optimize UI and user experience

app/config/config.py

@@ -1,6 +1,6 @@
 import os
 import socket
-import tomli
+import toml
 from loguru import logger
 
 root_dir = os.path.dirname(os.path.dirname(os.path.dirname(os.path.realpath(__file__))))
@@ -16,17 +16,17 @@ if not os.path.isfile(config_file):
 logger.info(f"load config from file: {config_file}")
 
 try:
-    with open(config_file, mode="rb") as fp:
+    _cfg = toml.load(config_file)
-        _cfg = tomli.load(fp)
 except Exception as e:
     logger.warning(f"load config failed: {str(e)}, try to load as utf-8-sig")
     with open(config_file, mode="r", encoding='utf-8-sig') as fp:
         _cfg_content = fp.read()
-        _cfg = tomli.loads(_cfg_content)
+        _cfg = toml.loads(_cfg_content)
 
 app = _cfg.get("app", {})
 whisper = _cfg.get("whisper", {})
 pexels = _cfg.get("pexels", {})
+ui = _cfg.get("ui", {})
 
 hostname = socket.gethostname()
 
@@ -47,9 +47,18 @@ ffmpeg_path = app.get("ffmpeg_path", "")
 if ffmpeg_path and os.path.isfile(ffmpeg_path):
     os.environ["IMAGEIO_FFMPEG_EXE"] = ffmpeg_path
 
+
 # __cfg = {
 #     "hostname": hostname,
 #     "listen_host": listen_host,
 #     "listen_port": listen_port,
 # }
 # logger.info(__cfg)
+
+
+def save_config():
+    with open(config_file, "w", encoding="utf-8") as f:
+        _cfg["app"] = app
+        _cfg["whisper"] = whisper
+        _cfg["pexels"] = pexels
+        f.write(toml.dumps(_cfg))

app/services/material.py

@@ -11,13 +11,14 @@ from app.models.schema import VideoAspect, VideoConcatMode, MaterialInfo
 from app.utils import utils
 
 requested_count = 0
-pexels_api_keys = config.app.get("pexels_api_keys")
-if not pexels_api_keys:
-    raise ValueError(
-        f"\n\n##### pexels_api_keys is not set #####\n\nPlease set it in the config.toml file: {config.config_file}\n\n{utils.to_json(config.app)}")
 
 
 def round_robin_api_key():
+    pexels_api_keys = config.app.get("pexels_api_keys")
+    if not pexels_api_keys:
+        raise ValueError(
+            f"\n\n##### pexels_api_keys is not set #####\n\nPlease set it in the config.toml file: {config.config_file}\n\n{utils.to_json(config.app)}")
+
     # if only one key is provided, return it
     if isinstance(pexels_api_keys, str):
         return pexels_api_keys

app/utils/utils.py

@@ -1,3 +1,4 @@
+import locale
 import os
 import platform
 import threading
@@ -174,3 +175,25 @@ def split_string_by_punctuations(s):
 def md5(text):
     import hashlib
     return hashlib.md5(text.encode('utf-8')).hexdigest()
+
+
+def get_system_locale():
+    try:
+        loc = locale.getdefaultlocale()
+        # zh_CN, zh_TW return zh
+        # en_US, en_GB return en
+        language_code = loc[0].split("_")[0]
+        return language_code
+    except Exception as e:
+        return "en"
+
+
+def load_locales(i18n_dir):
+    _locales = {}
+    for root, dirs, files in os.walk(i18n_dir):
+        for file in files:
+            if file.endswith(".json"):
+                lang = file.split(".")[0]
+                with open(os.path.join(root, file), "r", encoding="utf-8") as f:
+                    _locales[lang] = json.loads(f.read())
+    return _locales

webui.bat

@@ -1,4 +1,2 @@
-set CURRENT_DIR=%CD%
-set PYTHONPATH=%CURRENT_DIR%
 rem set HF_ENDPOINT=https://hf-mirror.com
-streamlit run .\webui\Main.py
+streamlit run .\webui\Main.py --browser.gatherUsageStats=False --server.enableCORS=True

webui.sh

@@ -1,7 +1,3 @@
-CURRENT_DIR=$(pwd)
-echo "***** Current directory: $CURRENT_DIR *****"
-export PYTHONPATH="${CURRENT_DIR}:$PYTHONPATH"
-
 # If you could not download the model from the official site, you can use the mirror site.
 # Just remove the comment of the following line .
 # 如果你无法从官方网站下载模型，你可以使用镜像网站。

webui/Main.py

@@ -9,15 +9,12 @@ if root_dir not in sys.path:
     print(sys.path)
     print("")
 
-import json
-import locale
 import streamlit as st
 
 import os
 from uuid import uuid4
 import platform
 import streamlit.components.v1 as components
-import toml
 from loguru import logger
 
 st.set_page_config(page_title="MoneyPrinterTurbo",
@@ -35,6 +32,7 @@ st.set_page_config(page_title="MoneyPrinterTurbo",
 from app.models.schema import VideoParams, VideoAspect, VideoConcatMode
 from app.services import task as tm, llm, voice
 from app.utils import utils
+from app.config import config
 
 hide_streamlit_style = """
 <style>#root > div:nth-child(1) > div > div > div > div > section > div {padding-top: 0rem;}</style>
@@ -46,33 +44,7 @@ font_dir = os.path.join(root_dir, "resource", "fonts")
 song_dir = os.path.join(root_dir, "resource", "songs")
 i18n_dir = os.path.join(root_dir, "webui", "i18n")
 config_file = os.path.join(root_dir, "webui", ".streamlit", "webui.toml")
-
+system_locale = utils.get_system_locale()
-
-def load_config() -> dict:
-    try:
-        return toml.load(config_file)
-    except Exception as e:
-        return {}
-
-
-cfg = load_config()
-
-
-def save_config():
-    with open(config_file, "w", encoding="utf-8") as f:
-        f.write(toml.dumps(cfg))
-
-
-def get_system_locale():
-    try:
-        loc = locale.getdefaultlocale()
-        # zh_CN, zh_TW return zh
-        # en_US, en_GB return en
-        language_code = loc[0].split("_")[0]
-        return language_code
-    except Exception as e:
-        return "en"
-
 
 if 'video_subject' not in st.session_state:
     st.session_state['video_subject'] = ''
@@ -81,7 +53,7 @@ if 'video_script' not in st.session_state:
 if 'video_terms' not in st.session_state:
     st.session_state['video_terms'] = ''
 if 'ui_language' not in st.session_state:
-    st.session_state['ui_language'] = cfg.get("ui_language", get_system_locale())
+    st.session_state['ui_language'] = config.ui.get("language", system_locale)
 
 
 def get_all_fonts():
@@ -163,19 +135,7 @@ def init_log():
 
 init_log()
 
-
+locales = utils.load_locales(i18n_dir)
-def load_locales():
-    locales = {}
-    for root, dirs, files in os.walk(i18n_dir):
-        for file in files:
-            if file.endswith(".json"):
-                lang = file.split(".")[0]
-                with open(os.path.join(root, file), "r", encoding="utf-8") as f:
-                    locales[lang] = json.loads(f.read())
-    return locales
-
-
-locales = load_locales()
 
 
 def tr(key):
@@ -183,20 +143,70 @@ def tr(key):
     return loc.get("Translation", {}).get(key, key)
 
 
-display_languages = []
+with st.expander(tr("Basic Settings"), expanded=True):
-selected_index = 0
+    config_panels = st.columns(3)
-for i, code in enumerate(locales.keys()):
+    left_config_panel = config_panels[0]
+    middle_config_panel = config_panels[1]
+    right_config_panel = config_panels[2]
+    with left_config_panel:
+        display_languages = []
+        selected_index = 0
+        for i, code in enumerate(locales.keys()):
             display_languages.append(f"{code} - {locales[code].get('Language')}")
             if code == st.session_state['ui_language']:
                 selected_index = i
 
-selected_language = st.selectbox("Language", options=display_languages, label_visibility='collapsed',
+        selected_language = st.selectbox(tr("Language"), options=display_languages,
                                          index=selected_index)
-if selected_language:
+        if selected_language:
             code = selected_language.split(" - ")[0].strip()
             st.session_state['ui_language'] = code
-    cfg['ui_language'] = code
+            config.ui['language'] = code
-    save_config()
+            config.save_config()
+
+    with middle_config_panel:
+        #   openai
+        #   moonshot (月之暗面)
+        #   oneapi
+        #   g4f
+        #   azure
+        #   qwen (通义千问)
+        #   gemini
+        #   ollama
+        llm_providers = ['OpenAI', 'Moonshot', 'Azure', 'Qwen', 'Gemini', 'Ollama', 'G4f', 'OneAPI']
+        saved_llm_provider = config.app.get("llm_provider", "OpenAI").lower()
+        saved_llm_provider_index = 0
+        for i, provider in enumerate(llm_providers):
+            if provider.lower() == saved_llm_provider:
+                saved_llm_provider_index = i
+                break
+
+        llm_provider = st.selectbox(tr("LLM Provider"), options=llm_providers, index=saved_llm_provider_index)
+        llm_provider = llm_provider.lower()
+        config.app["llm_provider"] = llm_provider
+        st.write(f"**{tr('LLM Provider')}:** {llm_provider}")
+
+        llm_api_key = config.app.get(f"{llm_provider}_api_key", "")
+        llm_base_url = config.app.get(f"{llm_provider}_base_url", "")
+        llm_model_name = config.app.get(f"{llm_provider}_model_name", "")
+        st_llm_api_key = st.text_input(tr("API Key"), value=llm_api_key, type="password")
+        st_llm_base_url = st.text_input(tr("Base Url"), value=llm_base_url)
+        st_llm_model_name = st.text_input(tr("Model Name"), value=llm_model_name)
+        if st_llm_api_key:
+            config.app[f"{llm_provider}_api_key"] = st_llm_api_key
+        if st_llm_base_url:
+            config.app[f"{llm_provider}_base_url"] = st_llm_base_url
+        if st_llm_model_name:
+            config.app[f"{llm_provider}_model_name"] = st_llm_model_name
+
+        config.save_config()
+
+    with right_config_panel:
+        pexels_api_key = config.app.get("pexels_api_keys", "")
+        pexels_api_key = st.text_input(tr("Pexels API Key"), value=pexels_api_key, type="password")
+        if pexels_api_key:
+            config.app["pexels_api_keys"] = pexels_api_key
+            config.save_config()
 
 panel = st.columns(3)
 left_panel = panel[0]
@@ -286,7 +296,7 @@ with middle_panel:
             replace("Male", tr("Male")).
             replace("Neural", "") for
             voice in voices}
-        saved_voice_name = cfg.get("voice_name", "")
+        saved_voice_name = config.ui.get("voice_name", "")
         saved_voice_name_index = 0
         if saved_voice_name in friendly_names:
             saved_voice_name_index = list(friendly_names.keys()).index(saved_voice_name)
@@ -302,8 +312,8 @@ with middle_panel:
 
         voice_name = list(friendly_names.keys())[list(friendly_names.values()).index(selected_friendly_name)]
         params.voice_name = voice_name
-        cfg['voice_name'] = voice_name
+        config.ui['voice_name'] = voice_name
-        save_config()
+        config.save_config()
 
         params.voice_volume = st.selectbox(tr("Speech Volume"),
                                            options=[0.6, 0.8, 1.0, 1.2, 1.5, 2.0, 3.0, 4.0, 5.0], index=2)
@@ -334,7 +344,13 @@ with right_panel:
         st.write(tr("Subtitle Settings"))
         params.subtitle_enabled = st.checkbox(tr("Enable Subtitles"), value=True)
         font_names = get_all_fonts()
-        params.font_name = st.selectbox(tr("Font"), font_names)
+        saved_font_name = config.ui.get("font_name", "")
+        saved_font_name_index = 0
+        if saved_font_name in font_names:
+            saved_font_name_index = font_names.index(saved_font_name)
+        params.font_name = st.selectbox(tr("Font"), font_names, index=saved_font_name_index)
+        config.ui['font_name'] = params.font_name
+        config.save_config()
 
         subtitle_positions = [
             (tr("Top"), "top"),
@@ -350,9 +366,14 @@ with right_panel:
 
         font_cols = st.columns([0.3, 0.7])
         with font_cols[0]:
-            params.text_fore_color = st.color_picker(tr("Font Color"), "#FFFFFF")
+            saved_text_fore_color = config.ui.get("text_fore_color", "#FFFFFF")
+            params.text_fore_color = st.color_picker(tr("Font Color"), saved_text_fore_color)
+            config.ui['text_fore_color'] = params.text_fore_color
+
         with font_cols[1]:
-            params.font_size = st.slider(tr("Font Size"), 30, 100, 60)
+            saved_font_size = config.ui.get("font_size", 60)
+            params.font_size = st.slider(tr("Font Size"), 30, 100, saved_font_size)
+            config.ui['font_size'] = params.font_size
 
         stroke_cols = st.columns([0.3, 0.7])
         with stroke_cols[0]:
@@ -362,12 +383,23 @@ with right_panel:
 
 start_button = st.button(tr("Generate Video"), use_container_width=True, type="primary")
 if start_button:
+    config.save_config()
     task_id = str(uuid4())
     if not params.video_subject and not params.video_script:
         st.error(tr("Video Script and Subject Cannot Both Be Empty"))
         scroll_to_bottom()
         st.stop()
 
+    if not config.app.get(f"{llm_provider}_api_key", ""):
+        st.error(tr("Please Enter the LLM API Key"))
+        scroll_to_bottom()
+        st.stop()
+
+    if not config.app.get("pexels_api_keys", ""):
+        st.error(tr("Please Enter the Pexels API Key"))
+        scroll_to_bottom()
+        st.stop()
+
     log_container = st.empty()
     log_records = []
 

webui/i18n/de.json

@@ -48,6 +48,15 @@
     "Generating Video": "Video wird erstellt, bitte warten...",
     "Start Generating Video": "Beginne mit der Generierung",
     "Video Generation Completed": "Video erfolgreich generiert",
-    "You can download the generated video from the following links": "Sie können das generierte Video über die folgenden Links herunterladen"
+    "You can download the generated video from the following links": "Sie können das generierte Video über die folgenden Links herunterladen",
+    "Basic Settings": "**Grunde Instellungen**",
+    "Pexels API Key": "Pexels API Key (:red[Required] [Get API Key](https://www.pexels.com/api/))",
+    "Language": "Language",
+    "LLM Provider": "LLM Provider",
+    "API Key": "API Key (:red[Required])",
+    "Base Url": "Base Url",
+    "Model Name": "Model Name",
+    "Please Enter the LLM API Key": "Please Enter the **LLM API Key**",
+    "Please Enter the Pexels API Key": "Please Enter the **Pexels API Key**"
   }
 }

webui/i18n/en.json

@@ -48,6 +48,15 @@
     "Generating Video": "Generating video, please wait...",
     "Start Generating Video": "Start Generating Video",
     "Video Generation Completed": "Video Generation Completed",
-    "You can download the generated video from the following links": "You can download the generated video from the following links"
+    "You can download the generated video from the following links": "You can download the generated video from the following links",
+    "Pexels API Key": "Pexels API Key (:red[Required] [Get API Key](https://www.pexels.com/api/))",
+    "Basic Settings": "**Basic Settings** (:blue[Click to expand])",
+    "Language": "Language",
+    "LLM Provider": "LLM Provider",
+    "API Key": "API Key (:red[Required])",
+    "Base Url": "Base Url",
+    "Model Name": "Model Name",
+    "Please Enter the LLM API Key": "Please Enter the **LLM API Key**",
+    "Please Enter the Pexels API Key": "Please Enter the **Pexels API Key**"
   }
 }

webui/i18n/zh.json

@@ -48,6 +48,15 @@
     "Generating Video": "正在生成视频，请稍候...",
     "Start Generating Video": "开始生成视频",
     "Video Generation Completed": "视频生成完成",
-    "You can download the generated video from the following links": "你可以从以下链接下载生成的视频"
+    "You can download the generated video from the following links": "你可以从以下链接下载生成的视频",
+    "Basic Settings": "**基础设置** (:blue[点击展开])",
+    "Language": "界面语言",
+    "Pexels API Key": "Pexels API Key (:red[必填] [点击获取](https://www.pexels.com/api/))",
+    "LLM Provider": "大模型提供商",
+    "API Key": "API Key (:red[必填，需要到大模型提供商的后台申请])",
+    "Base Url": "Base Url (可选)",
+    "Model Name": "模型名称 (:blue[需要到大模型提供商的后台确认被授权的模型名称])",
+    "Please Enter the LLM API Key": "请先填写大模型 **API Key**",
+    "Please Enter the Pexels API Key": "请先填写 **Pexels API Key**"
   }
 }