Transliterator fails at certain digraphs

Here is my code:

def digraph(chars):
als = "шжяеёющчШЖЯЕЁЮЩЧ"
new = {'sh':als[0],'zh':als[1],'ja':als[2],'je':als[3],'jo':als[4],
       'ju':als[5],'sx':als[6],'ch':als[7],'Sh':als[8],'Zh':als[9],
       'Ja':als[10],'Je':als[11],'Jo':als[12],'Ju':als[13],'Sx':als[14],
       'Ch':als[15],'SH':als[8],'ZH':als[9],'JA':als[10],'JE':als[11],
       'JO':als[12],'JU':als[13],'SX':als[14],'CH':als[15]}
try:
    return new[chars]
except:
    return "[Error]"

def trans_cyr(inp):
cyrillic = "абцдэфгхийклмнопрстувъыьзАБЦДЭФГХИЙКЛМНОПРСТУВЫЗ "
latin = "abcdefghijklmnoprstuv$y'zABCDEFGHIJKLMNOPRSTUVYZ "
digs = ['sh','zh','ja','je','jo','ju','sx','ch','Sh','Zh',
        'Ja','Je','Jo','Ju','Sx','Ch','SH','ZH','JA','JE','JO','JU','SX',
        'CH']
prevc = ""
for e, char in enumerate(inp):
    if(prevc != ""):
        comb = prevc + char
        newdig = digraph(comb)
        if(comb in digs):
            print(newdig, end="")
            prevc = ""
        else:
            pos = latin.index(char)
            posp = latin.index(inp[e - 1])
            if(inp[e-1] in "szjcSZJC"):
                print(cyrillic[posp] + cyrillic[pos], end="")
                prevc = ""
            else:
                prevc=""
                continue
    elif(char not in "szjcSZJC"):
        try:
            pos = latin.index(char)
            print(cyrillic[pos], end="")
        except:
            print(char, end="")
    else:
        prevc = char

while True:
cyrinp = input("\n> ")
trans_cyr(cyrinp)

The code is supposed to transliterate the Latin alphabet to Cyrillic, first by getting each character from the input (if it is not 'szjc' or their uppercase equivalents), getting the position of it using the index() function and then acquiring a Cyrillic equivalent in the same position as the latin one. However, Cyrillic has letters such as Я, Е, Ё, Ю, Ж, Ш, Щ, Ч, which are digraphs (ya, ye, yo, yu, zh, sh, shch (sx), ch) and therefore cannot be transliterated by just one character. Thus what I do, is check whether the current letter is equal to any of 'szjcSZJC', and if it is then I do not print it but instead assign it the name prevc if the next character combined with prevc is in the array 'digs'. Everything works perfectly, if I type in 'jojajo' it will print "ёяё" just as it should, but - if there is an unfinished digraph (c without h, s without h, x, z without h, j without a, e, u, and o) then the next digraph does not become transliterated. Example: sjo : If I enter sjo, my expected output would be сё, but instead I am getting сйо. Is there any way I can fix this?

Edit:

I wrote this code:

while True:
cyrillic = "абцдэфгхийклмнопрстувъыьзАБЦДЭФГХИЙКЛМНОПРСТУВЫЗ "
latin = "abcdefghijklmnoprstuv$y'zABCDEFGHIJKLMNOPRSTUVYZ "
als = "шжяеёющчШЖЯЕЁЮЩЧ"
new = {'sh':als[0],'zh':als[1],'ja':als[2],'je':als[3],'jo':als[4],
       'ju':als[5],'sx':als[6],'ch':als[7],'Sh':als[8],'Zh':als[9],
       'Ja':als[10],'Je':als[11],'Jo':als[12],'Ju':als[13],'Sx':als[14],
       'Ch':als[15],'SH':als[8],'ZH':als[9],'JA':als[10],'JE':als[11],
       'JO':als[12],'JU':als[13],'SX':als[14],'CH':als[15]}
inp = input("\n> ") + " "
digraph = ""
prevc = ""
for e, char in enumerate(inp):
    part_j = "jJ"
    part_v = "aeouAEOU"
    part_z = "zZ"
    part_h = "hH"
    part_s = "sS"
    part_x = "hxHX"
    part_c = "cC"
    if((char in part_j and inp[e+1] in part_v) or (char in part_z and inp[e+1] in part_h) or (char in part_s and inp[e+1] in part_x) or (char in part_c and inp[e+1] in part_h)):
        digraph = "yes"
    else:
        digraph = "no"

    if((char in part_v and inp[e-1] in part_j) or (char in part_h and inp[e-1] in part_z) or (char in part_x and inp[e-1] in part_s) or (char in part_h and inp[e-1] in part_c)):
        comb = inp[e-1] + char
        dig = new[comb]
        print(dig, end="")
    elif(digraph == "yes"):
        prevc = char
    else:
        try:
            print(cyrillic[latin.index(char)],end="")
        except:
            print(char, end="")

which appears to have the same sort of logic as the answer I selected as correct, and it works :)

Solution

Here is a solution that uses the same code logic as your approach, but is more clearly written.

CYRILLIC = u"абцдэфгхийклмнопрстувъыьзАБЦДЭФГХИЙКЛМНОПРСТУВЫЗ "
LATIN = u"abcdefghijklmnoprstuv$y'zABCDEFGHIJKLMNOPRSTUVYZ "

DIGRAPHS = u"шжяеёющчШЖЯЕЁЮЩЧШЖЯЕЁЮЩЧ"
LATIN_DIGRAPHS = [u'sh', u'zh', u'ja', u'je', u'jo', u'ju', u'sx', u'ch',
                  u'Sh', u'Zh', u'Ja', u'Je', u'Jo', u'Ju', u'Sx', u'Ch',
                  u'SH', u'ZH', u'JA', u'JE', u'JO', u'JU', u'SX', u'CH']

MAPPING = dict(zip(list(LATIN) + LATIN_DIGRAPHS, CYRILLIC + DIGRAPHS))
DIGRAPH_FIRST_LETTER = u'szjcSZJC'

def latin_to_cyrillic(word):
    translation = []
    possible_digraph = False

    for letter in word:
        if possible_digraph:
            combination = previous_letter + letter
            if combination in LATIN_DIGRAPHS:
                translation.append(MAPPING[combination])
                possible_digraph = False
            else:
                translation.append(MAPPING[previous_letter])
                if letter in DIGRAPH_FIRST_LETTER:
                    previous_letter = letter
                else:
                    translation.append(letter)
                    possible_digraph = False
        else:
            if letter in DIGRAPH_FIRST_LETTER:
                possible_digraph = True
                previous_letter = letter
            else:
                translation.append(MAPPING[letter])
    if possible_digraph:
        translation.append(MAPPING[previous_letter])
    return ''.join(translation)

print latin_to_cyrillic('sjo')
print latin_to_cyrillic('jojajo')

The logic is as follows.

For each letter, check whether the previous letter was the beginning of a digraph.
If yes, check whether the previous and current letters are a legitimate digraph and translate them. If not, translate the previous letter as a single letter and check whether the current letter is the beginning of a digraph. If it is, store it in the buffer and move on, else translate that, too.
If the previous letter was not the beginning of a digraph, check whether the current letter is the beginning of a digraph. If yes, store it in the buffer and move, else translate it.

Instead of finding the index of a letter in the latin alphabet and using that index for the cyrillic alphabet, you can simply use a dictionary that maps each letter from one language to the other. Simply create a list of all the latin symbols (single and digraphs) and another one for the cyrillic. You only have to ensure that the respective order of symbols for both is the same. dict(zip(alphabet1, alphabet2)) will then create a mapping for each letter of alphabet1 to that of the same index in alphabet2.