import os
import re

patt = re.compile (r"sub([0-9]+)_([0-9]+).(in|out)")
cnt = {  }
print ("将 subXXX_XXX.in/out 转化为 dataXXX.in/out 的脚本")
path = input ("转化目录（如果是当前目录，输入 .）：")
os.chdir (path)

for i in os.listdir ("."):
    if os.path.isdir (i):
        print ("跳过子目录 %s" % i)
        continue
    q = re.findall (patt, i)
    if not q or len (q) != 1:
        print ("%s 不匹配格式 subXXX_XXX.in 或 subXXX_XXX.out，跳过" % i)
        continue
    q = q[0]
    subid = int (q[0])
    pid = int (q[1])
    suff = q[2]
    try:
        cnt[subid] += 1
    except KeyError:
        cnt[subid] = 1
cnt[0] = 0
for i in range (1, len (cnt)):
    try:
        cnt[i] += cnt[i - 1]
    except KeyError:
        raise Exception ("错误：subtask 编号不是连续的！自 %d 处断开" % i)
for i in range (1, len (cnt)):
    if cnt[i] % 2 == 1:
        raise Exception ("Subtask #%d 中 in/out 不配对（文件数量是奇数）" % i)
    cnt[i] /= 2

for i in range (1, len (cnt)):
    os.mkdir ("subtask%d" % i)
    print ("已创建子目录 subtask%d" % i)
copied = 0
covered = []

for i in os.listdir ("."):
    if os.path.isdir (i):
        continue
    q = re.findall (patt, i)
    if not q or len (q) != 1:
        continue
    q = q[0]
    subid = int (q[0])
    pid = int (q[1])
    suff = q[2]

    src = i
    newname = "data%d.%s" % (cnt[subid - 1] + pid, suff)
    dst = "subtask%d/" % subid + newname

    print ("%s -> %s" % (src, dst), end='')
    if os.path.exists (dst):
        print ("（覆盖！）")
        covered.append (dst)
    else:
        print ("")

    os.replace (src, dst)
    copied += 1
print ("已复制 %d 个文件，覆盖了 %d 个文件" % (copied, len (covered)))
print ("被覆盖文件：")
if len (covered) == 0:
    print ("（无）")
else:
    for i in covered:
        print ("--> \t", i)