| 1 |
73
|
aaronmk
|
# String manipulation
|
| 2 |
|
|
|
| 3 |
1295
|
aaronmk
|
import codecs
|
| 4 |
861
|
aaronmk
|
import re
|
| 5 |
|
|
|
| 6 |
1232
|
aaronmk
|
import util
|
| 7 |
|
|
|
| 8 |
4028
|
aaronmk
|
##### Sentinel values
|
| 9 |
|
|
|
| 10 |
|
|
class NonInternedStr(str):
|
| 11 |
|
|
'''Each instance is unique and does not compare identical with `is`.'''
|
| 12 |
|
|
pass
|
| 13 |
|
|
|
| 14 |
|
|
none_str = NonInternedStr()
|
| 15 |
4032
|
aaronmk
|
isspace_none_str = '\r\n\t\v' # clone-safe, but must be compared with ==
|
| 16 |
4028
|
aaronmk
|
|
| 17 |
1622
|
aaronmk
|
##### Parsing
|
| 18 |
|
|
|
| 19 |
4115
|
aaronmk
|
def raw_extra_len(str_):
|
| 20 |
|
|
'''Calculates the difference between the character length and byte length of
|
| 21 |
|
|
a string. These lengths will differ for Unicode strings containing
|
| 22 |
|
|
multi-byte characters.'''
|
| 23 |
|
|
return len(to_raw_str(str_))-len(str_)
|
| 24 |
|
|
|
| 25 |
3749
|
aaronmk
|
def concat(str0, str1, max_len):
|
| 26 |
4113
|
aaronmk
|
# Use to_unicode so that substring does not split Unicode characters
|
| 27 |
|
|
str0, str1 = map(to_unicode, [str0, str1])
|
| 28 |
|
|
# Use to_raw_str() because Unicode characters can be multi-byte, and length
|
| 29 |
|
|
# limits often apply to the byte length, not the character length.
|
| 30 |
|
|
return str0[:max_len-len(to_raw_str(str1))]+str1
|
| 31 |
2932
|
aaronmk
|
|
| 32 |
1622
|
aaronmk
|
def split(sep, str_):
|
| 33 |
|
|
'''Returns [] if str_ == ""'''
|
| 34 |
|
|
if str_ == '': return []
|
| 35 |
|
|
else: return str_.split(sep)
|
| 36 |
|
|
|
| 37 |
2221
|
aaronmk
|
def remove_prefix(prefix, str_, removed_ref=None):
|
| 38 |
|
|
if removed_ref == None: removed_ref = [False]
|
| 39 |
|
|
|
| 40 |
|
|
removed_ref[0] = str_.startswith(prefix)
|
| 41 |
|
|
if removed_ref[0]: return str_[len(prefix):]
|
| 42 |
1622
|
aaronmk
|
else: return str_
|
| 43 |
|
|
|
| 44 |
1680
|
aaronmk
|
def remove_prefixes(prefixes, str_):
|
| 45 |
|
|
for prefix in prefixes: str_ = remove_prefix(prefix, str_)
|
| 46 |
|
|
return str_
|
| 47 |
|
|
|
| 48 |
|
|
def with_prefixes(prefixes, str_): return (p+str_ for p in prefixes)
|
| 49 |
|
|
|
| 50 |
2221
|
aaronmk
|
def remove_suffix(suffix, str_, removed_ref=None):
|
| 51 |
|
|
if removed_ref == None: removed_ref = [False]
|
| 52 |
|
|
|
| 53 |
|
|
removed_ref[0] = str_.endswith(suffix)
|
| 54 |
|
|
if removed_ref[0]: return str_[:-len(suffix)]
|
| 55 |
1622
|
aaronmk
|
else: return str_
|
| 56 |
|
|
|
| 57 |
1959
|
aaronmk
|
def contains_any(haystack, needles):
|
| 58 |
|
|
for needle in needles:
|
| 59 |
|
|
if haystack.find(needle) >= 0: return True
|
| 60 |
|
|
return False
|
| 61 |
|
|
|
| 62 |
1622
|
aaronmk
|
def overlaps(str0, str1): return str0.find(str1) >= 0 or str1.find(str0) >= 0
|
| 63 |
|
|
|
| 64 |
2761
|
aaronmk
|
##### Escaping
|
| 65 |
|
|
|
| 66 |
|
|
def esc_for_mogrify(query):
|
| 67 |
|
|
'''Escapes a query right before being passed to a mogrifying function.'''
|
| 68 |
|
|
return query.replace('%', '%%')
|
| 69 |
|
|
|
| 70 |
1401
|
aaronmk
|
##### Unicode
|
| 71 |
|
|
|
| 72 |
2882
|
aaronmk
|
def to_raw_str(str_):
|
| 73 |
|
|
if isinstance(str_, unicode): str_ = str_.encode('utf_8')
|
| 74 |
|
|
return str_
|
| 75 |
|
|
|
| 76 |
1295
|
aaronmk
|
unicode_reader = codecs.getreader('utf_8')
|
| 77 |
|
|
|
| 78 |
73
|
aaronmk
|
def to_unicode(str_):
|
| 79 |
|
|
if isinstance(str_, unicode): return str_
|
| 80 |
|
|
encodings = ['utf_8', 'latin_1']
|
| 81 |
|
|
for encoding in encodings:
|
| 82 |
|
|
try: return unicode(str_, encoding)
|
| 83 |
|
|
except UnicodeDecodeError, e: pass
|
| 84 |
|
|
raise AssertionError(encoding+' is not a catch-all encoding')
|
| 85 |
340
|
aaronmk
|
|
| 86 |
3748
|
aaronmk
|
def ustr(value):
|
| 87 |
1232
|
aaronmk
|
'''Like built-in str() but converts to unicode object'''
|
| 88 |
3748
|
aaronmk
|
if util.is_str(value): str_ = value # already a string
|
| 89 |
|
|
elif hasattr(value, '__str__'): str_ = value.__str__()
|
| 90 |
|
|
else: str_ = str(value)
|
| 91 |
|
|
return to_unicode(str_)
|
| 92 |
1232
|
aaronmk
|
|
| 93 |
3747
|
aaronmk
|
def urepr(value):
|
| 94 |
|
|
'''Like built-in repr() but converts to unicode object'''
|
| 95 |
|
|
if hasattr(value, '__repr__'): str_ = value.__repr__()
|
| 96 |
|
|
else: str_ = repr(value)
|
| 97 |
|
|
return to_unicode(str_)
|
| 98 |
|
|
|
| 99 |
2502
|
aaronmk
|
def repr_no_u(value):
|
| 100 |
|
|
'''Like built-in repr() but removes the "u" in `u'...'`'''
|
| 101 |
3747
|
aaronmk
|
return re.sub(r"^u(?=')", r'', urepr(value))
|
| 102 |
2502
|
aaronmk
|
|
| 103 |
1401
|
aaronmk
|
##### Line endings
|
| 104 |
|
|
|
| 105 |
1622
|
aaronmk
|
def extract_line_ending(line):
|
| 106 |
|
|
'''@return tuple (contents, ending)'''
|
| 107 |
|
|
contents = remove_suffix('\r', remove_suffix('\n', line))
|
| 108 |
|
|
return (contents, line[len(contents):])
|
| 109 |
714
|
aaronmk
|
|
| 110 |
1622
|
aaronmk
|
def remove_line_ending(line): return extract_line_ending(line)[0]
|
| 111 |
|
|
|
| 112 |
|
|
def ensure_newl(str_): return remove_line_ending(str_)+'\n'
|
| 113 |
|
|
|
| 114 |
856
|
aaronmk
|
def is_multiline(str_):
|
| 115 |
|
|
newl_idx = str_.find('\n')
|
| 116 |
|
|
return newl_idx >= 0 and newl_idx != len(str_)-1 # has newline before end
|
| 117 |
|
|
|
| 118 |
860
|
aaronmk
|
def remove_extra_newl(str_):
|
| 119 |
856
|
aaronmk
|
if is_multiline(str_): return str_
|
| 120 |
2480
|
aaronmk
|
else: return str_.rstrip('\n')
|
| 121 |
856
|
aaronmk
|
|
| 122 |
714
|
aaronmk
|
def std_newl(str_): return str_.replace('\r\n', '\n').replace('\r', '\n')
|
| 123 |
|
|
|
| 124 |
3255
|
aaronmk
|
def join_lines(lines): return ''.join((l+'\n' for l in lines))
|
| 125 |
|
|
|
| 126 |
1401
|
aaronmk
|
##### Whitespace
|
| 127 |
|
|
|
| 128 |
714
|
aaronmk
|
def cleanup(str_): return std_newl(str_.strip())
|
| 129 |
861
|
aaronmk
|
|
| 130 |
1364
|
aaronmk
|
def single_space(str_): return re.sub(r' {2,}', r' ', str_.strip())
|
| 131 |
|
|
|
| 132 |
861
|
aaronmk
|
def one_line(str_): return re.sub(r'\n *', r' ', cleanup(str_))
|
| 133 |
1761
|
aaronmk
|
|
| 134 |
|
|
##### Control characters
|
| 135 |
|
|
|
| 136 |
|
|
def is_ctrl(char):
|
| 137 |
|
|
'''Whether char is a (non-printable) control character'''
|
| 138 |
|
|
return ord(char) < 32 and not char.isspace()
|
| 139 |
|
|
|
| 140 |
|
|
def strip_ctrl(str_):
|
| 141 |
|
|
'''Strips (non-printable) control characters'''
|
| 142 |
|
|
return ''.join(filter(lambda c: not is_ctrl(c), str_))
|
| 143 |
2471
|
aaronmk
|
|
| 144 |
3172
|
aaronmk
|
##### Text
|
| 145 |
|
|
|
| 146 |
|
|
def first_word(str_): return str_.partition(' ')[0]
|
| 147 |
|
|
|
| 148 |
2471
|
aaronmk
|
##### Formatting
|
| 149 |
|
|
|
| 150 |
3466
|
aaronmk
|
def indent(str_, level=1, indent_str=' '):
|
| 151 |
|
|
indent_str *= level
|
| 152 |
|
|
return ('\n'.join((indent_str+l for l in str_.rstrip().split('\n'))))+'\n'
|
| 153 |
|
|
|
| 154 |
2477
|
aaronmk
|
def as_tt(str_): return '@'+str_+'@'
|
| 155 |
|
|
|
| 156 |
2475
|
aaronmk
|
def as_code(str_, lang=None, multiline=True):
|
| 157 |
2471
|
aaronmk
|
'''Wraps a string in Redmine tags to syntax-highlight it.'''
|
| 158 |
2480
|
aaronmk
|
str_ = '\n'+str_.rstrip('\n')+'\n'
|
| 159 |
2471
|
aaronmk
|
if lang != None: str_ = '<code class="'+lang+'">'+str_+'</code>'
|
| 160 |
2475
|
aaronmk
|
if multiline: str_ = '<pre>'+str_+'</pre>'
|
| 161 |
|
|
return str_
|
| 162 |
2477
|
aaronmk
|
|
| 163 |
2504
|
aaronmk
|
def as_inline_table(dict_, key_label='Output', value_label='Input', ustr=ustr):
|
| 164 |
2477
|
aaronmk
|
'''Wraps a dict in Redmine tags to format it as a table.'''
|
| 165 |
|
|
str_ = ''
|
| 166 |
2481
|
aaronmk
|
def row(entry): return (': '.join(entry))+'\n'
|
| 167 |
2477
|
aaronmk
|
str_ += row([key_label, value_label])
|
| 168 |
2481
|
aaronmk
|
for entry in dict_.iteritems(): str_ += row([ustr(v) for v in entry])
|
| 169 |
|
|
return '<pre>\n'+str_+'</pre>'
|
| 170 |
2482
|
aaronmk
|
|
| 171 |
2504
|
aaronmk
|
def as_table(dict_, key_label='Output', value_label='Input', ustr=ustr):
|
| 172 |
2482
|
aaronmk
|
'''Wraps a dict in Redmine tags to format it as a table.'''
|
| 173 |
|
|
str_ = ''
|
| 174 |
|
|
def row(entry): return ('|'.join(['']+entry+['']))+'\n'# '' for outer border
|
| 175 |
|
|
str_ += row([key_label, value_label])
|
| 176 |
|
|
for entry in dict_.iteritems(): str_ += row([as_tt(ustr(v)) for v in entry])
|
| 177 |
|
|
return '\n'+str_+' ' # space protects last \n so blank line ends table
|
