vim_syntax.py 8.83 KB
Newer Older
1
from __future__ import with_statement
2 3
# XXX(nnorwitz): what versions of python is this file supposed to work with?
# It uses the old print statement not in py3k.
4

5 6
import keyword
import exceptions
7
import builtins
8
from string import Template
9
from sys import subversion
10

11
comment_header = '''" Auto-generated Vim syntax file for Python (%s: r%s).
12
"
13
" To use: copy or symlink to ~/.vim/syntax/python.vim'''
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

statement_header = """
if exists("b:current_syntax")
  finish
endif"""

statement_footer = '''
" Uncomment the 'minlines' statement line and comment out the 'maxlines'
" statement line; changes behaviour to look at least 2000 lines previously for
" syntax matches instead of at most 200 lines
syn sync match pythonSync grouphere NONE "):$"
syn sync maxlines=200
"syn sync minlines=2000

let b:current_syntax = "python"'''

looping = ('for', 'while')
conditionals = ('if', 'elif', 'else')
boolean_ops = ('and', 'in', 'is', 'not', 'or')
import_stmts = ('import', 'from')
object_defs = ('def', 'class')

36
exception_names = sorted(exc for exc in dir(exceptions)
37 38 39 40 41 42
                                if not exc.startswith('__'))

# Need to include functions that start with '__' (e.g., __import__), but
# nothing that comes with modules (e.g., __name__), so just exclude anything in
# the 'exceptions' module since we want to ignore exceptions *and* what any
# module would have
43
builtin_names = sorted(builtin for builtin in dir(builtins)
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
                            if builtin not in dir(exceptions))

escapes = (r'+\\[abfnrtv\'"\\]+', r'"\\\o\{1,3}"', r'"\\x\x\{2}"',
            r'"\(\\u\x\{4}\|\\U\x\{8}\)"', r'"\\$"')

todos = ("TODO", "FIXME", "XXX")

# XXX codify?
numbers = (r'"\<0x\x\+[Ll]\=\>"', r'"\<\d\+[LljJ]\=\>"',
            '"\.\d\+\([eE][+-]\=\d\+\)\=[jJ]\=\>"',
            '"\<\d\+\.\([eE][+-]\=\d\+\)\=[jJ]\=\>"',
            '"\<\d\+\.\d\+\([eE][+-]\=\d\+\)\=[jJ]\=\>"')

contained = lambda x: "%s contained" % x

def str_regexes():
    """Generator to yield various combinations of strings regexes"""
    regex_template = Template('matchgroup=Normal ' +
                                'start=+[uU]\=${raw}${sep}+ ' +
                                'end=+${sep}+ ' +
                                '${skip} ' +
                                '${contains}')
    skip_regex = Template(r'skip=+\\\\\|\\${sep}+')
    for raw in ('', '[rR]'):
        for separator in ("'", '"', '"""', "'''"):
            if len(separator) == 1:
                skip = skip_regex.substitute(sep=separator)
            else:
                skip = ''
73
            contains = 'contains=pythonEscape' if not raw else ''
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
            yield regex_template.substitute(raw=raw, sep=separator, skip=skip,
                                            contains = contains)

space_errors = (r'excludenl "\S\s\+$"ms=s+1', r'" \+\t"', r'"\t\+ "')

statements = (
                ('',
                    # XXX Might need to change pythonStatement since have
                    # specific Repeat, Conditional, Operator, etc. for 'while',
                    # etc.
                    [("Statement", "pythonStatement", "keyword",
                        (kw for kw in keyword.kwlist
                            if kw not in (looping + conditionals + boolean_ops +
                                        import_stmts + object_defs))
                      ),
                     ("Statement", "pythonStatement", "keyword",
                         (' '.join(object_defs) +
                             ' nextgroup=pythonFunction skipwhite')),
                     ("Function","pythonFunction", "match",
                         contained('"[a-zA-Z_][a-zA-Z0-9_]*"')),
                     ("Repeat", "pythonRepeat", "keyword", looping),
                     ("Conditional", "pythonConditional", "keyword",
                         conditionals),
                     ("Operator", "pythonOperator", "keyword", boolean_ops),
                     ("PreCondit", "pythonPreCondit", "keyword", import_stmts),
                     ("Comment", "pythonComment", "match",
                         '"#.*$" contains=pythonTodo'),
                     ("Todo", "pythonTodo", "keyword",
                         contained(' '.join(todos))),
                     ("String", "pythonString", "region", str_regexes()),
                     ("Special", "pythonEscape", "match",
                         (contained(esc) for esc in escapes
                             if not '$' in esc)),
                     ("Special", "pythonEscape", "match", r'"\\$"'),
                    ]
                ),
                ("python_highlight_numbers",
                    [("Number", "pythonNumber", "match", numbers)]
                ),
                ("python_highlight_builtins",
                    [("Function", "pythonBuiltin", "keyword", builtin_names)]
                ),
                ("python_highlight_exceptions",
                    [("Exception", "pythonException", "keyword",
                        exception_names)]
                ),
                ("python_highlight_space_errors",
                    [("Error", "pythonSpaceError", "match",
                        ("display " + err for err in space_errors))]
                )
             )

def syn_prefix(type_, kind):
    return 'syn %s %s    ' % (type_, kind)

def fill_stmt(iterable, fill_len):
    """Yield a string that fills at most fill_len characters with strings
    returned by 'iterable' and separated by a space"""
    # Deal with trailing char to handle ' '.join() calculation
Tim Peters's avatar
Tim Peters committed
133
    fill_len += 1
134 135 136 137 138 139 140 141 142 143 144
    overflow = None
    it = iter(iterable)
    while True:
        buffer_ = []
        total_len = 0
        if overflow:
            buffer_.append(overflow)
            total_len += len(overflow) + 1
            overflow = None
        while total_len < fill_len:
            try:
145
                new_item = next(it)
146 147 148 149 150
                buffer_.append(new_item)
                total_len += len(new_item) + 1
            except StopIteration:
                if buffer_:
                    break
151 152 153
                if overflow:
                    yield overflow
                return
154 155 156 157 158 159 160 161 162 163
        if total_len > fill_len:
            overflow = buffer_.pop()
            total_len -= len(overflow) - 1
        ret = ' '.join(buffer_)
        assert len(ret) <= fill_len
        yield ret

FILL = 80

def main(file_path):
164
    with open(file_path, 'w') as FILE:
165
        # Comment for file
166
        print>>FILE, comment_header % subversion[1:]
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
        print>>FILE, ''
        # Statements at start of file
        print>>FILE, statement_header
        print>>FILE, ''
        # Generate case for python_highlight_all
        print>>FILE, 'if exists("python_highlight_all")'
        for statement_var, statement_parts in statements:
            if statement_var:
                print>>FILE, '  let %s = 1' % statement_var
        else:
            print>>FILE, 'endif'
            print>>FILE, ''
        # Generate Python groups
        for statement_var, statement_parts in statements:
            if statement_var:
                print>>FILE, 'if exists("%s")' % statement_var
                indent = '  '
            else:
                indent = ''
            for colour_group, group, type_, arguments in statement_parts:
                if not isinstance(arguments, basestring):
                    prefix = syn_prefix(type_, group)
                    if type_ == 'keyword':
                        stmt_iter = fill_stmt(arguments,
                                            FILL - len(prefix) - len(indent))
                        try:
                            while True:
194
                                print>>FILE, indent + prefix + next(stmt_iter)
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
                        except StopIteration:
                            print>>FILE, ''
                    else:
                        for argument in arguments:
                            print>>FILE, indent + prefix + argument
                        else:
                            print>>FILE, ''

                else:
                    print>>FILE, indent + syn_prefix(type_, group) + arguments
                    print>>FILE, ''
            else:
                if statement_var:
                    print>>FILE, 'endif'
                    print>>FILE, ''
            print>>FILE, ''
        # Associating Python group with Vim colour group
        for statement_var, statement_parts in statements:
            if statement_var:
                print>>FILE, '  if exists("%s")' % statement_var
                indent = '    '
            else:
                indent = '  '
            for colour_group, group, type_, arguments in statement_parts:
                print>>FILE, (indent + "hi def link %s %s" %
                                (group, colour_group))
            else:
                if statement_var:
                    print>>FILE, '  endif'
                print>>FILE, ''
        # Statements at the end of the file
        print>>FILE, statement_footer

if __name__ == '__main__':
    main("python.vim")