# String manipulation

import codecs

import re

import util

##### Unicode

unicode_reader = codecs.getreader('utf_8')

def to_unicode(str_):

    if isinstance(str_, unicode): return str_

    encodings = ['utf_8', 'latin_1']

    for encoding in encodings:

        try: return unicode(str_, encoding)

        except UnicodeDecodeError, e: pass

    raise AssertionError(encoding+' is not a catch-all encoding')

def ustr(val):

    '''Like built-in str() but converts to unicode object'''

    if not util.is_str(val): val = str(val)

    return to_unicode(val)

##### Line endings

def ensure_newl(str_): return str_.rstrip()+'\n'

def is_multiline(str_):

    newl_idx = str_.find('\n')

    return newl_idx >= 0 and newl_idx != len(str_)-1 # has newline before end

def remove_extra_newl(str_):

    if is_multiline(str_): return str_

    else: return str_.rstrip()

def std_newl(str_): return str_.replace('\r\n', '\n').replace('\r', '\n')

##### Whitespace

def cleanup(str_): return std_newl(str_.strip())

def single_space(str_): return re.sub(r' {2,}', r' ', str_.strip())

def one_line(str_): return re.sub(r'\n *', r' ', cleanup(str_))

##### Parsing

def split(sep, str_):

    '''Returns [] if str_ == ""'''

    if str_ == '': return []

    else: return str_.split(sep)

def remove_prefix(prefix, str_):

    if str_.startswith(prefix): return str_[len(prefix):]

    else: return str_

def remove_suffix(suffix, str_):

    if str_.endswith(suffix): return str_[:-len(suffix)]

    else: return str_

def remove_prefixes(prefixes, str_):

    for prefix in prefixes: str_ = remove_prefix(prefix, str_)

    return str_