| | 1 | import re |
| | 2 | from codecs import BOM_UTF8 |
| | 3 | from typing import Tuple |
| | 4 | |
| | 5 | from parso.python.tokenize import group |
| | 6 | |
| | 7 | unicode_bom = BOM_UTF8.decode('utf-8') |
| | 8 | |
| | 9 | |
| | 10 | class PrefixPart: |
| | 11 | def __init__(self, leaf, typ, value, spacing='', start_pos=None): |
| | 12 | assert start_pos is not None |
| | 13 | self.parent = leaf |
| | 14 | self.type = typ |
| | 15 | self.value = value |
| | 16 | self.spacing = spacing |
| | 17 | self.start_pos: Tuple[int, int] = start_pos |
| | 18 | |
| | 19 | @property |
| | 20 | def end_pos(self) -> Tuple[int, int]: |
| | 21 | if self.value.endswith('\n') or self.value.endswith('\r'): |
| | 22 | return self.start_pos[0] + 1, 0 |
| | 23 | if self.value == unicode_bom: |
| | 24 | # The bom doesn't have a length at the start of a Python file. |
| | 25 | return self.start_pos |
| | 26 | return self.start_pos[0], self.start_pos[1] + len(self.value) |
| | 27 | |
| | 28 | def create_spacing_part(self): |
| | 29 | column = self.start_pos[1] - len(self.spacing) |
| | 30 | return PrefixPart( |
| | 31 | self.parent, 'spacing', self.spacing, |
| | 32 | start_pos=(self.start_pos[0], column) |
| | 33 | ) |
| | 34 | |
| | 35 | def __repr__(self): |
| | 36 | return '%s(%s, %s, %s)' % ( |
| | 37 | self.__class__.__name__, |
| | 38 | self.type, |
| | 39 | repr(self.value), |
| | 40 | self.start_pos |
| | 41 | ) |
| | 42 | |
| | 43 | def search_ancestor(self, *node_types): |
| | 44 | node = self.parent |
| | 45 | while node is not None: |
| | 46 | if node.type in node_types: |
| | 47 | return node |
| | 48 | node = node.parent |
| | 49 | return None |
| | 50 | |
| | 51 | |
| | 52 | _comment = r'#[^\n\r\f]*' |
| | 53 | _backslash = r'\\\r?\n|\\\r' |
| | 54 | _newline = r'\r?\n|\r' |
| | 55 | _form_feed = r'\f' |
| | 56 | _only_spacing = '$' |
| | 57 | _spacing = r'[ \t]*' |
| | 58 | _bom = unicode_bom |
| | 59 | |
| | 60 | _regex = group( |
| | 61 | _comment, _backslash, _newline, _form_feed, _only_spacing, _bom, |
| | 62 | capture=True |
| | 63 | ) |
| | 64 | _regex = re.compile(group(_spacing, capture=True) + _regex) |
| | 65 | |
| | 66 | |
| | 67 | _types = { |
| | 68 | '#': 'comment', |
| | 69 | '\\': 'backslash', |
| | 70 | '\f': 'formfeed', |
| | 71 | '\n': 'newline', |
| | 72 | '\r': 'newline', |
| | 73 | unicode_bom: 'bom' |
| | 74 | } |
| | 75 | |
| | 76 | |
| | 77 | def split_prefix(leaf, start_pos): |
| | 78 | line, column = start_pos |
| | 79 | start = 0 |
| | 80 | value = spacing = '' |
| | 81 | bom = False |
| | 82 | while start != len(leaf.prefix): |
| | 83 | match = _regex.match(leaf.prefix, start) |
| | 84 | spacing = match.group(1) |
| | 85 | value = match.group(2) |
| | 86 | if not value: |
| | 87 | break |
| | 88 | type_ = _types[value[0]] |
| | 89 | yield PrefixPart( |
| | 90 | leaf, type_, value, spacing, |
| | 91 | start_pos=(line, column + start - int(bom) + len(spacing)) |
| | 92 | ) |
| | 93 | if type_ == 'bom': |
| | 94 | bom = True |
| | 95 | |
| | 96 | start = match.end(0) |
| | 97 | if value.endswith('\n') or value.endswith('\r'): |
| | 98 | line += 1 |
| | 99 | column = -start |
| | 100 | |
| | 101 | if value: |
| | 102 | spacing = '' |
| | 103 | yield PrefixPart( |
| | 104 | leaf, 'spacing', spacing, |
| | 105 | start_pos=(line, column + start) |
| | 106 | ) |
projects / config.git / blame_incremental