File indexing completed on 2024-05-12 05:47:03

 # -*- coding: UTF-8 -*-
 
 """
 Process letter accents in Serbian Cyrillic text.
 
 @author: Chusslove Illich (Часлав Илић) <caslav.ilic@gmx.net>
 @license: GPLv3
 """
 
 # All accented letters in Serbian Cyrillic, for a given non-accented letter.
 _accents = {
     "а": ("а̀", "а́", "а̏", "а̑", "а̄", "а̂", "â", "ȃ"),
     "А": ("А̀", "А́", "А̏", "А̑", "А̄", "А̂", "Â", "Ȃ"),
     # ...with Latin long-falling/genitive a in NFC, used sometimes as makeshift
     "е": ("ѐ", "е́", "е̏", "е̑", "е̄", "е̂", "ѐ"),
     "Е": ("Ѐ", "Е́", "Е̏", "Е̑", "Е̄", "Е̂", "Ѐ"),
     "и": ("ѝ", "и́", "и̏", "и̑", "ӣ", "и̂", "ѝ", "ӣ"),
     "И": ("Ѝ", "И́", "И̏", "И̑", "Ӣ", "И̂", "Ѝ", "Ӣ"),
     "о": ("о̀", "о́", "о̏", "о̑", "о̄", "о̂", "ȏ", "ô"),
     "О": ("О̀", "О́", "О̏", "О̑", "О̄", "О̂", "Ȏ", "Ô"),
     # ...with Latin long-falling/genitive o in NFC, used sometimes as makeshift
     "у": ("у̀", "у́", "у̏", "у̑", "ӯ", "у̂", "ӯ"),
     "У": ("У̀", "У́", "У̏", "У̑", "Ӯ", "У̂", "Ӯ"),
     "р": ("р̀", "р́", "р̏", "р̑", "р̄", "р̂"),
     "Р": ("Р̀", "Р́", "Р̏", "Р̑", "Р̄", "Р̂"),
 }
 
 # All accented letters bunched together,
 # and inverted mapping (base for each accented letter).
 _accents_flat = set()
 _accents_inverted = {}
 for base, accents in list(_accents.items()):
     _accents_flat.update(set(accents))
     for accent in accents:
         _accents_inverted[accent] = base
 del base, accents # do not pollute exports
 
 _max_accent_len = max(list(map(len, list(_accents_flat))))
 _min_accent_len = min(list(map(len, list(_accents_flat))))
 _accent_len_range = list(range(_max_accent_len, _min_accent_len - 1, -1))
 
 # FIXME: The graphing sequences with slashes and backslashes are far
 # too easy to happen accidentally; think of something better.
 _agraphs_unused = {
     r"\а": r"а̀",
     r"/а": r"а́",
     r"\\а": r"а̏",
     r"//а": r"а̑",
     r"~а": r"а̄",
     r"\А": r"А̀",
     r"/А": r"А́",
     r"\\А": r"А̏",
     r"//А": r"А̑",
     r"~А": r"А̄",
 
     r"\е": r"ѐ",
     r"/е": r"е́",
     r"\\е": r"е̏",
     r"//е": r"е̑",
     r"~е": r"е̄",
     r"\Е": r"Ѐ",
     r"/Е": r"Е́",
     r"\\Е": r"Е̏",
     r"//Е": r"Е̑",
     r"~Е": r"Е̄",
 
     r"\и": r"ѝ",
     r"/и": r"и́",
     r"\\и": r"и̏",
     r"//и": r"и̑",
     r"~и": r"ӣ",
     r"\И": r"Ѝ",
     r"/И": r"И́",
     r"\\И": r"И̏",
     r"//И": r"И̑",
     r"~И": r"Ӣ",
 
     r"\о": r"о̀",
     r"/о": r"о́",
     r"\\о": r"о̏",
     r"//о": r"о̑",
     r"~о": r"о̄",
     r"\О": r"О̀",
     r"/О": r"О́",
     r"\\О": r"О̏",
     r"//О": r"О̑",
     r"~О": r"О̄",
 
     r"\у": r"у̀",
     r"/у": r"у́",
     r"\\у": r"у̏",
     r"//у": r"у̑",
     r"~у": r"ӯ",
     r"\У": r"У̀",
     r"/У": r"У́",
     r"\\У": r"У̏",
     r"//У": r"У̑",
     r"~У": r"Ӯ",
 
     r"\р": r"р̀",
     r"/р": r"р́",
     r"\\р": r"р̏",
     r"//р": r"р̑",
     r"~р": r"р̄",
     r"\Р": r"Р̀",
     r"/Р": r"Р́",
     r"\\Р": r"Р̏",
     r"//Р": r"Р̑",
     r"~Р": r"Р̄",
 }
 
 _agraphs = {
     #ur"^а": ur"а̂",
     #ur"^о": ur"о̂",
     #ur"^А": ur"А̂",
     #ur"^О": ur"О̂",
     # ...use Latin NFC forms at places for the moment.
     r"^а" : r"â",
     r"^о" : r"ô",
     r"^А" : r"Â",
     r"^О" : r"Ô",
 }
 
 _max_agraph_len = max(list(map(len, list(_agraphs.keys()))))
 _min_agraph_len = min(list(map(len, list(_agraphs.keys()))))
 _agraph_len_range = list(range(_max_agraph_len, _min_agraph_len - 1, -1))
 
 
 def resolve_agraphs (text):
     """
     Convert accent graphs into real accented letters [type F1A hook].
 
     Accented Cyrillic letters still cannot be widely entered directly
     by keyboard, and in such cases this module allows converting graphical
     accent-letter representations into actual Unicode compositions.
 
     @note: At the moment, only genitive endings are supported.
 
     @return: text
     """
 
     return _apply_mapping(text, _agraphs, _agraph_len_range)
 
 
 def remove_accents (text):
     """
     Remove accents from all accented letters [type F1A hook].
 
     Sometimes it is convenient to operate on text without accents,
     e.g. when checking spelling.
 
     @return: text
     """
 
     return _apply_mapping(text, _accents_inverted, _accent_len_range)
 
 
 def _apply_mapping (text, mapping, mlenrange):
 
     p = 0
     pp = 0
     tsegs = []
     ltext = len(text)
     while p < ltext:
         for mlen in mlenrange:
             mapfrom = text[p:p + mlen]
             mapto = mapping.get(mapfrom)
             if mapto:
                 tsegs.append(text[pp:p])
                 tsegs.append(mapto)
                 p += mlen - 1
                 pp = p + 1
                 break
         p += 1
     tsegs.append(text[pp:p])
 
     return "".join(tsegs)