File size: 18,931 Bytes
8ad9e26
 
 
 
 
 
 
 
 
 
 
 
b6fb8c4
 
8ad9e26
 
 
 
 
 
 
 
 
fa4087a
8ad9e26
 
fa4087a
5cb0bc3
8ad9e26
 
 
 
 
 
 
 
8ba98ee
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5cb0bc3
 
 
8ba98ee
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c67c8ad
 
 
 
 
 
8ad9e26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5cb0bc3
8ad9e26
 
 
5cb0bc3
8ad9e26
 
 
 
 
 
 
5cb0bc3
 
 
 
8ad9e26
 
 
b6fb8c4
 
 
 
 
8ad9e26
 
8ba98ee
 
 
 
 
 
8ad9e26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
fa4087a
8ba98ee
 
8ad9e26
 
5cb0bc3
 
 
 
 
8ad9e26
 
 
 
 
8ba98ee
8ad9e26
8ba98ee
 
8ad9e26
 
 
 
 
 
 
8ba98ee
8ad9e26
 
 
 
 
 
 
 
 
fa4087a
8ad9e26
 
 
 
 
 
fa4087a
8ba98ee
5cb0bc3
 
 
 
8ad9e26
 
 
8ba98ee
8ad9e26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
fa4087a
8ad9e26
 
 
 
8ba98ee
8ad9e26
 
 
 
8ba98ee
8ad9e26
 
 
 
 
 
 
 
 
 
 
 
fa4087a
 
 
8ad9e26
 
fa4087a
 
 
8ad9e26
 
 
 
 
fa4087a
8ad9e26
 
 
 
 
 
 
976b4d3
 
8ad9e26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
fa4087a
 
8ad9e26
 
 
 
 
 
 
8ba98ee
8ad9e26
 
8ba98ee
8ad9e26
 
 
 
 
8ba98ee
8ad9e26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8ba98ee
8ad9e26
 
 
 
 
 
 
 
 
 
 
 
 
b6fb8c4
8ad9e26
 
 
 
b6fb8c4
 
 
 
8ba98ee
b6fb8c4
 
 
 
 
 
 
 
 
 
 
5cb0bc3
 
b6fb8c4
 
 
 
 
c67c8ad
 
 
 
 
b6fb8c4
 
 
 
 
 
 
 
 
 
 
 
5cb0bc3
b6fb8c4
 
 
c67c8ad
 
 
 
5cb0bc3
c67c8ad
b6fb8c4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
fa4087a
 
8ba98ee
 
fa4087a
 
 
 
 
 
8ba98ee
fa4087a
 
 
 
 
 
 
8ba98ee
fa4087a
 
 
 
 
8ba98ee
fa4087a
 
 
8ba98ee
 
 
 
 
 
 
 
 
 
c67c8ad
5cb0bc3
 
 
c67c8ad
 
 
 
 
 
5cb0bc3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
# -*- coding:utf-8 -*-
from __future__ import annotations
from typing import TYPE_CHECKING, Any, Callable, Dict, List, Tuple, Type
import logging
import json
import os
import datetime
import hashlib
import csv
import requests
import re
import html
import sys
import subprocess

import gradio as gr
from pypinyin import lazy_pinyin
import tiktoken
import mdtex2html
from markdown import markdown
from pygments import highlight
from pygments.lexers import get_lexer_by_name
from pygments.formatters import HtmlFormatter
import pandas as pd

from modules.presets import *
from . import shared
from modules.config import retrieve_proxy, hide_history_when_not_logged_in

if TYPE_CHECKING:
    from typing import TypedDict

    class DataframeData(TypedDict):
        headers: List[str]
        data: List[List[str | int | bool]]

def predict(current_model, *args):
    iter = current_model.predict(*args)
    for i in iter:
        yield i

def billing_info(current_model):
    return current_model.billing_info()

def set_key(current_model, *args):
    return current_model.set_key(*args)

def load_chat_history(current_model, *args):
    return current_model.load_chat_history(*args)

def interrupt(current_model, *args):
    return current_model.interrupt(*args)

def reset(current_model, *args):
    return current_model.reset(*args)

def retry(current_model, *args):
    iter = current_model.retry(*args)
    for i in iter:
        yield i

def delete_first_conversation(current_model, *args):
    return current_model.delete_first_conversation(*args)

def delete_last_conversation(current_model, *args):
    return current_model.delete_last_conversation(*args)

def set_system_prompt(current_model, *args):
    return current_model.set_system_prompt(*args)

def save_chat_history(current_model, *args):
    return current_model.save_chat_history(*args)

def export_markdown(current_model, *args):
    return current_model.export_markdown(*args)

def load_chat_history(current_model, *args):
    return current_model.load_chat_history(*args)

def upload_chat_history(current_model, *args):
    return current_model.load_chat_history(*args)

def set_token_upper_limit(current_model, *args):
    return current_model.set_token_upper_limit(*args)

def set_temperature(current_model, *args):
    current_model.set_temperature(*args)

def set_top_p(current_model, *args):
    current_model.set_top_p(*args)

def set_n_choices(current_model, *args):
    current_model.set_n_choices(*args)

def set_stop_sequence(current_model, *args):
    current_model.set_stop_sequence(*args)

def set_max_tokens(current_model, *args):
    current_model.set_max_tokens(*args)

def set_presence_penalty(current_model, *args):
    current_model.set_presence_penalty(*args)

def set_frequency_penalty(current_model, *args):
    current_model.set_frequency_penalty(*args)

def set_logit_bias(current_model, *args):
    current_model.set_logit_bias(*args)

def set_user_identifier(current_model, *args):
    current_model.set_user_identifier(*args)

def set_single_turn(current_model, *args):
    current_model.set_single_turn(*args)

def handle_file_upload(current_model, *args):
    return current_model.handle_file_upload(*args)

def like(current_model, *args):
    return current_model.like(*args)

def dislike(current_model, *args):
    return current_model.dislike(*args)


def count_token(message):
    encoding = tiktoken.get_encoding("cl100k_base")
    input_str = f"role: {message['role']}, content: {message['content']}"
    length = len(encoding.encode(input_str))
    return length


def markdown_to_html_with_syntax_highlight(md_str):
    def replacer(match):
        lang = match.group(1) or "text"
        code = match.group(2)

        try:
            lexer = get_lexer_by_name(lang, stripall=True)
        except ValueError:
            lexer = get_lexer_by_name("text", stripall=True)

        formatter = HtmlFormatter()
        highlighted_code = highlight(code, lexer, formatter)

        return f'<pre><code class="{lang}">{highlighted_code}</code></pre>'

    code_block_pattern = r"```(\w+)?\n([\s\S]+?)\n```"
    md_str = re.sub(code_block_pattern, replacer, md_str, flags=re.MULTILINE)

    html_str = markdown(md_str)
    return html_str


def normalize_markdown(md_text: str) -> str:
    lines = md_text.split("\n")
    normalized_lines = []
    inside_list = False

    for i, line in enumerate(lines):
        if re.match(r"^(\d+\.|-|\*|\+)\s", line.strip()):
            if not inside_list and i > 0 and lines[i - 1].strip() != "":
                normalized_lines.append("")
            inside_list = True
            normalized_lines.append(line)
        elif inside_list and line.strip() == "":
            if i < len(lines) - 1 and not re.match(
                r"^(\d+\.|-|\*|\+)\s", lines[i + 1].strip()
            ):
                normalized_lines.append(line)
            continue
        else:
            inside_list = False
            normalized_lines.append(line)

    return "\n".join(normalized_lines)


def convert_mdtext(md_text):
    code_block_pattern = re.compile(r"```(.*?)(?:```|$)", re.DOTALL)
    inline_code_pattern = re.compile(r"`(.*?)`", re.DOTALL)
    code_blocks = code_block_pattern.findall(md_text)
    non_code_parts = code_block_pattern.split(md_text)[::2]

    result = []
    raw = f'<div class="raw-message hideM">{html.escape(md_text)}</div>'
    for non_code, code in zip(non_code_parts, code_blocks + [""]):
        if non_code.strip():
            non_code = normalize_markdown(non_code)
            result.append(markdown(non_code, extensions=["tables"]))
        if code.strip():
            # _, code = detect_language(code)  # 暂时去除代码高亮功能,因为在大段代码的情况下会出现问题
            # code = code.replace("\n\n", "\n") # 暂时去除代码中的空行,因为在大段代码的情况下会出现问题
            code = f"\n```{code}\n\n```"
            code = markdown_to_html_with_syntax_highlight(code)
            result.append(code)
    result = "".join(result)
    output = f'<div class="md-message">{result}</div>'
    output += raw
    output += ALREADY_CONVERTED_MARK
    return output


def convert_asis(userinput):
    return (
        f'<p style="white-space:pre-wrap;">{html.escape(userinput)}</p>'
        + ALREADY_CONVERTED_MARK
    )


def detect_converted_mark(userinput):
    try:
        if userinput.endswith(ALREADY_CONVERTED_MARK):
            return True
        else:
            return False
    except:
        return True


def detect_language(code):
    if code.startswith("\n"):
        first_line = ""
    else:
        first_line = code.strip().split("\n", 1)[0]
    language = first_line.lower() if first_line else ""
    code_without_language = code[len(first_line) :].lstrip() if first_line else code
    return language, code_without_language


def construct_text(role, text):
    return {"role": role, "content": text}


def construct_user(text):
    return construct_text("user", text)


def construct_system(text):
    return construct_text("system", text)


def construct_assistant(text):
    return construct_text("assistant", text)


def save_file(filename, system, history, chatbot, user_name):
    logging.debug(f"{user_name} 保存对话历史中……")
    os.makedirs(os.path.join(HISTORY_DIR, user_name), exist_ok=True)
    if filename.endswith(".json"):
        json_s = {"system": system, "history": history, "chatbot": chatbot}
        if "/" in filename or "\\" in filename:
            history_file_path = filename
        else:
            history_file_path = os.path.join(HISTORY_DIR, user_name, filename)
        with open(history_file_path, "w") as f:
            json.dump(json_s, f)
    elif filename.endswith(".md"):
        md_s = f"system: \n- {system} \n"
        for data in history:
            md_s += f"\n{data['role']}: \n- {data['content']} \n"
        with open(os.path.join(HISTORY_DIR, user_name, filename), "w", encoding="utf8") as f:
            f.write(md_s)
    logging.debug(f"{user_name} 保存对话历史完毕")
    return os.path.join(HISTORY_DIR, user_name, filename)


def sorted_by_pinyin(list):
    return sorted(list, key=lambda char: lazy_pinyin(char)[0][0])


def get_file_names(dir, plain=False, filetypes=[".json"]):
    logging.debug(f"获取文件名列表,目录为{dir},文件类型为{filetypes},是否为纯文本列表{plain}")
    files = []
    try:
        for type in filetypes:
            files += [f for f in os.listdir(dir) if f.endswith(type)]
    except FileNotFoundError:
        files = []
    files = sorted_by_pinyin(files)
    if files == []:
        files = [""]
    logging.debug(f"files are:{files}")
    if plain:
        return files
    else:
        return gr.Dropdown.update(choices=files)


def get_history_names(plain=False, user_name=""):
    logging.debug(f"从用户 {user_name} 中获取历史记录文件名列表")
    if user_name == "" and hide_history_when_not_logged_in:
        return ""
    else:
        return get_file_names(os.path.join(HISTORY_DIR, user_name), plain)


def load_template(filename, mode=0):
    logging.debug(f"加载模板文件{filename},模式为{mode}(0为返回字典和下拉菜单,1为返回下拉菜单,2为返回字典)")
    lines = []
    if filename.endswith(".json"):
        with open(os.path.join(TEMPLATES_DIR, filename), "r", encoding="utf8") as f:
            lines = json.load(f)
        lines = [[i["act"], i["prompt"]] for i in lines]
    else:
        with open(
            os.path.join(TEMPLATES_DIR, filename), "r", encoding="utf8"
        ) as csvfile:
            reader = csv.reader(csvfile)
            lines = list(reader)
        lines = lines[1:]
    if mode == 1:
        return sorted_by_pinyin([row[0] for row in lines])
    elif mode == 2:
        return {row[0]: row[1] for row in lines}
    else:
        choices = sorted_by_pinyin([row[0] for row in lines])
        return {row[0]: row[1] for row in lines}, gr.Dropdown.update(
            choices=choices
        )


def get_template_names(plain=False):
    logging.debug("获取模板文件名列表")
    return get_file_names(TEMPLATES_DIR, plain, filetypes=[".csv", "json"])


def get_template_content(templates, selection, original_system_prompt):
    logging.debug(f"应用模板中,选择为{selection},原始系统提示为{original_system_prompt}")
    try:
        return templates[selection]
    except:
        return original_system_prompt


def reset_textbox():
    logging.debug("重置文本框")
    return gr.update(value="")


def reset_default():
    default_host = shared.state.reset_api_host()
    retrieve_proxy("")
    return gr.update(value=default_host), gr.update(value=""), "API-Host 和代理已重置"


def change_api_host(host):
    shared.state.set_api_host(host)
    msg = f"API-Host更改为了{host}"
    logging.info(msg)
    return msg


def change_proxy(proxy):
    retrieve_proxy(proxy)
    os.environ["HTTPS_PROXY"] = proxy
    msg = f"代理更改为了{proxy}"
    logging.info(msg)
    return msg


def hide_middle_chars(s):
    if s is None:
        return ""
    if len(s) <= 8:
        return s
    else:
        head = s[:4]
        tail = s[-4:]
        hidden = "*" * (len(s) - 8)
        return head + hidden + tail


def submit_key(key):
    key = key.strip()
    msg = f"API密钥更改为了{hide_middle_chars(key)}"
    logging.info(msg)
    return key, msg


def replace_today(prompt):
    today = datetime.datetime.today().strftime("%Y-%m-%d")
    return prompt.replace("{current_date}", today)


def get_geoip():
    try:
        with retrieve_proxy():
            response = requests.get("https://ipapi.co/json/", timeout=5)
        data = response.json()
    except:
        data = {"error": True, "reason": "连接ipapi失败"}
    if "error" in data.keys():
        logging.warning(f"无法获取IP地址信息。\n{data}")
        if data["reason"] == "RateLimited":
            return (
                i18n("您的IP区域:未知。")
            )
        else:
            return i18n("获取IP地理位置失败。原因:") + f"{data['reason']}" + i18n("。你仍然可以使用聊天功能。")
    else:
        country = data["country_name"]
        if country == "China":
            text = "**您的IP区域:中国。请立即检查代理设置,在不受支持的地区使用API可能导致账号被封禁。**"
        else:
            text = i18n("您的IP区域:") + f"{country}。"
        logging.info(text)
        return text


def find_n(lst, max_num):
    n = len(lst)
    total = sum(lst)

    if total < max_num:
        return n

    for i in range(len(lst)):
        if total - lst[i] < max_num:
            return n - i - 1
        total = total - lst[i]
    return 1


def start_outputing():
    logging.debug("显示取消按钮,隐藏发送按钮")
    return gr.Button.update(visible=False), gr.Button.update(visible=True)


def end_outputing():
    return (
        gr.Button.update(visible=True),
        gr.Button.update(visible=False),
    )


def cancel_outputing():
    logging.info("中止输出……")
    shared.state.interrupt()


def transfer_input(inputs):
    # 一次性返回,降低延迟
    textbox = reset_textbox()
    outputing = start_outputing()
    return (
        inputs,
        gr.update(value=""),
        gr.Button.update(visible=False),
        gr.Button.update(visible=True),
    )



def run(command, desc=None, errdesc=None, custom_env=None, live=False):
    if desc is not None:
        print(desc)
    if live:
        result = subprocess.run(command, shell=True, env=os.environ if custom_env is None else custom_env)
        if result.returncode != 0:
            raise RuntimeError(f"""{errdesc or 'Error running command'}.
                Command: {command}
                Error code: {result.returncode}""")

        return ""
    result = subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True, env=os.environ if custom_env is None else custom_env)
    if result.returncode != 0:
        message = f"""{errdesc or 'Error running command'}.
            Command: {command}
            Error code: {result.returncode}
            stdout: {result.stdout.decode(encoding="utf8", errors="ignore") if len(result.stdout)>0 else '<empty>'}
            stderr: {result.stderr.decode(encoding="utf8", errors="ignore") if len(result.stderr)>0 else '<empty>'}
            """
        raise RuntimeError(message)
    return result.stdout.decode(encoding="utf8", errors="ignore")

def versions_html():
    git = os.environ.get('GIT', "git")
    python_version = ".".join([str(x) for x in sys.version_info[0:3]])
    try:
        commit_hash = run(f"{git} rev-parse HEAD").strip()
    except Exception:
        commit_hash = "<none>"
    if commit_hash != "<none>":
        short_commit = commit_hash[0:7]
        commit_info = f"<a style=\"text-decoration:none;color:inherit\" href=\"https://github.com/GaiZhenbiao/ChuanhuChatGPT/commit/{short_commit}\">{short_commit}</a>"
    else:
        commit_info = "unknown \U0001F615"
    return f"""
        Python: <span title="{sys.version}">{python_version}</span>
         • 
        Gradio: {gr.__version__}
         • 
        <a style="text-decoration:none;color:inherit" href="https://github.com/GaiZhenbiao/ChuanhuChatGPT">ChuanhuChat</a>: {commit_info}
        """

def add_source_numbers(lst, source_name = "Source", use_source = True):
    if use_source:
        return [f'[{idx+1}]\t "{item[0]}"\n{source_name}: {item[1]}' for idx, item in enumerate(lst)]
    else:
        return [f'[{idx+1}]\t "{item}"' for idx, item in enumerate(lst)]

def add_details(lst):
    nodes = []
    for index, txt in enumerate(lst):
        brief = txt[:25].replace("\n", "")
        nodes.append(
            f"<details><summary>{brief}...</summary><p>{txt}</p></details>"
        )
    return nodes


def sheet_to_string(sheet, sheet_name = None):
    result = []
    for index, row in sheet.iterrows():
        row_string = ""
        for column in sheet.columns:
            row_string += f"{column}: {row[column]}, "
        row_string = row_string.rstrip(", ")
        row_string += "."
        result.append(row_string)
    return result

def excel_to_string(file_path):
    # 读取Excel文件中的所有工作表
    excel_file = pd.read_excel(file_path, engine='openpyxl', sheet_name=None)

    # 初始化结果字符串
    result = []

    # 遍历每一个工作表
    for sheet_name, sheet_data in excel_file.items():

        # 处理当前工作表并添加到结果字符串
        result += sheet_to_string(sheet_data, sheet_name=sheet_name)


    return result

def get_last_day_of_month(any_day):
    # The day 28 exists in every month. 4 days later, it's always next month
    next_month = any_day.replace(day=28) + datetime.timedelta(days=4)
    # subtracting the number of the current day brings us back one month
    return next_month - datetime.timedelta(days=next_month.day)

def get_model_source(model_name, alternative_source):
    if model_name == "gpt2-medium":
        return "https://huggingface.co/gpt2-medium"

def refresh_ui_elements_on_load(current_model, selected_model_name, user_name):
    current_model.set_user_identifier(user_name)
    return toggle_like_btn_visibility(selected_model_name), *current_model.auto_load()

def toggle_like_btn_visibility(selected_model_name):
    if selected_model_name == "xmchat":
        return gr.update(visible=True)
    else:
        return gr.update(visible=False)

def new_auto_history_filename(dirname):
    latest_file = get_latest_filepath(dirname)
    if latest_file:
        with open(os.path.join(dirname, latest_file), 'r') as f:
            if len(f.read()) == 0:
                return latest_file
    now = datetime.datetime.now().strftime('%Y-%m-%d_%H-%M-%S')
    return f'{now}.json'

def get_latest_filepath(dirname):
    pattern = re.compile(r'\d{4}-\d{2}-\d{2}_\d{2}-\d{2}-\d{2}')
    latest_time = None
    latest_file = None
    for filename in os.listdir(dirname):
        if os.path.isfile(os.path.join(dirname, filename)):
            match = pattern.search(filename)
            if match and match.group(0) == filename[:19]:
                time_str = filename[:19]
                filetime = datetime.datetime.strptime(time_str, '%Y-%m-%d_%H-%M-%S')
                if not latest_time or filetime > latest_time:
                    latest_time = filetime
                    latest_file = filename
    return latest_file

def get_history_filepath(username):
    dirname = os.path.join(HISTORY_DIR, username)
    os.makedirs(dirname, exist_ok=True)
    latest_file = get_latest_filepath(dirname)
    if not latest_file:
        latest_file = new_auto_history_filename(dirname)

    latest_file = os.path.join(dirname, latest_file)
    return latest_file