Update beschreibungen.py

beschreibungen.py

@@ -1,232 +1,82 @@
-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 markdown2
-import torch 
-import sys
-import gc
-from pygments.lexers import guess_lexer, ClassNotFound
-
 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 guess_lexer,get_lexer_by_name
-from pygments.formatters import HtmlFormatter
-from beschreibungen import *
-
-logging.basicConfig(
-    level=logging.INFO,
-    format="%(asctime)s [%(levelname)s] [%(filename)s:%(lineno)d] %(message)s",
-)
-
-
-def markdown_to_html_with_syntax_highlight(md_str):
-    def replacer(match):
-        lang = match.group(1) or "text"
-        code = match.group(2)
-        lang = lang.strip()
-        #print(1,lang)
-        if lang=="text":
-            lexer = guess_lexer(code)
-            lang = lexer.name
-            #print(2,lang)
-        try:
-            lexer = get_lexer_by_name(lang, stripall=True)
-        except ValueError:
-            lexer = get_lexer_by_name("python", stripall=True)
-        formatter = HtmlFormatter()
-        #print(3,lexer.name)
-        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"```(.*?)(?:```|$)", re.DOTALL)
-    inline_code_pattern = re.compile(r"`(.*?)`", re.DOTALL)
-    code_blocks = code_block_pattern.findall(md_text)
-    non_code_parts = code_block_pattern.split(md_text)[::2]
-
-    result = []
-    for non_code, code in zip(non_code_parts, code_blocks + [""]):
-        if non_code.strip():
-            non_code = normalize_markdown(non_code)
-            if inline_code_pattern.search(non_code):
-                result.append(markdown(non_code, extensions=["tables"]))
-            else:
-                result.append(mdtex2html.convert(non_code, extensions=["tables"]))
-        if code.strip():
-            code = f"\n```{code}\n\n```"
-            code = markdown_to_html_with_syntax_highlight(code)
-            result.append(code)
-    result = "".join(result)
-    result += ALREADY_CONVERTED_MARK
-    return result
-
-def convert_asis(userinput):
-    return f"<p style=\"white-space:pre-wrap;\">{html.escape(userinput)}</p>"+ALREADY_CONVERTED_MARK
-
-def detect_converted_mark(userinput):
-    if userinput.endswith(ALREADY_CONVERTED_MARK):
-        return True
-    else:
-        return False
-
-
-
-def detect_language(code):
-    if code.startswith("\n"):
-        first_line = ""
-    else:
-        first_line = code.strip().split("\n", 1)[0]
-    language = first_line.lower() if first_line else ""
-    code_without_language = code[len(first_line) :].lstrip() if first_line else code
-    return language, code_without_language
-
-def convert_to_markdown(text):
-    text = text.replace("$","&#36;")
-    def replace_leading_tabs_and_spaces(line):
-        new_line = []
-        
-        for char in line:
-            if char == "\t":
-                new_line.append("&#9;")
-            elif char == " ":
-                new_line.append("&nbsp;")
-            else:
-                break
-        return "".join(new_line) + line[len(new_line):]
-
-    markdown_text = ""
-    lines = text.split("\n")
-    in_code_block = False
-
-    for line in lines:
-        if in_code_block is False and line.startswith("```"):
-            in_code_block = True
-            markdown_text += f"{line}\n"
-        elif in_code_block is True and line.startswith("```"):
-            in_code_block = False
-            markdown_text += f"{line}\n"
-        elif in_code_block:
-            markdown_text += f"{line}\n"
-        else:
-            line = replace_leading_tabs_and_spaces(line)
-            line = re.sub(r"^(#)", r"\\\1", line)
-            markdown_text += f"{line}  \n"
-
-    return markdown_text
-
-def add_language_tag(text):
-    def detect_language(code_block):
-        try:
-            lexer = guess_lexer(code_block)
-            return lexer.name.lower()
-        except ClassNotFound:
-            return ""
-
-    code_block_pattern = re.compile(r"(```)(\w*\n[^`]+```)", re.MULTILINE)
-
-    def replacement(match):
-        code_block = match.group(2)
-        if match.group(2).startswith("\n"):
-            language = detect_language(code_block)
-            if language:
-                return f"```{language}{code_block}```"
-            else:
-                return f"```\n{code_block}```"
-        else:
-            return match.group(1) + code_block + "```"
-
-    text2 = code_block_pattern.sub(replacement, text)
-    return text2
-
-def delete_last_conversation(chatbot, history):
-    if len(chatbot) > 0:
-        chatbot.pop()
-
-    if len(history) > 0:
-        history.pop()
-        
-    return (
-        chatbot,
-        history,
-        "Delete Done",
-    )
-
-def reset_state():
-    return [], [], "Reset Done"
-
-def reset_textbox():
-    return gr.update(value=""),""
-
-def cancel_outputing():
-    return "Stop Done"
-
-
-class State:
-    interrupted = False
-
-    def interrupt(self):
-        self.interrupted = True
-
-    def recover(self):
-        self.interrupted = False
-shared_state = State()
-
-
-
-
-
-
 
 
-def is_stop_word_or_prefix(s: str, stop_words: list) -> bool:
-    for stop_word in stop_words:
-        if s.endswith(stop_word):
-            return True
-        for i in range(1, len(stop_word)):
-            if s.endswith(stop_word[:i]):
-                return True
-    return False
+title = """<h1 align="left" style="min-width:200px; margin-top:0;"> Chat mit Huging Face Chatbots</h1>"""
+description_top = """\
+<div align="left">
+<p> Aktuelles Modell: Hugging Face Chatbots</p>
+<p>
+Disclaimer: Die hier verwendeten Modelle sind Modelle vom Hugging Face Hub. Kommerzieller Gebrauch ist ausdrücklich verboten. Der Output des Modells ist nicht zensiert und die Meiningen im Output entsprechend nicht unbedingt der des Authors dieser App. Gebrauch auf eigene Gefahr! 
+</p >
+</div>
+"""
+description = """\
+<div align="center" style="margin:16px 0">
+Hier werden verschiedene Modelle vom Hugging Face Hub verwendet.
+</div>
+"""
+CONCURRENT_COUNT = 100
+
+
+ALREADY_CONVERTED_MARK = "<!-- ALREADY CONVERTED BY PARSER. -->"
+
+small_and_beautiful_theme = gr.themes.Soft(
+        primary_hue=gr.themes.Color(
+            c50="#02C160",
+            c100="rgba(2, 193, 96, 0.2)",
+            c200="#02C160",
+            c300="rgba(2, 193, 96, 0.32)",
+            c400="rgba(2, 193, 96, 0.32)",
+            c500="rgba(2, 193, 96, 1.0)",
+            c600="rgba(2, 193, 96, 1.0)",
+            c700="rgba(2, 193, 96, 0.32)",
+            c800="rgba(2, 193, 96, 0.32)",
+            c900="#02C160",
+            c950="#02C160",
+        ),
+        secondary_hue=gr.themes.Color(
+            c50="#576b95",
+            c100="#576b95",
+            c200="#576b95",
+            c300="#576b95",
+            c400="#576b95",
+            c500="#576b95",
+            c600="#576b95",
+            c700="#576b95",
+            c800="#576b95",
+            c900="#576b95",
+            c950="#576b95",
+        ),
+        neutral_hue=gr.themes.Color(
+            name="gray",
+            c50="#f9fafb",
+            c100="#f3f4f6",
+            c200="#e5e7eb",
+            c300="#d1d5db",
+            c400="#B2B2B2",
+            c500="#808080",
+            c600="#636363",
+            c700="#515151",
+            c800="#393939",
+            c900="#272727",
+            c950="#171717",
+        ),
+        radius_size=gr.themes.sizes.radius_sm,
+    ).set(
+        button_primary_background_fill="#06AE56",
+        button_primary_background_fill_dark="#06AE56",
+        button_primary_background_fill_hover="#07C863",
+        button_primary_border_color="#06AE56",
+        button_primary_border_color_dark="#06AE56",
+        button_primary_text_color="#FFFFFF",
+        button_primary_text_color_dark="#FFFFFF",
+        button_secondary_background_fill="#F2F2F2",
+        button_secondary_background_fill_dark="#2B2B2B",
+        button_secondary_text_color="#393939",
+        button_secondary_text_color_dark="#FFFFFF",
+        # background_fill_primary="#F7F7F7",
+        # background_fill_primary_dark="#1F1F1F",
+        block_title_text_color="*primary_500",
+        block_title_background_fill="*primary_100",
+        input_background_fill="#F6F6F6",
+    )