aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/python/Pygments/py3/pygments/lexers/tnt.py
blob: 2251373c5a1bfc5e05f496ff3f8332edc58e08a7 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
"""
    pygments.lexers.tnt
    ~~~~~~~~~~~~~~~~~~~

    Lexer for Typographic Number Theory.

    :copyright: Copyright 2006-2023 by the Pygments team, see AUTHORS.
    :license: BSD, see LICENSE for details.
"""

import re

from pygments.lexer import Lexer
from pygments.token import Text, Comment, Operator, Keyword, Name, Number, \
    Punctuation, Error

__all__ = ['TNTLexer']


class TNTLexer(Lexer):
    """
    Lexer for Typographic Number Theory, as described in the book
    Gödel, Escher, Bach, by Douglas R. Hofstadter

    .. versionadded:: 2.7
    """

    name = 'Typographic Number Theory'
    url = 'https://github.com/Kenny2github/language-tnt'
    aliases = ['tnt']
    filenames = ['*.tnt']

    cur = []

    LOGIC = set('⊃→]&∧^|∨Vv')
    OPERATORS = set('+.⋅*')
    VARIABLES = set('abcde')
    PRIMES = set("'′")
    NEGATORS = set('~!')
    QUANTIFIERS = set('AE∀∃')
    NUMBERS = set('0123456789')
    WHITESPACE = set('\t \v\n')

    RULES = re.compile('''(?xi)
        joining | separation | double-tilde | fantasy\\ rule
        | carry[- ]over(?:\\ of)?(?:\\ line)?\\ ([0-9]+) | detachment
        | contrapositive | De\\ Morgan | switcheroo
        | specification | generalization | interchange
        | existence | symmetry | transitivity
        | add\\ S | drop\\ S | induction
        | axiom\\ ([1-5]) | premise | push | pop
    ''')
    LINENOS = re.compile(r'(?:[0-9]+)(?:(?:, ?|,? and )(?:[0-9]+))*')
    COMMENT = re.compile(r'\[[^\n\]]+\]')

    def __init__(self, *args, **kwargs):
        Lexer.__init__(self, *args, **kwargs)
        self.cur = []

    def whitespace(self, start, text, required=False):
        """Tokenize whitespace."""
        end = start
        try:
            while text[end] in self.WHITESPACE:
                end += 1
        except IndexError:
            end = len(text)
        if required and end == start:
            raise AssertionError
        if end != start:
            self.cur.append((start, Text, text[start:end]))
        return end

    def variable(self, start, text):
        """Tokenize a variable."""
        if text[start] not in self.VARIABLES:
            raise AssertionError
        end = start+1
        while text[end] in self.PRIMES:
            end += 1
        self.cur.append((start, Name.Variable, text[start:end]))
        return end

    def term(self, start, text):
        """Tokenize a term."""
        if text[start] == 'S':  # S...S(...) or S...0
            end = start+1
            while text[end] == 'S':
                end += 1
            self.cur.append((start, Number.Integer, text[start:end]))
            return self.term(end, text)
        if text[start] == '0':  # the singleton 0
            self.cur.append((start, Number.Integer, text[start]))
            return start+1
        if text[start] in self.VARIABLES:  # a''...
            return self.variable(start, text)
        if text[start] == '(':  # (...+...)
            self.cur.append((start, Punctuation, text[start]))
            start = self.term(start+1, text)
            if text[start] not in self.OPERATORS:
                raise AssertionError
            self.cur.append((start, Operator, text[start]))
            start = self.term(start+1, text)
            if text[start] != ')':
                raise AssertionError
            self.cur.append((start, Punctuation, text[start]))
            return start+1
        raise AssertionError  # no matches

    def formula(self, start, text):
        """Tokenize a formula."""
        if text[start] in self.NEGATORS:  # ~<...>
            end = start+1
            while text[end] in self.NEGATORS:
                end += 1
            self.cur.append((start, Operator, text[start:end]))
            return self.formula(end, text)
        if text[start] in self.QUANTIFIERS:  # Aa:<...>
            self.cur.append((start, Keyword.Declaration, text[start]))
            start = self.variable(start+1, text)
            if text[start] != ':':
                raise AssertionError
            self.cur.append((start, Punctuation, text[start]))
            return self.formula(start+1, text)
        if text[start] == '<':  # <...&...>
            self.cur.append((start, Punctuation, text[start]))
            start = self.formula(start+1, text)
            if text[start] not in self.LOGIC:
                raise AssertionError
            self.cur.append((start, Operator, text[start]))
            start = self.formula(start+1, text)
            if text[start] != '>':
                raise AssertionError
            self.cur.append((start, Punctuation, text[start]))
            return start+1
        # ...=...
        start = self.term(start, text)
        if text[start] != '=':
            raise AssertionError
        self.cur.append((start, Operator, text[start]))
        start = self.term(start+1, text)
        return start

    def rule(self, start, text):
        """Tokenize a rule."""
        match = self.RULES.match(text, start)
        if match is None:
            raise AssertionError
        groups = sorted(match.regs[1:])  # exclude whole match
        for group in groups:
            if group[0] >= 0:  # this group matched
                self.cur.append((start, Keyword, text[start:group[0]]))
                self.cur.append((group[0], Number.Integer,
                                 text[group[0]:group[1]]))
                if group[1] != match.end():
                    self.cur.append((group[1], Keyword,
                                     text[group[1]:match.end()]))
                break
        else:
            self.cur.append((start, Keyword, text[start:match.end()]))
        return match.end()

    def lineno(self, start, text):
        """Tokenize a line referral."""
        end = start
        while text[end] not in self.NUMBERS:
            end += 1
        self.cur.append((start, Punctuation, text[start]))
        self.cur.append((start+1, Text, text[start+1:end]))
        start = end
        match = self.LINENOS.match(text, start)
        if match is None:
            raise AssertionError
        if text[match.end()] != ')':
            raise AssertionError
        self.cur.append((match.start(), Number.Integer, match.group(0)))
        self.cur.append((match.end(), Punctuation, text[match.end()]))
        return match.end() + 1

    def error_till_line_end(self, start, text):
        """Mark everything from ``start`` to the end of the line as Error."""
        end = start
        try:
            while text[end] != '\n':  # there's whitespace in rules
                end += 1
        except IndexError:
            end = len(text)
        if end != start:
            self.cur.append((start, Error, text[start:end]))
        end = self.whitespace(end, text)
        return end

    def get_tokens_unprocessed(self, text):
        """Returns a list of TNT tokens."""
        self.cur = []
        start = end = self.whitespace(0, text)
        while start <= end < len(text):
            try:
                # try line number
                while text[end] in self.NUMBERS:
                    end += 1
                if end != start:  # actual number present
                    self.cur.append((start, Number.Integer, text[start:end]))
                    # whitespace is required after a line number
                    orig = len(self.cur)
                    try:
                        start = end = self.whitespace(end, text, True)
                    except AssertionError:
                        del self.cur[orig:]
                        start = end = self.error_till_line_end(end, text)
                        continue
                # at this point it could be a comment
                match = self.COMMENT.match(text, start)
                if match is not None:
                    self.cur.append((start, Comment, text[start:match.end()]))
                    start = end = match.end()
                    # anything after the closing bracket is invalid
                    start = end = self.error_till_line_end(start, text)
                    # do not attempt to process the rest
                    continue
                del match
                if text[start] in '[]':  # fantasy push or pop
                    self.cur.append((start, Keyword, text[start]))
                    start += 1
                    end += 1
                else:
                    # one formula, possibly containing subformulae
                    orig = len(self.cur)
                    try:
                        start = end = self.formula(start, text)
                    except (AssertionError, RecursionError):  # not well-formed
                        del self.cur[orig:]
                        while text[end] not in self.WHITESPACE:
                            end += 1
                        self.cur.append((start, Error, text[start:end]))
                        start = end
                # skip whitespace after formula
                orig = len(self.cur)
                try:
                    start = end = self.whitespace(end, text, True)
                except AssertionError:
                    del self.cur[orig:]
                    start = end = self.error_till_line_end(start, text)
                    continue
                # rule proving this formula a theorem
                orig = len(self.cur)
                try:
                    start = end = self.rule(start, text)
                except AssertionError:
                    del self.cur[orig:]
                    start = end = self.error_till_line_end(start, text)
                    continue
                # skip whitespace after rule
                start = end = self.whitespace(end, text)
                # line marker
                if text[start] == '(':
                    orig = len(self.cur)
                    try:
                        start = end = self.lineno(start, text)
                    except AssertionError:
                        del self.cur[orig:]
                        start = end = self.error_till_line_end(start, text)
                        continue
                    start = end = self.whitespace(start, text)
            except IndexError:
                try:
                    del self.cur[orig:]
                except NameError:
                    pass  # if orig was never defined, fine
                self.error_till_line_end(start, text)
        return self.cur