krkr.py

from glob import glob
from os.path import basename
import codecs
import subprocess
import pickle
import sys
seen = set()

tr = subprocess.Popen(["./trans","-s","japanese","-no-auto","-brief"],stdout=subprocess.PIPE,stdin=subprocess.PIPE,bufsize=1)

transcache = pickle.load(open("transcache.p","rb"))
sys.stdout.write((u"\ufeff").encode('utf-8'))

def do_trans(l):
    if l.strip() == "":
        return l
    if l in transcache:
        return transcache[l]
    ol = l
    l=l.encode('utf-8') + "\r\n"
    tr.stdin.write(l)
    res=tr.stdout.readline().strip().decode('utf-8').replace('"',"'")
    transcache[ol] = res
    pickle.dump(transcache, open("transcache.p","wb"))
    return res

def trans(l):
    if l.strip()=="":
        return ""
    parts = []
    # has a tag in it
    while '[' in l:
        p, l = l.split('[',1)
        parts.append(do_trans(p) + '[')
        p, l = l.split(']',1)
        parts.append(p + ']')
    parts.append(do_trans(l))
    return "".join(parts)


def pp(x):
    print(x.encode('utf-8'))

def quotes(l):
    return '"' + l.replace('"','""') + '"'

def body_flush(body):
    if not len(body[1]): return
    parts = []
    for ln in body[1]:
        parts.append(trans(ln))
    pp('"'+body[0]+'",'+quotes("\n".join(body[1]).rstrip()) + "," + quotes("\n".join(parts)))
    body[0] = None
    body[1] = []

def process_file(base,path):
    f = codecs.open(path, encoding='utf-16')
    special = [';', '@', '*','[','/']
    body = [None,[]]
    lineno = 0
    inscript = False
    for tl in f:
        tl = tl.rstrip('\n').rstrip('\r')
        l=tl.strip()
        goodtag = l.startswith('[r]') or l.startswith('[font')
        lineno+=1
        if body[0]:
            if not goodtag and (len(l) and (l[0] in special)):
                body_flush(body)
                continue
            body[1].append(tl)
        else:
            if l == "@iscript" or l == '[iscript]':
                inscript = True
            if l == "@endscript" or l == '[endscript]':
                inscript = False
                continue
            if inscript:
                continue
            if not len(l): continue
            if not goodtag and (l[0] in special):
                continue
            body[0] = "%s:%d" % (base,lineno)
            body[1] = []
            body[1].append(tl)

    body_flush(body)

for d in glob("data/patch/scenario/*.ks") + glob("data/scenario/*.ks") + glob("data/map/*.ks"):
    bn = basename(d)
    if bn in seen: continue
    seen.add(bn)
    process_file(bn,d)