PyPI - realtimex-deeptutor - Versions diffs - 0.5.0.post1__tar.gz → 0.5.0.post2__tar.gz - Mend

realtimex-deeptutor 0.5.0.post1tar.gz → 0.5.0.post2tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (297) hide show

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: realtimex-deeptutor
-Version: 0.5.0.post1
+Version: 0.5.0.post2
 Summary: RealTimeX DeepTutor - Intelligent learning companion with multi-agent collaboration and LightRAG
 License: Apache-2.0
 Requires-Python: >=3.10
@@ -26,6 +26,7 @@ Requires-Dist: lightrag-hku>=1.0.0
 Requires-Dist: PyMuPDF>=1.26.0
 Requires-Dist: numpy<2.0.0,>=1.24.0
 Requires-Dist: arxiv>=2.0.0
+Requires-Dist: realtimex-sdk>=1.1.2
 Provides-Extra: anthropic
 Requires-Dist: anthropic>=0.30.0; extra == "anthropic"
 Provides-Extra: dashscope
@@ -90,8 +91,6 @@ Dynamic: license-file
 ---
 ### 📰 News
-> **[2026.1.15]** DeepTutor [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) is out! Fixed multiple environment configuration and stability issues. We recommend everyone to pull the latest version! 🎉
 > **[2026.1.1]** Happy New Year! Join our [Discord Community](https://discord.gg/zpP9cssj), [Wechat Community](https://github.com/HKUDS/DeepTutor/issues/78), or [Discussions](https://github.com/HKUDS/DeepTutor/discussions) - shape the future of DeepTutor! 💬
 > **[2025.12.30]** Visit our [Official Website](https://hkuds.github.io/DeepTutor/) for more details!
@@ -100,10 +99,16 @@ Dynamic: license-file
 ### 📦 Releases
-> **[2026.1.15]** Release [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) - Unified LLM & Embedding services, RAG pipeline selection, and major enhancements to Home, History, QuestionGen & Settings modules -- Thanks to all the contributors!
+> **[2026.1.23]** Release [v0.6.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.6.0) - Frontend session persistence, full Chinese support, Docker deployment updates, and minor bug fixes -- Thanks for all the feedback!
 <details>
 <summary>History releases</summary>
+> **[2026.1.18]** Release [v0.5.2](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.1) - Enhance RAG pipeline with Docling support and improve CI/CD workflows with several minor bugs fixed -- Thanks to all the feedbacks!
+> **[2026.1.15]** Release [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) - Unified LLM & Embedding services, RAG pipeline selection, and major enhancements to Home, History, QuestionGen & Settings modules -- Thanks to all the contributors!
 > **[2026.1.9]** Release [v0.4.1](https://github.com/HKUDS/DeepTutor/releases/tag/v0.4.1) with LLM Provider system overhaul, Question Generation robustness improvements, and codebase cleanup - Thanks to all the contributors!
 > **[2026.1.9]** Release [v0.4.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.4.0) with new code structure, multiple llm & embeddings support - Thanks to all the contributors!
@@ -390,14 +395,14 @@ cp .env.example .env
 **Quick Start** — Build from source:
 ```bash
-docker compose up --build -d    # Build and start (~5-10 min first run)
-docker compose logs -f          # View logs
+docker compose up                  # Build and start (~11 min first run on mac mini M4)
+docker compose build --no-cache    # Clear cache and rebuild after pull the newest repo
 ```
 **Or use pre-built image** (faster):
 ```bash
-# Linux/macOS (AMD64)
+# Works on all platforms - Docker auto-detects your architecture
 docker run -d --name deeptutor \
   -p 8001:8001 -p 3782:3782 \
   --env-file .env \
@@ -405,7 +410,6 @@ docker run -d --name deeptutor \
   -v $(pwd)/config:/app/config:ro \
   ghcr.io/hkuds/deeptutor:latest
-# Apple Silicon (ARM64): use ghcr.io/hkuds/deeptutor:latest-arm64
 # Windows PowerShell: use ${PWD} instead of $(pwd)
 ```
@@ -421,14 +425,16 @@ docker compose up --build # Rebuild after changes
 <details>
 <summary>📋 <b>More Docker Options</b> (Pre-built images, Cloud deployment, Custom ports)</summary>
-**Pre-built Image Architecture Reference:**
+**Pre-built Image Tags:**
-| Architecture | Image Tag | Use Case |
-|:-------------|:----------|:---------|
-| **AMD64** | `ghcr.io/hkuds/deeptutor:latest` | Intel/AMD (most servers, Windows/Linux PCs) |
-| **ARM64** | `ghcr.io/hkuds/deeptutor:latest-arm64` | Apple Silicon, AWS Graviton, Raspberry Pi |
+| Tag | Architectures | Description |
+|:----|:--------------|:------------|
+| `:latest` | AMD64 + ARM64 | Latest stable release (auto-detects your architecture) |
+| `:v0.5.x` | AMD64 + ARM64 | Specific version (auto-detects your architecture) |
+| `:v0.5.x-amd64` | AMD64 only | Explicit AMD64 image |
+| `:v0.5.x-arm64` | ARM64 only | Explicit ARM64 image |
-> 💡 Run `uname -m` to check: `x86_64` = AMD64, `arm64`/`aarch64` = ARM64
+> 💡 The `:latest` tag is a **multi-architecture image** — Docker automatically pulls the correct version for your system (Intel/AMD or Apple Silicon/ARM)
 **Cloud Deployment** — Must set external API URL:
@@ -1605,8 +1611,9 @@ We hope DeepTutor could become a gift for the community. 🎁
 This project is licensed under the ***[AGPL-3.0 License](LICENSE)***.
-*✨ Thanks for visiting **DeepTutor**!*
-<img src="https://visitor-badge.laobi.icu/badge?page_id=HKUDS.DeepTutor&style=for-the-badge&color=00d4ff" alt="Views">
+<p align="center">
+  <em> Thanks for visiting ✨ DeepTutor!</em><br><br>
+  <img src="https://visitor-badge.laobi.icu/badge?page_id=HKUDS.DeepTutor&style=for-the-badge&color=00d4ff" alt="Views">
+</p>
 </div>

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/README.md RENAMED Viewed

@@ -37,8 +37,6 @@
 ---
 ### 📰 News
-> **[2026.1.15]** DeepTutor [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) is out! Fixed multiple environment configuration and stability issues. We recommend everyone to pull the latest version! 🎉
 > **[2026.1.1]** Happy New Year! Join our [Discord Community](https://discord.gg/zpP9cssj), [Wechat Community](https://github.com/HKUDS/DeepTutor/issues/78), or [Discussions](https://github.com/HKUDS/DeepTutor/discussions) - shape the future of DeepTutor! 💬
 > **[2025.12.30]** Visit our [Official Website](https://hkuds.github.io/DeepTutor/) for more details!
@@ -47,10 +45,16 @@
 ### 📦 Releases
-> **[2026.1.15]** Release [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) - Unified LLM & Embedding services, RAG pipeline selection, and major enhancements to Home, History, QuestionGen & Settings modules -- Thanks to all the contributors!
+> **[2026.1.23]** Release [v0.6.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.6.0) - Frontend session persistence, full Chinese support, Docker deployment updates, and minor bug fixes -- Thanks for all the feedback!
 <details>
 <summary>History releases</summary>
+> **[2026.1.18]** Release [v0.5.2](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.1) - Enhance RAG pipeline with Docling support and improve CI/CD workflows with several minor bugs fixed -- Thanks to all the feedbacks!
+> **[2026.1.15]** Release [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) - Unified LLM & Embedding services, RAG pipeline selection, and major enhancements to Home, History, QuestionGen & Settings modules -- Thanks to all the contributors!
 > **[2026.1.9]** Release [v0.4.1](https://github.com/HKUDS/DeepTutor/releases/tag/v0.4.1) with LLM Provider system overhaul, Question Generation robustness improvements, and codebase cleanup - Thanks to all the contributors!
 > **[2026.1.9]** Release [v0.4.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.4.0) with new code structure, multiple llm & embeddings support - Thanks to all the contributors!
@@ -337,14 +341,14 @@ cp .env.example .env
 **Quick Start** — Build from source:
 ```bash
-docker compose up --build -d    # Build and start (~5-10 min first run)
-docker compose logs -f          # View logs
+docker compose up                  # Build and start (~11 min first run on mac mini M4)
+docker compose build --no-cache    # Clear cache and rebuild after pull the newest repo
 ```
 **Or use pre-built image** (faster):
 ```bash
-# Linux/macOS (AMD64)
+# Works on all platforms - Docker auto-detects your architecture
 docker run -d --name deeptutor \
   -p 8001:8001 -p 3782:3782 \
   --env-file .env \
@@ -352,7 +356,6 @@ docker run -d --name deeptutor \
   -v $(pwd)/config:/app/config:ro \
   ghcr.io/hkuds/deeptutor:latest
-# Apple Silicon (ARM64): use ghcr.io/hkuds/deeptutor:latest-arm64
 # Windows PowerShell: use ${PWD} instead of $(pwd)
 ```
@@ -368,14 +371,16 @@ docker compose up --build # Rebuild after changes
 <details>
 <summary>📋 <b>More Docker Options</b> (Pre-built images, Cloud deployment, Custom ports)</summary>
-**Pre-built Image Architecture Reference:**
+**Pre-built Image Tags:**
-| Architecture | Image Tag | Use Case |
-|:-------------|:----------|:---------|
-| **AMD64** | `ghcr.io/hkuds/deeptutor:latest` | Intel/AMD (most servers, Windows/Linux PCs) |
-| **ARM64** | `ghcr.io/hkuds/deeptutor:latest-arm64` | Apple Silicon, AWS Graviton, Raspberry Pi |
+| Tag | Architectures | Description |
+|:----|:--------------|:------------|
+| `:latest` | AMD64 + ARM64 | Latest stable release (auto-detects your architecture) |
+| `:v0.5.x` | AMD64 + ARM64 | Specific version (auto-detects your architecture) |
+| `:v0.5.x-amd64` | AMD64 only | Explicit AMD64 image |
+| `:v0.5.x-arm64` | ARM64 only | Explicit ARM64 image |
-> 💡 Run `uname -m` to check: `x86_64` = AMD64, `arm64`/`aarch64` = ARM64
+> 💡 The `:latest` tag is a **multi-architecture image** — Docker automatically pulls the correct version for your system (Intel/AMD or Apple Silicon/ARM)
 **Cloud Deployment** — Must set external API URL:
@@ -1552,8 +1557,9 @@ We hope DeepTutor could become a gift for the community. 🎁
 This project is licensed under the ***[AGPL-3.0 License](LICENSE)***.
-*✨ Thanks for visiting **DeepTutor**!*
-<img src="https://visitor-badge.laobi.icu/badge?page_id=HKUDS.DeepTutor&style=for-the-badge&color=00d4ff" alt="Views">
+<p align="center">
+  <em> Thanks for visiting ✨ DeepTutor!</em><br><br>
+  <img src="https://visitor-badge.laobi.icu/badge?page_id=HKUDS.DeepTutor&style=for-the-badge&color=00d4ff" alt="Views">
+</p>
 </div>

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/pyproject.toml RENAMED Viewed

@@ -7,7 +7,7 @@ build-backend = "setuptools.build_meta"
 [project]
 name = "realtimex-deeptutor"
-version = "0.5.0.post1"
+version = "0.5.0.post2"
 description = "RealTimeX DeepTutor - Intelligent learning companion with multi-agent collaboration and LightRAG"
 requires-python = ">=3.10"
 readme = "README.md"
@@ -43,6 +43,8 @@ dependencies = [
     "numpy>=1.24.0,<2.0.0",
     # Academic and research tools
     "arxiv>=2.0.0",
+    # RealTimeX SDK
+    "realtimex-sdk>=1.1.2",
 ]
 # Optional dependencies for different providers and features
@@ -79,7 +81,12 @@ dev = [
 ]
 [project.scripts]
-deeptutor = "src.api.run_server:main"
+# Main entry point for full-stack startup (uvx realtimex-deeptutor)
+realtimex-deeptutor = "src.cli.start:main"
+deeptutor = "src.cli.start:main"  # Shorter alias
+# Backend-only mode (existing, for advanced users)
+deeptutor-backend = "src.api.run_server:main"
 [tool.setuptools.packages.find]
 where = ["."]
@@ -231,7 +238,8 @@ split-on-trailing-comma = true
 # ============================================
 [tool.ruff.lint.per-file-ignores]
 # Allow longer lines in test files
-"**/test_*.py" = ["E501", "PLR2004"]
+# F401: Allow unused imports for availability checks in tests
+"**/test_*.py" = ["E501", "PLR2004", "F401"]
 "**/__init__.py" = ["F401"]  # Allow unused imports in __init__.py
 # ============================================

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/realtimex_deeptutor.egg-info/PKG-INFO RENAMED Viewed

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: realtimex-deeptutor
-Version: 0.5.0.post1
+Version: 0.5.0.post2
 Summary: RealTimeX DeepTutor - Intelligent learning companion with multi-agent collaboration and LightRAG
 License: Apache-2.0
 Requires-Python: >=3.10
@@ -26,6 +26,7 @@ Requires-Dist: lightrag-hku>=1.0.0
 Requires-Dist: PyMuPDF>=1.26.0
 Requires-Dist: numpy<2.0.0,>=1.24.0
 Requires-Dist: arxiv>=2.0.0
+Requires-Dist: realtimex-sdk>=1.1.2
 Provides-Extra: anthropic
 Requires-Dist: anthropic>=0.30.0; extra == "anthropic"
 Provides-Extra: dashscope
@@ -90,8 +91,6 @@ Dynamic: license-file
 ---
 ### 📰 News
-> **[2026.1.15]** DeepTutor [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) is out! Fixed multiple environment configuration and stability issues. We recommend everyone to pull the latest version! 🎉
 > **[2026.1.1]** Happy New Year! Join our [Discord Community](https://discord.gg/zpP9cssj), [Wechat Community](https://github.com/HKUDS/DeepTutor/issues/78), or [Discussions](https://github.com/HKUDS/DeepTutor/discussions) - shape the future of DeepTutor! 💬
 > **[2025.12.30]** Visit our [Official Website](https://hkuds.github.io/DeepTutor/) for more details!
@@ -100,10 +99,16 @@ Dynamic: license-file
 ### 📦 Releases
-> **[2026.1.15]** Release [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) - Unified LLM & Embedding services, RAG pipeline selection, and major enhancements to Home, History, QuestionGen & Settings modules -- Thanks to all the contributors!
+> **[2026.1.23]** Release [v0.6.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.6.0) - Frontend session persistence, full Chinese support, Docker deployment updates, and minor bug fixes -- Thanks for all the feedback!
 <details>
 <summary>History releases</summary>
+> **[2026.1.18]** Release [v0.5.2](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.1) - Enhance RAG pipeline with Docling support and improve CI/CD workflows with several minor bugs fixed -- Thanks to all the feedbacks!
+> **[2026.1.15]** Release [v0.5.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.5.0) - Unified LLM & Embedding services, RAG pipeline selection, and major enhancements to Home, History, QuestionGen & Settings modules -- Thanks to all the contributors!
 > **[2026.1.9]** Release [v0.4.1](https://github.com/HKUDS/DeepTutor/releases/tag/v0.4.1) with LLM Provider system overhaul, Question Generation robustness improvements, and codebase cleanup - Thanks to all the contributors!
 > **[2026.1.9]** Release [v0.4.0](https://github.com/HKUDS/DeepTutor/releases/tag/v0.4.0) with new code structure, multiple llm & embeddings support - Thanks to all the contributors!
@@ -390,14 +395,14 @@ cp .env.example .env
 **Quick Start** — Build from source:
 ```bash
-docker compose up --build -d    # Build and start (~5-10 min first run)
-docker compose logs -f          # View logs
+docker compose up                  # Build and start (~11 min first run on mac mini M4)
+docker compose build --no-cache    # Clear cache and rebuild after pull the newest repo
 ```
 **Or use pre-built image** (faster):
 ```bash
-# Linux/macOS (AMD64)
+# Works on all platforms - Docker auto-detects your architecture
 docker run -d --name deeptutor \
   -p 8001:8001 -p 3782:3782 \
   --env-file .env \
@@ -405,7 +410,6 @@ docker run -d --name deeptutor \
   -v $(pwd)/config:/app/config:ro \
   ghcr.io/hkuds/deeptutor:latest
-# Apple Silicon (ARM64): use ghcr.io/hkuds/deeptutor:latest-arm64
 # Windows PowerShell: use ${PWD} instead of $(pwd)
 ```
@@ -421,14 +425,16 @@ docker compose up --build # Rebuild after changes
 <details>
 <summary>📋 <b>More Docker Options</b> (Pre-built images, Cloud deployment, Custom ports)</summary>
-**Pre-built Image Architecture Reference:**
+**Pre-built Image Tags:**
-| Architecture | Image Tag | Use Case |
-|:-------------|:----------|:---------|
-| **AMD64** | `ghcr.io/hkuds/deeptutor:latest` | Intel/AMD (most servers, Windows/Linux PCs) |
-| **ARM64** | `ghcr.io/hkuds/deeptutor:latest-arm64` | Apple Silicon, AWS Graviton, Raspberry Pi |
+| Tag | Architectures | Description |
+|:----|:--------------|:------------|
+| `:latest` | AMD64 + ARM64 | Latest stable release (auto-detects your architecture) |
+| `:v0.5.x` | AMD64 + ARM64 | Specific version (auto-detects your architecture) |
+| `:v0.5.x-amd64` | AMD64 only | Explicit AMD64 image |
+| `:v0.5.x-arm64` | ARM64 only | Explicit ARM64 image |
-> 💡 Run `uname -m` to check: `x86_64` = AMD64, `arm64`/`aarch64` = ARM64
+> 💡 The `:latest` tag is a **multi-architecture image** — Docker automatically pulls the correct version for your system (Intel/AMD or Apple Silicon/ARM)
 **Cloud Deployment** — Must set external API URL:
@@ -1605,8 +1611,9 @@ We hope DeepTutor could become a gift for the community. 🎁
 This project is licensed under the ***[AGPL-3.0 License](LICENSE)***.
-*✨ Thanks for visiting **DeepTutor**!*
-<img src="https://visitor-badge.laobi.icu/badge?page_id=HKUDS.DeepTutor&style=for-the-badge&color=00d4ff" alt="Views">
+<p align="center">
+  <em> Thanks for visiting ✨ DeepTutor!</em><br><br>
+  <img src="https://visitor-badge.laobi.icu/badge?page_id=HKUDS.DeepTutor&style=for-the-badge&color=00d4ff" alt="Views">
+</p>
 </div>

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/realtimex_deeptutor.egg-info/SOURCES.txt RENAMED Viewed

@@ -88,6 +88,7 @@ src/agents/research/utils/json_utils.py
 src/agents/research/utils/token_tracker.py
 src/agents/solve/__init__.py
 src/agents/solve/main_solver.py
+src/agents/solve/session_manager.py
 src/agents/solve/analysis_loop/__init__.py
 src/agents/solve/analysis_loop/investigate_agent.py
 src/agents/solve/analysis_loop/note_agent.py
@@ -137,6 +138,7 @@ src/api/routers/ideagen.py
 src/api/routers/knowledge.py
 src/api/routers/notebook.py
 src/api/routers/question.py
+src/api/routers/realtimex.py
 src/api/routers/research.py
 src/api/routers/settings.py
 src/api/routers/solve.py
@@ -146,6 +148,8 @@ src/api/utils/log_interceptor.py
 src/api/utils/notebook_manager.py
 src/api/utils/progress_broadcaster.py
 src/api/utils/task_id_manager.py
+src/cli/__init__.py
+src/cli/start.py
 src/config/__init__.py
 src/config/accessors.py
 src/config/constants.py
@@ -190,6 +194,7 @@ src/services/embedding/adapters/cohere.py
 src/services/embedding/adapters/jina.py
 src/services/embedding/adapters/ollama.py
 src/services/embedding/adapters/openai_compatible.py
+src/services/embedding/adapters/realtimex.py
 src/services/llm/__init__.py
 src/services/llm/capabilities.py
 src/services/llm/client.py
@@ -199,6 +204,7 @@ src/services/llm/error_mapping.py
 src/services/llm/exceptions.py
 src/services/llm/factory.py
 src/services/llm/local_provider.py
+src/services/llm/realtimex_provider.py
 src/services/llm/registry.py
 src/services/llm/telemetry.py
 src/services/llm/types.py
@@ -240,10 +246,12 @@ src/services/rag/components/retrievers/dense.py
 src/services/rag/components/retrievers/hybrid.py
 src/services/rag/components/retrievers/lightrag.py
 src/services/rag/pipelines/__init__.py
-src/services/rag/pipelines/academic.py
 src/services/rag/pipelines/lightrag.py
 src/services/rag/pipelines/llamaindex.py
 src/services/rag/pipelines/raganything.py
+src/services/rag/pipelines/raganything_docling.py
+src/services/rag/utils/__init__.py
+src/services/rag/utils/image_migration.py
 src/services/search/__init__.py
 src/services/search/base.py
 src/services/search/consolidation.py
@@ -255,6 +263,8 @@ src/services/search/providers/jina.py
 src/services/search/providers/perplexity.py
 src/services/search/providers/serper.py
 src/services/search/providers/tavily.py
+src/services/settings/__init__.py
+src/services/settings/interface_settings.py
 src/services/setup/__init__.py
 src/services/setup/init.py
 src/services/tts/__init__.py
@@ -276,4 +286,5 @@ src/utils/document_validator.py
 src/utils/error_rate_tracker.py
 src/utils/error_utils.py
 src/utils/json_parser.py
+src/utils/realtimex.py
 src/utils/network/circuit_breaker.py

realtimex_deeptutor-0.5.0.post2/realtimex_deeptutor.egg-info/entry_points.txt ADDED Viewed

@@ -0,0 +1,4 @@
+[console_scripts]
+deeptutor = src.cli.start:main
+deeptutor-backend = src.api.run_server:main
+realtimex-deeptutor = src.cli.start:main

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/realtimex_deeptutor.egg-info/requires.txt RENAMED Viewed

@@ -18,6 +18,7 @@ lightrag-hku>=1.0.0
 PyMuPDF>=1.26.0
 numpy<2.0.0,>=1.24.0
 arxiv>=2.0.0
+realtimex-sdk>=1.1.2
 [academic]
 arxiv>=2.0.0

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/src/__init__.py RENAMED Viewed

@@ -17,7 +17,7 @@ Core Modules:
 Usage:
     # Import as a package for integration with other services
     from src.api.main import app
     # Or use individual modules
     from src.services.rag import RAGService
     from src.agents import ChatAgent
@@ -30,7 +30,7 @@ __package_name__ = "realtimex-deeptutor"
 # These are imported lazily to avoid circular imports and heavy startup
 __all__ = [
     "agents",
-    "api",
+    "api",
     "services",
     "knowledge",
     "logging",

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/src/agents/ideagen/material_organizer_agent.py RENAMED Viewed

@@ -25,6 +25,7 @@ class MaterialOrganizerAgent(BaseAgent):
         language: str = "en",
         api_key: str | None = None,
         base_url: str | None = None,
+        api_version: str | None = None,
         model: str | None = None,
     ):
         super().__init__(
@@ -32,6 +33,7 @@ class MaterialOrganizerAgent(BaseAgent):
             agent_name="material_organizer",
             api_key=api_key,
             base_url=base_url,
+            api_version=api_version,
             model=model,
             language=language,
         )

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/src/agents/solve/__init__.py RENAMED Viewed

@@ -42,6 +42,9 @@ from .memory import (
     ToolCallRecord,
 )
+# Session management
+from .session_manager import SolverSessionManager, get_solver_session_manager
 # Solve loop
 from .solve_loop import (
     ManagerAgent,
@@ -77,4 +80,7 @@ __all__ = [
     "ToolAgent",
     # Main Controller
     "MainSolver",
+    # Session Management
+    "SolverSessionManager",
+    "get_solver_session_manager",
 ]

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/src/agents/solve/main_solver.py RENAMED Viewed

@@ -43,6 +43,7 @@ class MainSolver:
         api_key: str | None = None,
         base_url: str | None = None,
         api_version: str | None = None,
+        language: str | None = None,
         kb_name: str = "ai_textbook",
         output_base_dir: str | None = None,
     ):
@@ -55,6 +56,7 @@ class MainSolver:
             api_key: API key (if not provided, read from environment)
             base_url: API URL (if not provided, read from environment)
             api_version: API version (if not provided, read from environment)
+            language: Preferred language for prompts ("en"/"zh"/"cn")
             kb_name: Knowledge base name
             output_base_dir: Output base directory (optional, overrides config)
         """
@@ -63,6 +65,7 @@ class MainSolver:
         self._api_key = api_key
         self._base_url = base_url
         self._api_version = api_version
+        self._language = language
         self._kb_name = kb_name
         self._output_base_dir = output_base_dir
@@ -108,6 +111,7 @@ class MainSolver:
         api_version = self._api_version
         kb_name = self._kb_name
         output_base_dir = self._output_base_dir
+        language = self._language
         # Load config from config directory (main.yaml unified config)
         if config_path is None:
@@ -156,6 +160,11 @@ class MainSolver:
         if self.config is None or not isinstance(self.config, dict):
             self.config = {}
+        # Override system language from UI if provided
+        if language:
+            self.config.setdefault("system", {})
+            self.config["system"]["language"] = parse_language(language)
         # Override output directory config
         if output_base_dir:
             if "system" not in self.config:

{realtimex_deeptutor-0.5.0.post1 → realtimex_deeptutor-0.5.0.post2}/src/agents/solve/prompts/zh/analysis_loop/investigate_agent.yaml RENAMED Viewed

@@ -11,18 +11,19 @@ system: |
   4. **聚焦核心**：只关注解题必须的"硬知识"（定义、公式、定理）。除非用户明确要求，否则不要查询应用案例或背景故事。
   # 工具使用指南
-  - `rag_naive`: **首选**。用于查询明确的术语定义、核心公式、或验证某个概念是否存在。速度快。
+  - `query_item`: **编号条目优先**。当用户问题中**明确提到具体编号条目**（例如“公式 2.1.2 / Equation (2.1.2)”、“定理 3.1 / Theorem 3.1”、“图 1.2 / Figure 1.2”）时，**必须优先使用该工具（ALWAYS use this first）**。请从问题中提取标识符并直接查询（常见格式：方程用 `(X.X.X)`，图用 `Figure X.X`，定理用 `Theorem X.X`）。这是检索编号内容最直接、最准确的方式。
+  - `rag_naive`: **通用查询首选**。用于查询明确的术语定义、核心公式，或验证某个概念是否存在。速度快。
   - `rag_hybrid`: 用于需要综合多个概念、探究复杂实体关系或深层原理的场景。
   - `web_search`: **慎用**。仅当信息极有可能超出教材范围（如最新新闻、特定技术参数、开源库用法）时使用。
-  - `query_item`: **专用**。仅当你明确知道需要获取某个特定编号（如 "Fig 1.2", "Theorem 3.1"）的内容时使用。
   - `none`: 当现有信息足以支撑解题，或缺口无法通过检索填补（如需要用户提供）时，返回此状态。
   # 思考路径
-  1. **解析需求**：解题真正缺的是什么？是某个变量的定义？是某个定理的公式？还是某个常数的数值？
-  2. **核查存量**：查看`已有查询内容`。答案是否已经存在？或者是否隐含在已知信息中？
-  3. **构建查询**：如果确有缺口，构建原子化的、互不重叠的查询请求。
-  4. **判断终止**：如果核心定义和公式都已齐备，果断停止。
-  5. **及时止损**：如果某话题的第一轮查询没有返回有用信息（结果为空、不相关或质量低），说明知识库中可能没有相关内容，应该放弃该话题，换其他话题，或直接进入下一阶段。不要对同一个话题反复查询。
+  1. **识别编号条目**：**首先扫描用户问题中是否存在明确的编号引用**（例如“公式 2.1.2 / Equation (2.1.2)”、“定理 3.1 / Theorem 3.1”、“图 1.2 / Figure 1.2”）。若存在，提取标识符并立即使用 `query_item`。编号条目优先于语义检索。
+  2. **解析需求**：解题真正缺的是什么？是某个变量的定义？是某个定理的公式？还是某个常数的数值？
+  3. **核查存量**：查看`已有查询内容`。答案是否已经存在？或者是否隐含在已知信息中？
+  4. **构建查询**：如果确有缺口，构建原子化的、互不重叠的查询请求。
+  5. **判断终止**：如果核心定义和公式都已齐备，果断停止。
+  6. **⚠️ 关键 - 及时止损（Know When to Cut Losses）**：如果某话题的第一轮查询没有返回有用信息（结果为空、不相关或质量低），通常意味着知识库不包含相关内容。**你必须立即放弃该话题**，改查其他话题，或直接进入下一阶段。**不要对同一个话题反复查询**——这会浪费资源并拖慢解题。
   # 输出格式
   直接输出 JSON 对象（无 Markdown 格式）：
@@ -46,6 +47,7 @@ user_template: |
   ## 任务
   分析是否存在阻碍解题的知识缺口。
+  - **编号条目优先**：若问题提到具体编号条目（公式/定理/图/方程编号如“2.1.2”“3.1”等），**立即使用 `query_item`** 并传入提取的标识符（例如方程用 `(2.1.2)`，图用 `Figure 2.1`）。
   - 检查重复：新查询不得与已有内容重叠。
   - 灵活决策：如果前序工具查询出错，考虑通过其他工具来达到相同的目的。
   - 及时止损：如果某话题的第一轮查询没有返回有用信息，不要反复查询该话题，应该换其他话题或进入下一阶段。

realtimex-deeptutor 0.5.0.post1__tar.gz → 0.5.0.post2__tar.gz

realtimex-deeptutor 0.5.0.post1tar.gz → 0.5.0.post2tar.gz