修复已知bug和优化使用体验
This commit is contained in:
Eric-Terminal
52
web_app.py
52
web_app.py
@@ -51,6 +51,9 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
|
||||
app = Flask(__name__)
|
||||
app.config["MAX_CONTENT_LENGTH"] = 100 * 1024 * 1024 # 100 MB payload ceiling
|
||||
app.logger.handlers.clear()
|
||||
app.logger.propagate = True
|
||||
logger = logging.getLogger("essay_corrector.web")
|
||||
|
||||
api_service = ApiService(config_manager)
|
||||
config_lock = threading.Lock()
|
||||
@@ -156,7 +159,7 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
config_manager.save()
|
||||
|
||||
except Exception as exc: # pylint: disable=broad-except
|
||||
logging.exception("文件处理失败: %s", saved_path)
|
||||
logger.exception("文件处理失败: %s", saved_path)
|
||||
error = str(exc)
|
||||
logs.append(f"处理失败: {error}")
|
||||
|
||||
@@ -253,14 +256,17 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
"VlmApiKey": "",
|
||||
"HasVlmApiKey": has_vlm_key,
|
||||
"VlmModel": config_manager.get("VlmModel", ""),
|
||||
"VlmTemperature": config_manager.get("VlmTemperature", 0.0),
|
||||
"LlmUrl": config_manager.get("LlmUrl", ""),
|
||||
"LlmApiKey": "",
|
||||
"HasLlmApiKey": has_llm_key,
|
||||
"LlmModel": config_manager.get("LlmModel", ""),
|
||||
"LlmTemperature": config_manager.get("LlmTemperature", 0.0),
|
||||
"SensitivityFactor": config_manager.get("SensitivityFactor", "1.0"),
|
||||
"MaxWorkers": config_manager.get("MaxWorkers", 4),
|
||||
"MaxRetries": config_manager.get("MaxRetries", 3),
|
||||
"RetryDelay": config_manager.get("RetryDelay", 5),
|
||||
"RequestTimeout": config_manager.get("RequestTimeout", 120),
|
||||
"SaveMarkdown": _as_bool(config_manager.get("SaveMarkdown", True), True),
|
||||
"RenderMarkdown": _as_bool(config_manager.get("RenderMarkdown", True), True),
|
||||
"AutoUpdateCheck": _as_bool(config_manager.get("AutoUpdateCheck", True), True),
|
||||
@@ -289,6 +295,11 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
"LlmApiKey",
|
||||
]
|
||||
int_fields = ["MaxWorkers", "MaxRetries", "RetryDelay"]
|
||||
float_fields = {
|
||||
"RequestTimeout": (1.0, None),
|
||||
"VlmTemperature": (0.0, 2.0),
|
||||
"LlmTemperature": (0.0, 2.0),
|
||||
}
|
||||
bool_fields = ["SaveMarkdown", "RenderMarkdown", "AutoUpdateCheck"]
|
||||
|
||||
updates: Dict[str, Any] = {}
|
||||
@@ -318,6 +329,19 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
except (TypeError, ValueError):
|
||||
return jsonify({"error": f"{key} 需要是整数"}), 400
|
||||
|
||||
for key, bounds in float_fields.items():
|
||||
if key in payload and payload[key] not in (None, ""):
|
||||
try:
|
||||
value = float(payload[key])
|
||||
except (TypeError, ValueError):
|
||||
return jsonify({"error": f"{key} 需要是数字"}), 400
|
||||
min_val, max_val = bounds
|
||||
if min_val is not None and value < min_val:
|
||||
return jsonify({"error": f"{key} 不能小于 {min_val}"}), 400
|
||||
if max_val is not None and value > max_val:
|
||||
return jsonify({"error": f"{key} 不能大于 {max_val}"}), 400
|
||||
updates[key] = value
|
||||
|
||||
if "SensitivityFactor" in payload and payload["SensitivityFactor"] not in (None, ""):
|
||||
try:
|
||||
updates["SensitivityFactor"] = float(payload["SensitivityFactor"])
|
||||
@@ -854,6 +878,9 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
<label>VLM 模型
|
||||
<input type="text" name="vlm_model" autocomplete="off" />
|
||||
</label>
|
||||
<label>VLM 温度 (0-2)
|
||||
<input type="number" name="vlm_temperature" min="0" max="2" step="0.1" />
|
||||
</label>
|
||||
<label>LLM URL
|
||||
<input type="text" name="llm_url" autocomplete="off" />
|
||||
</label>
|
||||
@@ -863,6 +890,9 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
<label>LLM 模型
|
||||
<input type="text" name="llm_model" autocomplete="off" />
|
||||
</label>
|
||||
<label>LLM 温度 (0-2)
|
||||
<input type="number" name="llm_temperature" min="0" max="2" step="0.1" />
|
||||
</label>
|
||||
<label>手写敏感度 (建议 1.0)
|
||||
<input type="text" name="sensitivity_factor" autocomplete="off" />
|
||||
</label>
|
||||
@@ -875,6 +905,9 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
<label>重试延迟 (秒)
|
||||
<input type="number" name="retry_delay" min="1" />
|
||||
</label>
|
||||
<label>请求超时时间 (秒)
|
||||
<input type="number" name="request_timeout" min="1" step="1" />
|
||||
</label>
|
||||
<label>输出目录
|
||||
<input type="text" name="output_directory" autocomplete="off" />
|
||||
</label>
|
||||
@@ -898,6 +931,17 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
<div class="view" data-view-section="about">
|
||||
<div class="section">
|
||||
<h2>关于与更新</h2>
|
||||
<div class="about-card">
|
||||
<h3>AI 作文批改助手</h3>
|
||||
<p>一款专注于英语作文批改的 Web 应用，整合视觉语言模型（VLM）与大语言模型（LLM），帮助教师与学生高效获得结构化反馈。</p>
|
||||
<ul>
|
||||
<li>两阶段流水线：先识别手写文本与书写分，再生成全中文批改报告。</li>
|
||||
<li>任务分离：所有图片自动归档到独立 run id，方便回溯、分享与比对。</li>
|
||||
<li>Prompt 可编辑：浏览器内直接替换评分模板，快速适配不同考试场景。</li>
|
||||
<li>安全可控：API Key 本地加密保存，Token 用量实时累计并在界面呈现。</li>
|
||||
</ul>
|
||||
<p class="muted">作者：Eric_Terminal · 项目主页：<a href="https://github.com/Eric-Terminal/Pro_llm_correct" target="_blank" rel="noopener">GitHub</a></p>
|
||||
</div>
|
||||
<div class="about-card">
|
||||
<div><strong>当前版本：</strong><span id="about-current">{{ current_version }}</span></div>
|
||||
<div id="about-latest">正在获取最新版本信息...</div>
|
||||
@@ -987,13 +1031,16 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
settingsForm.vlm_url.value = data.VlmUrl || '';
|
||||
settingsForm.vlm_api_key.value = '';
|
||||
settingsForm.vlm_model.value = data.VlmModel || '';
|
||||
settingsForm.vlm_temperature.value = data.VlmTemperature ?? 0;
|
||||
settingsForm.llm_url.value = data.LlmUrl || '';
|
||||
settingsForm.llm_api_key.value = '';
|
||||
settingsForm.llm_model.value = data.LlmModel || '';
|
||||
settingsForm.llm_temperature.value = data.LlmTemperature ?? 0;
|
||||
settingsForm.sensitivity_factor.value = data.SensitivityFactor || '';
|
||||
settingsForm.max_workers.value = data.MaxWorkers || 4;
|
||||
settingsForm.max_retries.value = data.MaxRetries || 3;
|
||||
settingsForm.retry_delay.value = data.RetryDelay || 5;
|
||||
settingsForm.request_timeout.value = data.RequestTimeout ?? 120;
|
||||
settingsForm.output_directory.value = data.OutputDirectory || '{{ default_output_dir }}';
|
||||
settingsForm.save_markdown.checked = !!data.SaveMarkdown;
|
||||
settingsForm.render_markdown.checked = !!data.RenderMarkdown;
|
||||
@@ -1051,13 +1098,16 @@ def create_app(config_manager: ConfigManager) -> Flask:
|
||||
VlmUrl: settingsForm.vlm_url.value.trim(),
|
||||
VlmApiKey: settingsForm.vlm_api_key.value.trim(),
|
||||
VlmModel: settingsForm.vlm_model.value.trim(),
|
||||
VlmTemperature: settingsForm.vlm_temperature.value,
|
||||
LlmUrl: settingsForm.llm_url.value.trim(),
|
||||
LlmApiKey: settingsForm.llm_api_key.value.trim(),
|
||||
LlmModel: settingsForm.llm_model.value.trim(),
|
||||
LlmTemperature: settingsForm.llm_temperature.value,
|
||||
SensitivityFactor: settingsForm.sensitivity_factor.value.trim(),
|
||||
MaxWorkers: settingsForm.max_workers.value,
|
||||
MaxRetries: settingsForm.max_retries.value,
|
||||
RetryDelay: settingsForm.retry_delay.value,
|
||||
RequestTimeout: settingsForm.request_timeout.value,
|
||||
OutputDirectory: settingsForm.output_directory.value.trim(),
|
||||
SaveMarkdown: settingsForm.save_markdown.checked,
|
||||
RenderMarkdown: settingsForm.render_markdown.checked,
|
||||
|
||||
Reference in New Issue
Block a user