| 1 |
73
|
aaronmk
|
# String manipulation
|
| 2 |
|
|
|
| 3 |
1295
|
aaronmk
|
import codecs
|
| 4 |
861
|
aaronmk
|
import re
|
| 5 |
|
|
|
| 6 |
1232
|
aaronmk
|
import util
|
| 7 |
|
|
|
| 8 |
1622
|
aaronmk
|
##### Parsing
|
| 9 |
|
|
|
| 10 |
|
|
def split(sep, str_):
|
| 11 |
|
|
'''Returns [] if str_ == ""'''
|
| 12 |
|
|
if str_ == '': return []
|
| 13 |
|
|
else: return str_.split(sep)
|
| 14 |
|
|
|
| 15 |
|
|
def remove_prefix(prefix, str_):
|
| 16 |
|
|
if str_.startswith(prefix): return str_[len(prefix):]
|
| 17 |
|
|
else: return str_
|
| 18 |
|
|
|
| 19 |
1680
|
aaronmk
|
def remove_prefixes(prefixes, str_):
|
| 20 |
|
|
for prefix in prefixes: str_ = remove_prefix(prefix, str_)
|
| 21 |
|
|
return str_
|
| 22 |
|
|
|
| 23 |
|
|
def with_prefixes(prefixes, str_): return (p+str_ for p in prefixes)
|
| 24 |
|
|
|
| 25 |
1622
|
aaronmk
|
def remove_suffix(suffix, str_):
|
| 26 |
|
|
if str_.endswith(suffix): return str_[:-len(suffix)]
|
| 27 |
|
|
else: return str_
|
| 28 |
|
|
|
| 29 |
|
|
def overlaps(str0, str1): return str0.find(str1) >= 0 or str1.find(str0) >= 0
|
| 30 |
|
|
|
| 31 |
1401
|
aaronmk
|
##### Unicode
|
| 32 |
|
|
|
| 33 |
1295
|
aaronmk
|
unicode_reader = codecs.getreader('utf_8')
|
| 34 |
|
|
|
| 35 |
73
|
aaronmk
|
def to_unicode(str_):
|
| 36 |
|
|
if isinstance(str_, unicode): return str_
|
| 37 |
|
|
encodings = ['utf_8', 'latin_1']
|
| 38 |
|
|
for encoding in encodings:
|
| 39 |
|
|
try: return unicode(str_, encoding)
|
| 40 |
|
|
except UnicodeDecodeError, e: pass
|
| 41 |
|
|
raise AssertionError(encoding+' is not a catch-all encoding')
|
| 42 |
340
|
aaronmk
|
|
| 43 |
1232
|
aaronmk
|
def ustr(val):
|
| 44 |
|
|
'''Like built-in str() but converts to unicode object'''
|
| 45 |
|
|
if not util.is_str(val): val = str(val)
|
| 46 |
|
|
return to_unicode(val)
|
| 47 |
|
|
|
| 48 |
1401
|
aaronmk
|
##### Line endings
|
| 49 |
|
|
|
| 50 |
1622
|
aaronmk
|
def extract_line_ending(line):
|
| 51 |
|
|
'''@return tuple (contents, ending)'''
|
| 52 |
|
|
contents = remove_suffix('\r', remove_suffix('\n', line))
|
| 53 |
|
|
return (contents, line[len(contents):])
|
| 54 |
714
|
aaronmk
|
|
| 55 |
1622
|
aaronmk
|
def remove_line_ending(line): return extract_line_ending(line)[0]
|
| 56 |
|
|
|
| 57 |
|
|
def ensure_newl(str_): return remove_line_ending(str_)+'\n'
|
| 58 |
|
|
|
| 59 |
856
|
aaronmk
|
def is_multiline(str_):
|
| 60 |
|
|
newl_idx = str_.find('\n')
|
| 61 |
|
|
return newl_idx >= 0 and newl_idx != len(str_)-1 # has newline before end
|
| 62 |
|
|
|
| 63 |
860
|
aaronmk
|
def remove_extra_newl(str_):
|
| 64 |
856
|
aaronmk
|
if is_multiline(str_): return str_
|
| 65 |
|
|
else: return str_.rstrip()
|
| 66 |
|
|
|
| 67 |
714
|
aaronmk
|
def std_newl(str_): return str_.replace('\r\n', '\n').replace('\r', '\n')
|
| 68 |
|
|
|
| 69 |
1401
|
aaronmk
|
##### Whitespace
|
| 70 |
|
|
|
| 71 |
714
|
aaronmk
|
def cleanup(str_): return std_newl(str_.strip())
|
| 72 |
861
|
aaronmk
|
|
| 73 |
1364
|
aaronmk
|
def single_space(str_): return re.sub(r' {2,}', r' ', str_.strip())
|
| 74 |
|
|
|
| 75 |
861
|
aaronmk
|
def one_line(str_): return re.sub(r'\n *', r' ', cleanup(str_))
|
| 76 |
1761
|
aaronmk
|
|
| 77 |
|
|
##### Control characters
|
| 78 |
|
|
|
| 79 |
|
|
def is_ctrl(char):
|
| 80 |
|
|
'''Whether char is a (non-printable) control character'''
|
| 81 |
|
|
return ord(char) < 32 and not char.isspace()
|
| 82 |
|
|
|
| 83 |
|
|
def strip_ctrl(str_):
|
| 84 |
|
|
'''Strips (non-printable) control characters'''
|
| 85 |
|
|
return ''.join(filter(lambda c: not is_ctrl(c), str_))
|
