Sadmin
f20080305d
Определение специфических рекомендаций (матрица профилей §8):
1. Ж/д слой (закрыт мёртвый railway ×2.5 у РАС):
- /api/v1/water/{case_id} отдаёт railway=rail как LineString
(без service/industrial/military веток), кэш общий v2;
- railway_warning «перекрыть/проверить немедленно» по профилям;
- SearchMap: Polyline слой ж/д (тёмно-красный), счётчики 💧/🚂.
2. cant_swim → профиль не_умеет_плавать (water ×3.0, без изменения
радиуса, critical_warning «обследовать водоёмы НЕМЕДЛЕННО»):
- раньше чекбокс влиял только на текст, в скоринге был пробел;
- derive в analyze._derive_profiles — работает и для closed_cases.
3. unmodeled_profiles: ДЦП/слабое зрение/слух — честная пометка
«вне поведенческой модели» с пояснением (vector_tasks B12:
профили без аналога не выдавать за учтённые); блок на фронте
в карточке здоровья.
Площадь воды: сферический эксцесс, проверен на квадрате 53° (744017 м²
vs 743272 точного). Тесты: 202 passed (новый test_cant_swim_profile).
103 lines
3.2 KiB
Python
103 lines
3.2 KiB
Python
"""
|
|
pygments.regexopt
|
|
~~~~~~~~~~~~~~~~~
|
|
|
|
An algorithm that generates optimized regexes for matching long lists of
|
|
literal strings.
|
|
|
|
:copyright: Copyright 2006-present by the Pygments team, see AUTHORS.
|
|
:license: BSD, see LICENSE for details.
|
|
"""
|
|
|
|
import re
|
|
from re import escape
|
|
from itertools import groupby
|
|
from operator import itemgetter
|
|
|
|
CS_ESCAPE = re.compile(r'[\[\^\\\-\]]')
|
|
FIRST_ELEMENT = itemgetter(0)
|
|
|
|
|
|
def commonprefix(m):
|
|
"""Given an iterable of strings, returns the longest common leading substring"""
|
|
if not m:
|
|
return ""
|
|
s1 = min(m)
|
|
s2 = max(m)
|
|
for i, c in enumerate(s1):
|
|
if c != s2[i]:
|
|
return s1[:i]
|
|
return s1
|
|
|
|
|
|
def make_charset(letters):
|
|
return '[' + CS_ESCAPE.sub(lambda m: '\\' + m.group(), ''.join(letters)) + ']'
|
|
|
|
|
|
def regex_opt_inner(strings, open_paren):
|
|
"""Return a regex that matches any string in the sorted list of strings."""
|
|
close_paren = open_paren and ')' or ''
|
|
# print strings, repr(open_paren)
|
|
if not strings:
|
|
# print '-> nothing left'
|
|
return ''
|
|
first = strings[0]
|
|
if len(strings) == 1:
|
|
# print '-> only 1 string'
|
|
return open_paren + escape(first) + close_paren
|
|
if not first:
|
|
# print '-> first string empty'
|
|
return open_paren + regex_opt_inner(strings[1:], '(?:') \
|
|
+ '?' + close_paren
|
|
if len(first) == 1:
|
|
# multiple one-char strings? make a charset
|
|
oneletter = []
|
|
rest = []
|
|
for s in strings:
|
|
if len(s) == 1:
|
|
oneletter.append(s)
|
|
else:
|
|
rest.append(s)
|
|
if len(oneletter) > 1: # do we have more than one oneletter string?
|
|
if rest:
|
|
# print '-> 1-character + rest'
|
|
return open_paren + regex_opt_inner(rest, '') + '|' \
|
|
+ make_charset(oneletter) + close_paren
|
|
# print '-> only 1-character'
|
|
return open_paren + make_charset(oneletter) + close_paren
|
|
prefix = commonprefix(strings)
|
|
if prefix:
|
|
plen = len(prefix)
|
|
# we have a prefix for all strings
|
|
# print '-> prefix:', prefix
|
|
return open_paren + escape(prefix) \
|
|
+ regex_opt_inner([s[plen:] for s in strings], '(?:') \
|
|
+ close_paren
|
|
# is there a suffix?
|
|
strings_rev = [s[::-1] for s in strings]
|
|
suffix = commonprefix(strings_rev)
|
|
if suffix:
|
|
slen = len(suffix)
|
|
# print '-> suffix:', suffix[::-1]
|
|
return open_paren \
|
|
+ regex_opt_inner(sorted(s[:-slen] for s in strings), '(?:') \
|
|
+ escape(suffix[::-1]) + close_paren
|
|
# recurse on common 1-string prefixes
|
|
# print '-> last resort'
|
|
return open_paren + \
|
|
'|'.join(regex_opt_inner(list(group[1]), '')
|
|
for group in groupby(strings, lambda s: s[0] == first[0])) \
|
|
+ close_paren
|
|
|
|
|
|
def regex_opt(strings, prefix='', suffix=''):
|
|
"""Return a compiled regex that matches any string in the given list.
|
|
|
|
The strings to match must be literal strings, not regexes. They will be
|
|
regex-escaped.
|
|
|
|
*prefix* and *suffix* are pre- and appended to the final regex.
|
|
"""
|
|
strings = sorted(set(strings))
|
|
return prefix + regex_opt_inner(strings, '(') + suffix
|