# String manipulation

import codecs

import re

import util

##### Parsing

def split(sep, str_):

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

    if str_ == '': return []

    else: return str_.split(sep)

def remove_prefix(prefix, str_, removed_ref=None):

    if removed_ref == None: removed_ref = [False]

    removed_ref[0] = str_.startswith(prefix)

    if removed_ref[0]: return str_[len(prefix):]

    else: return str_

def remove_prefixes(prefixes, str_):

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

    return str_

def with_prefixes(prefixes, str_): return (p+str_ for p in prefixes)

def remove_suffix(suffix, str_, removed_ref=None):

    if removed_ref == None: removed_ref = [False]

    removed_ref[0] = str_.endswith(suffix)

    if removed_ref[0]: return str_[:-len(suffix)]

    else: return str_

def contains_any(haystack, needles):

    for needle in needles:

        if haystack.find(needle) >= 0: return True

    return False

def overlaps(str0, str1): return str0.find(str1) >= 0 or str1.find(str0) >= 0

##### 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 extract_line_ending(line):

    '''@return tuple (contents, ending)'''

    contents = remove_suffix('\r', remove_suffix('\n', line))

    return (contents, line[len(contents):])

def remove_line_ending(line): return extract_line_ending(line)[0]

def ensure_newl(str_): return remove_line_ending(str_)+'\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('\n')

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_))

##### Control characters

def is_ctrl(char):

    '''Whether char is a (non-printable) control character'''

    return ord(char) < 32 and not char.isspace()

def strip_ctrl(str_):

    '''Strips (non-printable) control characters'''

    return ''.join(filter(lambda c: not is_ctrl(c), str_))

##### Formatting

def as_tt(str_): return '@'+str_+'@'

def as_code(str_, lang=None, multiline=True):

    '''Wraps a string in Redmine tags to syntax-highlight it.'''

    str_ = '\n'+str_.rstrip('\n')+'\n'

    if lang != None: str_ = '<code class="'+lang+'">'+str_+'</code>'

    if multiline: str_ = '<pre>'+str_+'</pre>'

    return str_

def as_inline_table(dict_, key_label='Output', value_label='Input'):

    '''Wraps a dict in Redmine tags to format it as a table.'''

    str_ = ''

    def row(entry): return (': '.join(entry))+'\n'

    str_ += row([key_label, value_label])

    for entry in dict_.iteritems(): str_ += row([ustr(v) for v in entry])

    return '<pre>\n'+str_+'</pre>'

def as_table(dict_, key_label='Output', value_label='Input'):

    '''Wraps a dict in Redmine tags to format it as a table.'''

    str_ = ''

    def row(entry): return ('|'.join(['']+entry+['']))+'\n'# '' for outer border

    str_ += row([key_label, value_label])

    for entry in dict_.iteritems(): str_ += row([as_tt(ustr(v)) for v in entry])

    return '\n'+str_+' ' # space protects last \n so blank line ends table