asl2ibus_table.py

#!/usr/bin/env python
# coding: utf-8

# In[1]:


from xml.dom import minidom

with open('ogsl.asl', 'r') as infile:
    asl = infile.read()


# In[2]:


reps = [
['š','sz'],
['Š','SZ'],
['ṣ','s,'],
['Ṣ','S,'],
['ṭ','t,'],
['Ṭ','T,'],
['ś',"s'"],
['Ś',"S'"],
['ʾ',"'"],
['₀','0'],
['₁','1'],
['₂','2'],
['₃','3'],
['₄','4'],
['₅','5'],
['₆','6'],
['₇','7'],
['₈','8'],
['₉','9'],
['ₓ','x'],
['ḫ','h,'],
['Ḫ','H,'],
['ŋ','j'],
['Ŋ','J'],
['û','u'],
['ā','aa'],
['ē','ee'],
['ī','ii'],
['ū','uu'],
    ['×','x'],
    ['°','0'],
    ['·','.'],
    ['⁺','+'],
    ['⁻','-'],
    ['𒑱',':'],
]

def to_ascii(s):
    for fro, to in reps:
        s=s.replace(fro, to)
    return s
    


# In[3]:


def unistr2unicode(x):
    x=x.replace('X', "x58") # replace X with unicode X
    x=x.replace('None', "x58") # replace None with unicode X (|UD.MA₂.AB×(U.U.U).ŠIR|)
    return ''.join(chr(int(y[1:], base=16)) for y in x.split('.'))
print(unistr2unicode("x12000.x12094.x1223E"))
print(unistr2unicode("x12000"))


# In[4]:


def maybeadd(reading, unic, sign='', prio = 1000):
    reading=to_ascii(reading).strip('|')
    sign=to_ascii(sign).strip('|')
    if reading not in mapping:
        mapping[reading]={(unic,sign,prio)}
    else:
        mapping[reading].add((unic,sign,prio))


# In[5]:


mapping = {}

funic = None
unic = None
in_form = False
listcodes = []
in_sign = False

for line in asl.splitlines():
    if line =="":
        continue
    if line.startswith("@sign"):
        in_sign = True
        sign = line.split()[1]
                
    if line.strip()=="@end sign" :
        if in_sign:
            if sign and unic and not in_form:
                maybeadd(sign, unic, sign)
            if sign and funic and in_form: # end sign also ends last form
                maybeadd(sign, funic, sign)
            if listcodes != [] and unic:
                for listcode in listcodes:
                    maybeadd(listcode, unic, sign, 500)
        listcodes=[]
        in_form = False
        unic = None
        funic = None
        sign = None
        in_sign = False
            
    if line.startswith("@list") and not in_form:
        listcode = line.split()[1]
        listcodes.append(listcode)
        
    if line.startswith("@v"):
        reading = line.split()[1]
        if sign and unic and not in_form:
            maybeadd(reading, unic, sign)
        if sign and funic and  in_form:
            maybeadd(reading, funic, sign)
        
    if line.startswith("@ucode"):
        if in_form:
            funic = unistr2unicode(line.split()[1])
        else:
            unic = unistr2unicode(line.split()[1])
    #if line.startswith("@uname"):
    #    uname = unistr2unicode(line.split(maxsplit=1))[1]
        
    if line.startswith("@form"):
        in_form = True
        foo, formcode, sign, *foo = line.split()
        
    if line=="@end form":
        in_form = False
        if sign and funic:
            maybeadd(sign, funic)
        funic = None

    if line.startswith("@form"):
        pass
print(mapping['darengal'])
print(mapping['LAK797'])
print(mapping["'u4"])
print(mapping["libiszx"])
#print(mapping["MZL224"])


# In[6]:


valid_chars = ''.join(sorted(list(set(''.join(k for k in mapping.keys())))))
#''.join(k for k in mapping.keys())
valid_chars
print(list(mapping["'u4"]))


# In[7]:


inf = ""
from calendar import timegm
from time import gmtime

with open("ibus-table-cuneiform_pre.txt",'r') as infile:
    inf = infile.read()
    inf = inf.format(VALID_INPUTS= valid_chars, SERIAL=timegm(gmtime()))
with open("ibus-table-cuneiform.txt",'w') as outfile:
    outfile.write(inf)
    for reading, vals in mapping.items():
        if len(vals)>1:
            for unic,name,prio in sorted(list(vals)):
                outfile.write(f"{reading}({name})\t{unic}\t{prio}\t### {name}\n")
        else:
            unic,name, prio = vals.pop()
            outfile.write(f"{reading}\t{unic}\t{prio}\t### {name}\n")
    outfile.write("END_TABLE")
print("Written")