# -*- coding: utf-8 -*-
import re, os
import codecs
from time import clock as now 

#import boot as PATH
zhPattern = re.compile(u'"[^"]*[\u4e00-\u9fa5]+[^"]*"')
zhPattern2 = re.compile(u'\[\[[^\"\[\]]*[\u4e00-\u9fa5]+[^\"\[\]]*\]\]')
base_path = os.path.abspath("../src")
base_path2 = os.path.abspath("done")
file_type = [".lua"]
all_chinese = {}
all_chinese_short = {}
start=now()  
def addOne(one, file_name):
    global all_chinese
    global all_chinese_short
    args = re.compile(u"[0-9.]+").findall(one)
    short = re.sub(u"[0-9.]+", "@@", one)
    args2 = re.compile(u"[^\u4e00-\u9fa5]+").findall(one)
    short2 = re.sub(u"[^\u4e00-\u9fa5]+", "##", one)
    if not args:
        args = []
    all_chinese[one] = {"base":one, "short":short, "args":args, "len":len(one), "args2":args2, "short2":short2, "file_name":file_name}
    all_chinese_short[short2] = one

def check(file_name, fn):
    print(file_name)
    global all_chinese
    i = 0       
    for line in codecs.open(file_name, "r", "utf-8"):
        line = re.sub("--.+", "", line)
        line = re.sub("print.+", "", line)
        i += 1
        match = zhPattern.findall(line)
        if match:
            for one in match:
                addOne(one[1:-1], fn[:-3] + str(i))
        match = zhPattern2.findall(line)
        if match:
            for one in match:
                addOne(one[2:-2], fn[:-3] + str(i))

def loop_path(path):
    global base_path
    for fn in os.listdir(path):
        file = os.path.join(path, fn)
        if os.path.isdir(file):
            loop_path(file)
            continue
        elif os.path.splitext(fn)[1] in file_type:
            check(file, path[len(base_path)+1:]+"/"+fn)
loop_path(base_path)
all_short = {}
for key in all_chinese:
    short = all_chinese[key]["short2"]
    all_short[short] = [all_chinese[key]["file_name"], key]
all_short = sorted(all_short.iteritems(), key=lambda asd:asd[1][0], reverse = False)
f2 = codecs.open("all_word_short.txt", "w", "utf-8")
f3 = codecs.open("all_word_base.txt", "w", "utf-8")
for one in all_short:
    # print(one[0], one[1])
    f2.write("%s\t%s\t%s\n" % (one[1][1], one[0], one[1][0]))
    f3.write("%s\n" % (one[0]))
f2.close()
f3.close()

f = codecs.open("all_word.txt", "w", "utf-8")
for key in all_chinese:
    one = all_chinese[key]
    f.write("%s\n" % (one["base"]))
f.close()
print "finish!" 
end = now()  
time_last = end - start  
print 'use time:',time_last,'sec'