aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/python/ipython/py2/IPython/utils/tokenutil.py
blob: f52d3b76583d8b62add3953a6e5ec78500002a3b (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
"""Token-related utilities"""

# Copyright (c) IPython Development Team.
# Distributed under the terms of the Modified BSD License.

from __future__ import absolute_import, print_function

from collections import namedtuple
from io import StringIO
from keyword import iskeyword

from . import tokenize2
from .py3compat import cast_unicode_py2

Token = namedtuple('Token', ['token', 'text', 'start', 'end', 'line'])

def generate_tokens(readline):
    """wrap generate_tokens to catch EOF errors"""
    try:
        for token in tokenize2.generate_tokens(readline):
            yield token
    except tokenize2.TokenError:
        # catch EOF error
        return

def line_at_cursor(cell, cursor_pos=0):
    """Return the line in a cell at a given cursor position
    
    Used for calling line-based APIs that don't support multi-line input, yet.
    
    Parameters
    ----------
    
    cell: str
        multiline block of text
    cursor_pos: integer
        the cursor position
    
    Returns
    -------
    
    (line, offset): (text, integer)
        The line with the current cursor, and the character offset of the start of the line.
    """
    offset = 0
    lines = cell.splitlines(True)
    for line in lines:
        next_offset = offset + len(line)
        if next_offset >= cursor_pos:
            break
        offset = next_offset
    else:
        line = ""
    return (line, offset)

def token_at_cursor(cell, cursor_pos=0):
    """Get the token at a given cursor
    
    Used for introspection.
    
    Function calls are prioritized, so the token for the callable will be returned
    if the cursor is anywhere inside the call.
    
    Parameters
    ----------
    
    cell : unicode
        A block of Python code
    cursor_pos : int
        The location of the cursor in the block where the token should be found
    """
    cell = cast_unicode_py2(cell)
    names = []
    tokens = []
    call_names = []
    
    offsets = {1: 0} # lines start at 1
    for tup in generate_tokens(StringIO(cell).readline):
        
        tok = Token(*tup)
        
        # token, text, start, end, line = tup
        start_line, start_col = tok.start
        end_line, end_col = tok.end
        if end_line + 1 not in offsets:
            # keep track of offsets for each line
            lines = tok.line.splitlines(True)
            for lineno, line in zip(range(start_line + 1, end_line + 2), lines):
                if lineno not in offsets:
                    offsets[lineno] = offsets[lineno-1] + len(line)
        
        offset = offsets[start_line]
        # allow '|foo' to find 'foo' at the beginning of a line
        boundary = cursor_pos + 1 if start_col == 0 else cursor_pos
        if offset + start_col >= boundary:
            # current token starts after the cursor,
            # don't consume it
            break
        
        if tok.token == tokenize2.NAME and not iskeyword(tok.text):
            if names and tokens and tokens[-1].token == tokenize2.OP and tokens[-1].text == '.':
                names[-1] = "%s.%s" % (names[-1], tok.text)
            else:
                names.append(tok.text)
        elif tok.token == tokenize2.OP:
            if tok.text == '=' and names:
                # don't inspect the lhs of an assignment
                names.pop(-1)
            if tok.text == '(' and names:
                # if we are inside a function call, inspect the function
                call_names.append(names[-1])
            elif tok.text == ')' and call_names:
                call_names.pop(-1)
        
        tokens.append(tok)
        
        if offsets[end_line] + end_col > cursor_pos:
            # we found the cursor, stop reading
            break
        
    if call_names:
        return call_names[-1]
    elif names:
        return names[-1]
    else:
        return ''