Upload 3 files

v-corpus-zh/橘子班/高考恋爱100天/0.txt.gz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f82b72f1ec279163ac40c35b55a1f880edfff6c129b318ec626a2e26db0f143
+size 430060

v-corpus-zh/橘子班/高考恋爱100天/extractdata.py

@@ -0,0 +1,250 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# https://github.com/gumblex/extract-gk100d
+
+# conda create -n gk100d conda-forge::python
+# conda activate gk100d
+# pip install python-magic-bin
+# pip install python-magic
+# pip show python-magic
+
+# envs\gk100d\Lib\site-packages\magic
+# 修改 `loader.py` ，在 `def _lib_candidates():` 中添加
+# ```python
+# def _lib_candidates():
+#  yield r'C:\Users\limou\miniconda3\envs\gk100d\Lib\site-packages\magic\libmagic\libmagic.dll'
+# ```
+
+# python extractdata.py -e data.xp3 output/
+# conda remove -n gk100d --all
+
+import os
+import io
+import sys
+import zlib
+import struct
+import shutil
+import itertools
+import mimetypes
+import collections
+
+xor = lambda t, k: bytes(x^y for x,y in zip(t, itertools.cycle(k)))
+
+def assert_string(infile, value):
+    assert infile.read(len(value)) == value
+
+
+class AttrDict(dict):
+    def __init__(self, *args, **kwargs):
+        super(AttrDict, self).__init__(*args, **kwargs)
+        self.__dict__ = self
+
+
+SegmChunk = collections.namedtuple('SegmChunk', 'flags offset size size_comp')
+
+MAGICS = (
+('ogg', b'\x4f\x67\x67\x53', b'\x00\x02\x00\x00\x00\x00\x00\x00'),
+('wmv', b'\x30\x26\xB2\x75', b'\x8E\x66\xCF\x11\xA6\xD9\x00\xAA'),
+('tjs', b'\xFF\xFE\x76\x00', b'\x61\x00\x72\x00\x20\x00\x61\x00'),
+)
+
+class Xp3File:
+    MAGIC = b'XP3\r\n\x20\x0A\x1A\x8B\x67\x01'
+    KEYHEAD = b'\x0c\xf0\x04a\x00JB\x00'
+
+    def __init__(self, filename):
+        self.filename = filename
+        self.fp = open(self.filename, 'rb')
+        self.table = []
+        assert_string(self.fp, self.MAGIC)
+        self.fp.seek(19)
+        self.version = 2 if self._read_int(4) == 1 else 1
+        self.fp.seek(len(self.MAGIC))
+        if self.version == 2:
+            self.additional_header_offset = self._read_int(8)
+            self.minor_version = self._read_int(4)
+            assert self.minor_version == 1
+            self.fp.seek(self.additional_header_offset)
+            # 80
+            self.flags = self._read_int(1)
+            # 00 00 00 00 00 00 00 00
+            self.fp.read(8)
+            self.table_offset = self._read_int(8)
+        else:
+            self.table_offset = self._read_int(8)
+        self.read_table()
+
+    def read_table(self):
+        self.fp.seek(self.table_offset)
+        self.table_compressed = self._read_int(1)
+        self.table_size_comp = self._read_int(8)
+        if self.table_compressed:
+            self.table_size_orig = self._read_int(8)
+        else:
+            self.table_size_orig = self.table_size_comp
+        table_data = self.fp.read(self.table_size_comp)
+        if self.table_compressed:
+            table_data = zlib.decompress(table_data)
+        self.table_raw = table_data
+        stream = io.BytesIO(table_data)
+        while 1:
+            magic = stream.read(4)
+            if not magic:
+                break
+            assert magic == b'File'
+            d = AttrDict()
+            d.length = self._read_int(8, stream)
+            assert_string(stream, b'info')
+            d.info_length = self._read_int(8, stream)
+            d.flags = self._read_int(4, stream)
+            d.size = self._read_int(8, stream)
+            d.size_comp = self._read_int(8, stream)
+            offset = stream.tell()
+            # should be name
+            name_length = self._read_int(2, stream)
+            if d.info_length == 22 + name_length*2:
+                d.name = stream.read(name_length*2).decode('utf-16-le', 'ignore')
+                d.name_good = True
+            else:
+                stream.seek(offset)
+                d.name = stream.read(12)
+                d.name_good = False
+            assert_string(stream, b'segm')
+            d.segm_number = self._read_int(8, stream) // 28
+            d.segm = []
+            for i in range(d.segm_number):
+                d.segm.append(SegmChunk(
+                    self._read_int(4, stream),
+                    self._read_int(8, stream),
+                    self._read_int(8, stream),
+                    self._read_int(8, stream)
+                ))
+            assert_string(stream, b'adlr')
+            # length
+            self._read_int(8, stream)
+            d.adlr = stream.read(4)
+            self.table.append(d)
+
+    def get(self, num, decrypt=True):
+        fileinfo = self.table[num]
+        outbuffer = io.BytesIO()
+        for segm in fileinfo.segm:
+            self.fp.seek(segm.offset)
+            data = self.fp.read(segm.size_comp)
+            if segm.flags & 7: # compressed
+                data = zlib.decompress(data)
+            if decrypt:
+                data = self.decrypt(fileinfo, data)
+            assert len(data) == segm.size
+            outbuffer.write(data)
+        outbuffer.seek(0)
+        return outbuffer
+
+    def extract(self, out):
+        for k, fileinfo in enumerate(self.table, 1):
+            outbuffer = self.get(k-1)
+            if fileinfo.name_good:
+                filename = fileinfo.name
+                if len(filename.encode('utf-8')) > 128:
+                    fn, ext = os.path.splitext(filename)
+                    filename = fn[:64] + ext
+            else:
+                ext = self.detect_ext(outbuffer.read(1024))
+                filename = ('%04d' % k) + (ext or '.txt')
+            outbuffer.seek(0)
+            dirname = os.path.dirname(filename)
+            os.makedirs(os.path.join(out, dirname), exist_ok=True)
+            with open(os.path.join(out, filename), 'wb') as f:
+                shutil.copyfileobj(outbuffer, f)
+            print('Extracted %s' % filename)
+
+    def detect_ext(self, data):
+        import magic
+        try:
+            fmagic = magic.detect_from_content(data)
+        except AttributeError as e:
+            print(e)
+            return '.ukn'
+        if fmagic.mime_type.startswith('text/'):
+            if fmagic.encoding == 'unknown-8bit':
+                ext = '.bin'
+            else:
+                text = data.decode(fmagic.encoding)
+                if '@return' in text or '*start' in text or '.ks' in text or '[w]' in text:
+                    ext = '.ks'
+                elif '.tjs' in text or '%[' in text or '];' in text:
+                    ext = '.tjs'
+                else:
+                    ext = '.txt'
+        else:
+            ext = mimetypes.guess_extension(fmagic.mime_type)
+            if ext == '.jpeg':
+                ext = '.jpg'
+            elif ext == '.oga':
+                ext = '.ogg'
+            elif ext == '.asf':
+                ext = '.wmv'
+        return ext
+
+    def _read_int(self, size, infile=None, endian='<', signed=False):
+        inttypes = {1: 'B', 2: 'H', 4: 'I', 8: 'Q'}
+        d = (infile or self.fp).read(size)
+        if signed:
+            return struct.unpack(endian + inttypes[size].lower(), d)[0]
+        else:
+            return struct.unpack(endian + inttypes[size], d)[0]
+
+    def decrypt(self, fileinfo, data):
+        return data
+
+    def close(self):
+        self.fp.close()
+
+    def __del__(self):
+        self.fp.close()
+
+class EncryptedXp3File(Xp3File):
+
+    def load_key(self):
+        if hasattr(self, 'keyhead'):
+            return
+        # Gaokao Love 100 Days Disc Ver.
+        elif self.table_offset == 1631288384:
+            self.keyhead = b'\x0c\xf0\x04a\x00JB\x00'
+        else:
+            self.keyhead = b'\x1d\xef[\xa3\x00\xcaA\x00'
+
+    def detect_key(self):
+        for k, fileinfo in enumerate(self.table):
+            segm = fileinfo.segm[0]
+            self.fp.seek(segm.offset)
+            if segm.flags & 7: # compressed
+                data = self.fp.read(segm.size_comp)
+                data = zlib.decompress(data)[:12]
+            else:
+                data = self.fp.read(12)
+            for ftype, head1, head2 in MAGICS:
+                if xor(head1, fileinfo.adlr) == data[:4]:
+                    self.keyhead = xor(head2, data[4:])
+                    return self.keyhead
+
+    def decrypt(self, fileinfo, data):
+        self.load_key()
+        return xor(data, fileinfo.adlr + self.keyhead)
+
+
+def main():
+    args = sys.argv[1:]
+    encrypted = False
+    if args[0] == '-e':
+        encrypted = True
+        args.pop(0)
+        xp3file = EncryptedXp3File(args[0])
+    else:
+        xp3file = Xp3File(args[0])
+    print('File loaded.')
+    xp3file.extract(args[1])
+
+if __name__ == '__main__':
+    main()

v-corpus-zh/橘子班/高考恋爱100天/gk100d_ks.py

@@ -0,0 +1,203 @@
+import sys
+sys.path.append('..')
+
+from common import \
+    get_all_files_in_directory, \
+    startsWithAny, getCmdArgsMap, \
+    clearT, removeWait, log_process
+# https://huggingface.co/datasets/Limour/archvie/blob/main/v-corpus-zh/common.py
+
+# =================
+
+a = get_all_files_in_directory(r'E:\galgame\gk100d\ks', ext='.ks')
+b = r'E:\galgame\gk100d\text'
+
+# =================
+
+sc = {}
+_n = {
+  "【木  馨" : "木馨",
+  "【？ ？" : "?",
+  "【我" : "我",
+  "【曲  琪" : "曲琪",
+  "【乔  伊" : "乔伊",
+  "【魏  通" : "魏通",
+  "【梅老师" : "梅老师",
+  "【母  亲" : "母亲",
+  "【罗小涵" : "罗小涵",
+  "【父  亲" : "父亲",
+  "【西安老师" : "西安老师",
+  "【男生 A" : "男生A",
+  "【男生 B" : "男生B",
+  "【女生 A" : "女生A",
+  "【女生A" : "女生A",
+  "【老师A" : "老师A",
+  "【女生 B" : "女生B",
+  "【老  师" : "老师",
+  "【店老板" : "店老板",
+  "【副团长" : "副团长",
+  "【女生 C" : "女生C",
+  "【老  鼠" : "老鼠",
+  "【团  长" : "团长",
+  "【工作人员" : "工作人员",
+  "【木馨的母亲" : "木馨的母亲",
+  "【校  医" : "校医",
+  "【学员 A" : "学员A",
+  "【学员 B" : "学员B",
+  "【众学员" : "众学员",
+  "【教  官" : "教官",
+  "【学员 C" : "学员C",
+  "【学员 D" : "学员D",
+  "【保洁员" : "保洁员",
+  "【女  生" : "女生",
+  "【小商贩" : "小商贩",
+  "【男生A" : "男生A",
+  "【路过的学生" : "路过的学生",
+  "【演讲者" : "演讲者",
+  "【韩状元" : "韩状元",
+  "【外校学生" : "外校学生",
+  "【高一学生" : "高一学生",
+  "【小卖部大叔" : "小卖部大叔",
+  "【外校生A" : "外校生A",
+  "【外校生B" : "外校生B",
+  "【外校生C" : "外校生C",
+  "【外校生D" : "外校生D",
+  "【服务员" : "服务员",
+  "【皇阿玛" : "皇阿玛",
+  "【女同学A " : "女同学A",
+  "【众  人" : "众人",
+  "【男生B" : "男生B",
+  "【女生B" : "女生B",
+  "【众同学" : "众同学",
+  "【同学A" : "同学A",
+  "【女生" : "女生",
+  "【中年女性" : "中年女性",
+  "【老师B " : "老师B",
+  "【老师A " : "老师A",
+  "【女生A " : "女生A",
+  "【女生B " : "女生B",
+  "【小涵的母亲" : "小涵的母亲",
+  "【旁边的小孩" : "旁边的小孩",
+  "【罗小涵\"" : "罗小涵",
+  "【男生C" : "男生C",
+  "【女生C" : "女生C",
+  "【？？" : "?",
+  "【西安老师\"" : "西安老师",
+  "【女生C " : "女生C",
+  "【男生A " : "男生A",
+  "【男生B " : "男生B",
+  "【男生C " : "男生C",
+  "【同学B" : "同学B",
+  "【流浪汉" : "流浪汉",
+  "【清洁工" : "清洁工",
+  "【服务生" : "服务生",
+  "【木馨" : "木馨",
+  "【医  生" : "医生",
+  "【护  士" : "护士",
+  "【网吧青年" : "网吧青年",
+  "【老  板" : "老板",
+  "【网吧老板" : "网吧老板",
+  "【警  察" : "警察",
+  "【书店老板" : "书店老板",
+  "【同学 A" : "同学A",
+  "【店  员" : "店员",
+  "【记  者" : "记者",
+  "【警  卫" : "警卫",
+  "【父 亲" : "父亲",
+  "【工作人员A" : "工作人员A",
+  "【工作人员B" : "工作人员B",
+  "【工作人员\"" : "工作人员",
+  "【肌肉男" : "肌肉男",
+  "【推销人员" : "推销人员",
+  "【矮个子" : "矮个子",
+  "【神秘人" : "神秘人",
+  "【志愿者" : "志愿者",
+  "【白发老人" : "白发老人",
+  "【店  长" : "店长",
+  "【队  友" : "队友",
+  "【店员A" : "店员A",
+  "【店员B" : "店员B",
+  "【店长" : "店长",
+  "【顾客" : "顾客",
+  "【众店员" : "众店员",
+  "【木馨的父亲" : "木馨的父亲",
+  "【室友A" : "室友A",
+  "【室友B" : "室友B",
+  "【室友C" : "室友C",
+  "【学  长" : "学长"
+}
+
+# =================
+
+for path in a:
+    name: str = path[path.rindex('\\') + 1:]
+    name = '0'
+    if name not in sc:
+        sc[name] = []
+        print(name)
+
+    # =================
+    try:
+        f = open(path, 'r', encoding='utf-16-le')
+        data = list(x.rstrip() for x in f.readlines())
+    except UnicodeDecodeError:
+        f = open(path, 'r', encoding='gbk')
+        data = list(x.rstrip() for x in f.readlines())
+    finally:
+        f.close()
+    # =================
+    if any(('{' in line) for line in data):
+        continue
+    # =================
+    data[0] = data[0].lstrip('\ufeff')
+    # print(data[0][0])
+    def _filter(x: str):
+        if not x:
+            return False
+        if startsWithAny(x, {'*', ';', '[', '//'}):
+            return False
+        return True
+    data = list(filter(_filter, data))
+
+    # =================
+    n = ''
+    w_i = -1
+    while w_i < len(data) - 1:
+        w_i += 1
+        line: str = data[w_i]
+        # =================
+        if line.startswith('@dia'):
+            n = ''
+            continue
+        elif line.startswith('@主角'):
+            n = '我'
+            continue
+        elif line.startswith('@npc'):
+            tmp = getCmdArgsMap(line)
+            if 'id' in tmp:
+                n = tmp['id']
+            else:
+                print(line)
+            continue
+        elif line.startswith('@'):
+            n = ''
+            continue
+        if line.isascii():
+            continue
+        if n:
+            d = log_process(f'【{n}】{removeWait(line)}', line, _n)
+        else:
+            d = log_process(f'{removeWait(line)}', line, _n)
+        if d:
+            sc[name].append(d)
+
+# =================
+
+for k, v in sc.items():
+    if v:
+        with open(b + f'\\{k}.txt', 'w', encoding='utf-8') as f:
+            f.write('\n'.join(v))
+
+# =================
+import json
+tmp = json.dumps(_n, ensure_ascii=False, indent=4)