIER, "identifier_0", 25, 5, 1)
|
|
assert tokens[21] == Token(TokenKind.WHITESPACE, "\t \t", 37, 5, 13)
|
|
assert tokens[22] == Token(TokenKind.NUMBER, "10.15", 41, 5, 17)
|
|
assert tokens[23] == Token(TokenKind.WHITESPACE, " ", 46, 5, 22)
|
|
assert tokens[24] == Token(TokenKind.NUMBER, "10", 47, 5, 23)
|
|
assert tokens[25] == Token(TokenKind.WHITESPACE, " ", 49, 5, 25)
|
|
assert tokens[26] == Token(TokenKind.STRING, "'string\n'", 50, 5, 26)
|
|
assert tokens[27] == Token(TokenKind.WHITESPACE, " ", 59, 6, 1)
|
|
assert tokens[28] == Token(TokenKind.STRING, '"another string"', 60, 6, 2)
|
|
assert tokens[29] == Token(TokenKind.EQUALS, '=', 76, 6, 18)
|
|
assert tokens[30] == Token(TokenKind.VBAR, '|', 77, 6, 19)
|
|
assert tokens[31] == Token(TokenKind.AMPER, '&', 78, 6, 20)
|
|
assert tokens[32] == Token(TokenKind.LESS, '<', 79, 6, 21)
|
|
assert tokens[33] == Token(TokenKind.GREATER, '>', 80, 6, 22)
|
|
|
|
|
|
@pytest.mark.parametrize("text, expected", [
|
|
("_ident", True),
|
|
("ident", True),
|
|
("ident123", True),
|
|
("ident_123", True),
|
|
("ident-like-this", True),
|
|
("àèùéû", True),
|
|
("011254", False),
|
|
("0abcd", False),
|
|
("-abcd", False)
|
|
])
|
|
def test_i_can_tokenize_identifiers(text, expected):
|
|
tokens = list(Tokenizer(text))
|
|
comparison = tokens[0].type == TokenKind.IDENTIFIER
|
|
assert comparison == expected
|
|
|
|
|
|
@pytest.mark.parametrize("text, error_text, index, line, column", [
|
|
("'string", "'string", 7, 1, 8),
|
|
('"string', '"string', 7, 1, 8),
|
|
('"a" + "string', '"string', 13, 1, 14),
|
|
('"a"\n\n"string', '"string', 12, 3, 8),
|
|
])
|
|
def test_i_can_detect_unfinished_strings(text, error_text, index, line, column):
|
|
with pytest.raises(LexerError) as e:
|
|
list(Tokenizer(text))
|
|
assert e.value.text == error_text
|
|
assert e.value.index == index
|
|
assert e.value.line == line
|
|
assert e.value.column == column
|
|
|
|
|
|
@pytest.mark.parametrize("text, expected_text, expected_newlines", [
|
|
("'foo'", "'foo'", 0),
|
|
('"foo"', '"foo"', 0),
|
|
("'foo\rbar'", "'foo\rbar'", 1),
|
|
("'foo\nbar'", "'foo\nbar'", 1),
|
|
("'foo\n\rbar'", "'foo\n\rbar'", 1),
|
|
("'foo\r\nbar'", "'foo\r\nbar'", 1),
|
|
("'foo\r\rbar'", "'foo\r\rbar'", 2),
|
|
("'foo\n\nbar'", "'foo\n\nbar'", 2),
|
|
("'foo\r\n\n\rbar'", "'foo\r\n\n\rbar'", 2),
|
|
("'\rfoo\rbar\r'", "'\rfoo\rbar\r'", 3),
|
|
("'\nfoo\nbar\n'", "'\nfoo\nbar\n'", 3),
|
|
("'\n\rfoo\r\n'", "'\n\rfoo\r\n'", 2),
|
|
(r"'foo\'bar'", r"'foo\'bar'", 0),
|
|
(r'"foo\"bar"', r'"foo\"bar"', 0),
|
|
('"foo"bar"', '"foo"', 0),
|
|
("'foo'bar'", "'foo'", 0),
|
|
])
|
|
def test_i_can_parse_strings(text, expected_text, expected_newlines):
|
|
lexer = Tokenizer(text)
|
|
text_found, nb_of_newlines = lexer.eat_string(0, 1, 1)
|
|
|
|
assert nb_of_newlines == expected_newlines
|
|
assert text_found == expected_text
|
|
|
|
|
|
@pytest.mark.parametrize("text", [
|
|
"1", "3.1415", "0.5", "01", "-5", "-5.10"
|
|
])
|
|
def test_i_can_parse_numbers(text):
|
|
tokens = list(Tokenizer(text))
|
|
assert tokens[0].type == TokenKind.NUMBER
|
|
assert tokens[0].value == text
|
|
|
|
|
|
@pytest.mark.parametrize("text, expected", [
|
|
("def", Keywords.DEF),
|
|
("concept", Keywords.CONCEPT),
|
|
("as", Keywords.AS),
|
|
("pre", Keywords.PRE),
|
|
("post", Keywords.POST)
|
|
])
|
|
def test_i_can_recognize_keywords(text, expected):
|
|
tokens = list(Tokenizer(text))
|
|
assert tokens[0].type == TokenKind.KEYWORD
|
|
assert tokens[0].value == expected
|