summaryrefslogtreecommitdiff
path: root/pygments/lexers/postgres.py
blob: b4dc9df048c6e8306d641171baba7ffa32ac7d9e (plain)
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
"""
    pygments.lexers.postgres
    ~~~~~~~~~~~~~~~~~~~~~~~~

    Lexers for PostgreSQL-specific SQL and psql interactive session.

    :copyright: Copyright 2011 by Daniele Varrazzo.
    :license: BSD, see LICENSE for details.
"""

import re
import sys
from copy import deepcopy

from pygments.lexer import Lexer, RegexLexer, include, bygroups, using, \
     this, do_insertions
from pygments.token import Error, Punctuation, Literal, Token, \
     Text, Comment, Operator, Keyword, Name, String, Number, Generic
from pygments.lexers import get_lexer_by_name, ClassNotFound

from pygments.lexers._postgres_builtins import (
    KEYWORDS, DATATYPES, PSEUDO_TYPES)


__all__ = [ 'PostgresLexer', 'PostgresConsoleLexer' ]

line_re  = re.compile('.*?\n')

language_re = re.compile(r"\s+LANGUAGE\s+'?(\w+)'?")

class PostgresLexer(RegexLexer):
    """
    Lexer for the PostgreSQL dialect of SQL.
    """

    name = 'PostgreSQL SQL dialect'
    aliases = ['postgresql', 'postgres']
    mimetypes = ['text/x-postgresql']

    def get_tokens_unprocessed(self, text, *args):
        # Have a copy of the entire text to be used by `language_callback`.
        self.text = text
        for x in RegexLexer.get_tokens_unprocessed(self, text, *args):
            yield x

    def language_callback(self, match):
        lexer = None
        # TODO: the language can also be before the string
        m = language_re.match(self.text[match.end():])
        if m is not None:
            lexer = self._get_lexer(m.group(1))

        if lexer:
            yield (match.start(1), String, match.group(1))
            for x in lexer.get_tokens_unprocessed(match.group(2)):
                yield x
            yield (match.start(3), String, match.group(3))

        else:
            yield (match.start(), String, match.group())

    def _get_lexer(self, lang):
        if lang == 'sql':
            return get_lexer_by_name('postgresql', **self.options)

        tries = [ lang ]
        if lang.startswith('pl'):
            tries.append(lang[2:])
        if lang.endswith('u'):
            tries.append(lang[:-1])
        if lang.startswith('pl') and lang.endswith('u'):
            tries.append(lang[2:-1])

        for l in tries:
            try:
                return get_lexer_by_name(l, **self.options)
            except ClassNotFound:
                pass
        else:
            # TODO: better logging
            print >>sys.stderr, "language not found:", lang
            return None

    flags = re.IGNORECASE
    tokens = {
        'root': [
            (r'\s+', Text),
            (r'--.*?\n', Comment.Single),
            (r'/\*', Comment.Multiline, 'multiline-comments'),
            (r'(' + '|'.join([s.replace(" ", "\s+")
                for s in DATATYPES + PSEUDO_TYPES])
                  + r')\b', Name.Builtin),
            (r'(' + '|'.join(KEYWORDS) + r')\b', Keyword),
            (r'[+*/<>=~!@#%^&|`?^-]+', Operator),
            (r'::', Operator),  # cast
            (r'\$\d+', Name.Variable),
            (r'([0-9]*\.[0-9]*|[0-9]+)(e[+-]?[0-9]+)?', Number.Float),
            (r'[0-9]+', Number.Integer),
            # TODO: Backslash escapes?
            (r"E?'(''|[^'])*'", String.Single),
            (r'"(""|[^"])*"', String.Name), # quoted identifier
            (r'(?ms)(\$[^\$]*\$)(.*?)(\1)', language_callback),
            (r'[a-zA-Z_][a-zA-Z0-9_]*', Name),

            # psql variable in SQL
            (r""":(['"]?)[a-z][a-z0-9_]*\b\1""", Name.Variable),

            (r'[;:()\[\],\.]', Punctuation),
        ],
        'multiline-comments': [
            (r'/\*', Comment.Multiline, 'multiline-comments'),
            (r'\*/', Comment.Multiline, '#pop'),
            (r'[^/\*]+', Comment.Multiline),
            (r'[/*]', Comment.Multiline)
        ],
    }


class PsqlRegexLexer(PostgresLexer):
    """
    Extend the PostgresLexer adding support specific for psql commands.

    This is not a complete psql lexer yet as it lacks prompt support
    and output rendering.
    """
    name = 'PostgreSQL console - regexp based lexer'
    aliases = []    # not public
    tokens = deepcopy(PostgresLexer.tokens)
    tokens['root'].append(
        (r'\\[^\s]+', Keyword.Pseudo, 'psql-command'))
    tokens['psql-command'] = [
        (r'\n', Text, 'root'),
        (r'\s+', Text),
        (r'\\[^\s]+', Keyword.Pseudo),
        (r""":(['"]?)[a-z][a-z0-9_]*\b\1""", Name.Variable),
        (r"'(''|[^'])*'", String.Single),
        (r"`([^`])*`", String.Backtick),
        (r"[^\s]+", String.Symbol),
    ]


re_prompt = re.compile(r'^.*?[=\-\(][#>]')
re_psql_command = re.compile(r'\s*\\')
re_end_command = re.compile(r';\s*(--.*?)?$')
re_psql_command = re.compile(r'(\s*)(\\.+?)(\s+)$')
re_error = re.compile(r'(ERROR|FATAL):')
re_message = re.compile(
    r'((?:DEBUG|INFO|NOTICE|WARNING|ERROR|'
    r'FATAL|HINT|DETAIL|LINE [0-9]+):)(.*?\n)')
re_charhint = re.compile(r'\s*\^\s*\n')

def lookahead(x):
    """Wrap an iterator and allow pushing back an item."""
    for i in x:
        while 1:
            i = yield i
            if i is None:
                break
            yield i


class PostgresConsoleLexer(Lexer):
    """
    Lexer for psql sessions.
    """

    name = 'PostgreSQL console (psql)'
    aliases = ['psql', 'postgresql-console', 'postgres-console']
    mimetypes = ['text/x-postgresql-psql']

    def get_tokens_unprocessed(self, data):
        sql = PsqlRegexLexer(**self.options)

        lines = lookahead(line_re.findall(data))

        # prompt-output cycle
        while 1:

            # consume the lines of the command: start with an optional prompt
            # and continue until the end of command is detected
            curcode = ''
            insertions = []
            while 1:
                try:
                    line = lines.next()
                except StopIteration:
                    # allow the emission of partially collected items
                    # the repl loop will be broken below
                    break

                # Identify a shell prompt in case of psql commandline example
                if line.startswith('$'):
                    lexer = get_lexer_by_name('console', **self.options)
                    for x in lexer.get_tokens_unprocessed(line):
                        yield x
                    break

                # Identify a psql prompt
                mprompt = re_prompt.match(line)
                if mprompt is not None:
                    insertions.append((len(curcode),
                                       [(0, Generic.Prompt, mprompt.group())]))
                    curcode += line[len(mprompt.group()):]
                else:
                    curcode += line

                # Check if this is the end of the command
                # TODO: better handle multiline comments at the end with
                # a lexer with an external state?
                if re_psql_command.match(curcode) \
                or re_end_command.search(curcode):
                    break

            # Emit the combined stream of command and prompt(s)
            for item in do_insertions(insertions,
                    sql.get_tokens_unprocessed(curcode)):
                yield item

            # Emit the output lines
            out_token = Generic.Output
            while 1:
                line = lines.next()
                mprompt = re_prompt.match(line)
                if mprompt is not None:
                    # push the line back to have it processed by the prompt
                    lines.send(line)
                    break

                mmsg = re_message.match(line)
                if mmsg is not None:
                    if mmsg.group(1).startswith("ERROR") \
                    or mmsg.group(1).startswith("FATAL"):
                        out_token = Generic.Error
                    yield (mmsg.start(1), Generic.Strong, mmsg.group(1))
                    yield (mmsg.start(2), out_token, mmsg.group(2))
                elif re_charhint.match(line):
                    yield (0, out_token, line)
                else:
                    yield (0, Generic.Output, line)