增加绕过tts的设置和设置菜单 - #11
Open
huiyuanXP wants to merge 7 commits into
Open
Conversation
|
推，實用 |
Author
|
使软件可以解析任意格式的pdf文件并完成翻译和rag构建 任意格式解析模式删除了查找作者,引用等论文独有格式的代码,使整体不会出现解析错误的问题 |
This was referenced Apr 26, 2025
- config.py: API密钥从硬编码改为app_settings.json文件读取 - settings_bar.py: 新增API设置对话框 - TTS_manager.py: 适配新配置读取方式 - rag_processor.py: 修复langchain import路径 - .gitignore: 忽略app_settings.json和data/output目录 Made-with: Cursor
引用跳转功能： - rag_retriever: 返回citations列表，context段落带编号 - AI_professor_chat: RAG检索时在user message注入引用格式要求 - threads/AI_manager: 信号链路传递citations列表 - message_bubble: AI消息中 *ref-[N] 渲染为可点击citation pill - chat_widget: 点击pill跳转到文章对应位置 导师风格切换： - AI_professor_chat: 支持多角色prompt切换和清空对话 - chat_widget: 标题栏增加风格切换菜单 - AI_manager: 转发set_character/clear_conversation
Made-with: Cursor
jialuncheng
added a commit
to jialuncheng/mad-professor-public
that referenced
this pull request
May 21, 2026
依 v2 報告 §4.1 25 狀況決策樹 + §3.1 黑名單 + §5 domain 仲裁 + §6 相似度。 md_restore_processor.process 不再單看 JSON tree data['title']，改為融合： - raw (md_processor first-#) - metadata.title (含 .source: pdf_metadata/llm_page1/both_agree/...) - domain (Commit 0 後從 metadata.domain 經 pipeline_core 取) - doc_type (resume 走 candidate_name 短路) 決策核心： - resume + candidate_name → 短路（LYiHub#1） - both_agree → metadata 勝（LYiHub#9，雙路確認） - llm_page1 + sim → metadata（LYiHub#11，精確版本） - llm_page1 + 不 sim + raw 黑名單 → metadata（LYiHub#12，raw 雜訊） - llm_page1 + 不 sim + raw 合法 → domain 仲裁（LYiHub#13） - pdf_metadata + 不 sim + raw 合法 → raw 勝（LYiHub#15，baron 核心觀察： 99% PDF metadata 是「人沒改」的垃圾預設值） - llm_page1 (conflict ...) → 同 LYiHub#11-LYiHub#13 但低信心 - 全空 → fallback original_filename(去 .pdf) → paper_uuid processor/md_restore_processor.py: - 新增模組級 TITLE_BLACKLIST_EXACT (29 entries) + TITLE_BLACKLIST_PATTERN (8 entries) - 新增 _title_in_blacklist / _title_sim / _tokenize_for_domain / _dom_match - 新增 _resolve_title 主融合邏輯（涵蓋 25 狀況決策樹） - process() 簽名加 metadata/doc_type/domain/original_filename/paper_uuid（皆有 預設值，舊 caller 不傳則退回 raw-only 行為） - 寫 # title 前先呼叫 _resolve_title 取得融合結果 + log 信心度 pipeline_core.py: - process() 將 original_filename 存到 self._original_filename - _stage_md_restore 傳 metadata / doc_type / domain / original_filename / paper_uuid 給 RestoreProcessor tests/test_md_restore_processor.py（新）：13 個融合決策測試 - _title_in_blacklist 4 個 - _title_sim 1 個 - _dom_match 2 個 - _resolve_title 主決策樹 6 個（resume 短路、黑名單、both_agree、 pdf_metadata LYiHub#15、domain 仲裁 3 case、全空 fallback） 注意： - 不動 md_processor.parse() 抓 first-# 策略（仍是 raw 資料源） - 不動 sections 寫入流（原 raw data['title'] 已被 md_processor 抽出、 不在 sections 內、無內文重複問題） - 不重抽既有 paper，只對新上傳生效；既有 final_*_zh.md 不重生 效果預期（新上傳）： - 800-vdc (technical, raw='Contents'): metadata 勝（黑名單） - 1763... (slides, raw='第 1 頁'): metadata 勝（pattern 黑名單） - DeHunt (resume, candidate_name='DeHunt'): 短路取候選人姓名 - academic 正常: both_agree 或 llm sim raw → metadata py_compile 通過；pytest 39 passed 3 skipped；check_doc_type_registry exit 0。
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
理由:
当前的tts十分不稳定
仍建议打开AI_manager 228行注释掉的代码:
# self.ai_sentence_ready.emit(sentence, self.current_request_id)避免回答不显示