eranmi commited on
Commit
c1d2cf1
·
2 Parent(s): c936336085a23f

Merge branch 'main' of https://github.com/AnnaErsh/RAGnTeX

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
.dockerignore ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ __pycache__
2
+ *.pyc
3
+ *.pyo
4
+ *.pyd
5
+ *.db
6
+ .env
7
+ .venv
.github/workflows/check.yml ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: Check file size
2
+ on:
3
+ pull_request:
4
+ branches: [main]
5
+
6
+ # to run this workflow manually from the Actions tab
7
+ workflow_dispatch:
8
+
9
+ jobs:
10
+ sync-to-hub:
11
+ runs-on: ubuntu-latest
12
+ steps:
13
+ - name: Check large files
14
+ uses: ActionsDesk/lfs-warning@v2.0
15
+ with:
16
+ filesizelimit: 10485760 # this is 10MB
.github/workflows/ci.yml ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: CI
2
+
3
+ on:
4
+ push:
5
+ branches: [main, dev]
6
+ pull_request:
7
+ branches: [main, dev]
8
+
9
+ jobs:
10
+ lint:
11
+ runs-on: ubuntu-latest
12
+
13
+ steps:
14
+ - uses: actions/checkout@v4
15
+
16
+ - name: Set up Python
17
+ uses: actions/setup-python@v5
18
+ with:
19
+ python-version: "3.10"
20
+
21
+ - name: Install Poetry
22
+ uses: abatilo/actions-poetry@v2
23
+ with:
24
+ poetry-version: "2.1.1"
25
+
26
+ - name: Verify Poetry Installation
27
+ run: poetry --version
28
+
29
+ - name: Install dependencies
30
+ run: poetry install --with dev
31
+
32
+ - name: Run Black and isort
33
+ run: |
34
+ poetry run black .
35
+ poetry run isort .
36
+
37
+ - name: Commit and push changes
38
+ uses: stefanzweifel/git-auto-commit-action@v5
39
+ with:
40
+ commit_message: "style: Automatic formatting via CI (Black + isort)"
41
+ branch: ${{ github.head_ref || github.ref_name }}
42
+ file_pattern: "*.py"
43
+
44
+ - name: Run Mypy
45
+ run: poetry run mypy .
46
+
47
+ - name: Run Pylint
48
+ run: |
49
+ poetry run pylint . | tee pylint-output.txt
50
+ score=$(tail -n 2 pylint-output.txt | grep -oP "rated at \K[0-9.]+")
51
+ echo "Pylint score: $score"
52
+ awk -v s="$score" 'BEGIN {exit (s >= 9.5) ? 0 : 1}'
53
+
54
+ - name: Run Bandit security scan
55
+ run: poetry run bandit -r src/
56
+
57
+ - name: Run Safety vulnerability check (warnings only)
58
+ run: |
59
+ poetry run safety check --full-report || true
60
+
61
+ - name: Build docs
62
+ working-directory: docs
63
+ run: poetry run sphinx-build -b html . _build/html
.github/workflows/hf_sync.yml ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: Sync to Hugging Face hub
2
+ on:
3
+ push:
4
+ branches: [main]
5
+ # to run this workflow manually from the Actions tab
6
+ workflow_dispatch:
7
+
8
+ jobs:
9
+ sync-to-hub:
10
+ runs-on: ubuntu-latest
11
+ steps:
12
+ - uses: actions/checkout@v3
13
+ with:
14
+ fetch-depth: 0
15
+ lfs: true
16
+ - name: Push to hub
17
+ env:
18
+ HF_SECRET: ${{ secrets.HF_SECRET }}
19
+ run: git push https://eranmi:$HF_SECRET@huggingface.co/spaces/eranmi/RAGnTeX main
Dockerfile ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Use official Python base image
2
+ FROM python:3.10-slim
3
+
4
+ # Set environment variables
5
+ ENV PYTHONDONTWRITEBYTECODE=1
6
+ ENV PYTHONUNBUFFERED=1
7
+
8
+ # Set work directory
9
+ WORKDIR /app
10
+
11
+ # Install system dependencies
12
+ RUN apt-get update && apt-get install -y \
13
+ build-essential \
14
+ libgl1-mesa-glx \
15
+ libglib2.0-0 \
16
+ texlive-latex-base \
17
+ texlive-fonts-recommended \
18
+ texlive-latex-extra \
19
+ && rm -rf /var/lib/apt/lists/*
20
+
21
+ # Clean up
22
+ RUN rm -rf /var/lib/apt/lists/*
23
+
24
+ # Install Poetry
25
+ RUN pip install --upgrade pip
26
+ RUN pip install poetry
27
+
28
+ # Copy only poetry files first (for caching)
29
+ COPY pyproject.toml poetry.lock* /app/
30
+
31
+ # Install dependencies
32
+ RUN poetry config virtualenvs.create false \
33
+ && poetry install --no-root --no-interaction --no-ansi
34
+
35
+ # Copy the rest of the application
36
+ COPY . /app
37
+
38
+ # Expose Gradio port
39
+ EXPOSE 7860
40
+
41
+ # Command to run the app
42
+ CMD ["python", "ragntex.py"]
LICENSE ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ MIT License
2
+
3
+ Copyright (c) 2025 RAGnTeX
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
README.md CHANGED
@@ -1,3 +1,12 @@
 
 
 
 
 
 
 
 
 
1
  # 📚 From Text to Visuals: Auto-Generating LaTeX Beamer Presentations with GenAI
2
 
3
  In this project, we explore how generative AI can automate the creation of professional-looking presentation slides—directly from extensive collections of PDF documents.
@@ -42,3 +51,24 @@ The model outputs fully formatted compilable LaTeX code following a clear and co
42
  - **Summary**: Key takeaways  
43
 
44
  It also incorporates relevant images extracted from the documents, organizing them into LaTeX Beamer-friendly layouts (e.g., *Core Idea 2* and *Core Idea 3* slide formats).
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ title: RAGnTeX
3
+ emoji: 📄
4
+ colorFrom: indigo
5
+ colorTo: pink
6
+ sdk: docker
7
+ pinned: false
8
+ ---
9
+
10
  # 📚 From Text to Visuals: Auto-Generating LaTeX Beamer Presentations with GenAI
11
 
12
  In this project, we explore how generative AI can automate the creation of professional-looking presentation slides—directly from extensive collections of PDF documents.
 
51
  - **Summary**: Key takeaways  
52
 
53
  It also incorporates relevant images extracted from the documents, organizing them into LaTeX Beamer-friendly layouts (e.g., *Core Idea 2* and *Core Idea 3* slide formats).
54
+
55
+ ### Prerequisites
56
+
57
+ This project requires the LaTeX toolchain, specifically `pdflatex`, to generate PDF presentations.
58
+
59
+ Please install it before running the project:
60
+
61
+ ### On Debian/Ubuntu:
62
+
63
+ ```bash
64
+ sudo apt-get update && sudo apt-get install -y texlive-latex-base texlive-fonts-recommended texlive-latex-extra
65
+
66
+
67
+ ### Run docker:
68
+ ```bash
69
+ docker run \
70
+ --env-file .env \
71
+ -e IN_DOCKER=true \
72
+ -p 7860:7860 \
73
+ -v /path/to/your/output:/app/output \
74
+ ragntex
docs/Makefile ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Minimal makefile for Sphinx documentation
2
+ #
3
+
4
+ # You can set these variables from the command line, and also
5
+ # from the environment for the first two.
6
+ SPHINXOPTS ?=
7
+ SPHINXBUILD ?= sphinx-build
8
+ SOURCEDIR = .
9
+ BUILDDIR = _build
10
+
11
+ # Put it first so that "make" without argument is like "make help".
12
+ help:
13
+ @$(SPHINXBUILD) -M help "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
14
+
15
+ .PHONY: help Makefile
16
+
17
+ # Catch-all target: route all unknown targets to Sphinx using the new
18
+ # "make mode" option. $(O) is meant as a shortcut for $(SPHINXOPTS).
19
+ %: Makefile
20
+ @$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
docs/conf.py ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Configuration file for the Sphinx documentation builder.
2
+ #
3
+ # For the full list of built-in configuration values, see the documentation:
4
+ # https://www.sphinx-doc.org/en/master/usage/configuration.html
5
+
6
+ # -- Project information -----------------------------------------------------
7
+ # https://www.sphinx-doc.org/en/master/usage/configuration.html#project-information
8
+ import os
9
+ import sys
10
+
11
+ project = "RAGnTeX"
12
+ copyright = "2025, Anna Ershova, Kajetan Niewczas"
13
+ author = "Anna Ershova, Kajetan Niewczas"
14
+ # release = "0.1.0"
15
+
16
+ # ── allow “import ragntex …” ────────────────────────────────────────────────
17
+ sys.path.insert(0, os.path.abspath("../src"))
18
+ # -- General configuration ---------------------------------------------------
19
+ # https://www.sphinx-doc.org/en/master/usage/configuration.html#general-configuration
20
+
21
+ extensions = [
22
+ "sphinx.ext.autodoc",
23
+ "sphinx.ext.napoleon",
24
+ "sphinx_autodoc_typehints",
25
+ ]
26
+
27
+
28
+ templates_path = ["_templates"]
29
+ exclude_patterns = ["_build", "Thumbs.db", ".DS_Store"]
30
+
31
+ # -- Options for Napoleon ----------------------------------------------------
32
+
33
+ napoleon_google_docstring = True
34
+ napoleon_numpy_docstring = False
35
+
36
+ # -- Options for HTML output -------------------------------------------------
37
+ # https://www.sphinx-doc.org/en/master/usage/configuration.html#options-for-html-output
38
+
39
+ html_theme = "furo"
40
+ html_static_path = ["_static"]
docs/index.rst ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ .. RAGnTeX documentation master file, created by
2
+ sphinx-quickstart on Fri Jun 13 12:43:29 2025.
3
+ You can adapt this file completely to your liking, but it should at least
4
+ contain the root `toctree` directive.
5
+ RAGnTeX documentation
6
+ =====================
7
+
8
+ 📚 From Text to Visuals: Auto-Generating LaTeX Beamer Presentations with GenAI
9
+ ------------------------------------------------------------------------------
10
+
11
+ In this project, we explore how generative AI can automate the creation of professional-looking presentation slides—directly from extensive collections of PDF documents.
12
+
13
+ ---
14
+
15
+ 🧠 Use Case
16
+ ===========
17
+
18
+ Creating slide decks from dense documents (like whitepapers or scientific articles) is a time-consuming and cognitively heavy task. Our goal is to streamline this process using generative AI.
19
+
20
+ We built an AI assistant that transforms document collections into LaTeX Beamer presentations—complete with structure, content, and visuals—based on a user-defined topic.
21
+
22
+ ---
23
+
24
+ 🔍 How It Works
25
+ ==============
26
+
27
+ 💿 Database Creation
28
+ -------------------
29
+
30
+ We start by creating a vector database from a collection of PDF documents (e.g., arXiv papers). Each document is processed to extract text chunks, and for each chunk, we store:
31
+
32
+ - **Embeddings** of the text from the file
33
+ - **Associated metadata** (like images, their captions, how many images there are)
34
+
35
+ 🗣️ User Prompt
36
+ --------------
37
+
38
+ A user enters a natural language query like:
39
+
40
+ ::
41
+
42
+ "I need a presentation about AI agents."
43
+
44
+ 📄 Document Retrieval
45
+ --------------------
46
+
47
+ We use **Chroma** as our vector database to retrieve the most relevant PDF documents based on the query.
48
+
49
+ 🧠 Content Understanding
50
+ ------------------------
51
+
52
+ Using **Gemini 2.0 Flash** (``google-genai==1.7.0``), the assistant analyzes the retrieved content and extracts the most relevant ideas and insights.
53
+
54
+ 🎞️ Slide Generation (LaTeX Beamer)
55
+ ----------------------------------
56
+
57
+ The model outputs fully formatted compilable LaTeX code following a clear and consistent structure:
58
+
59
+ - **Introduction**: Topic overview and motivation
60
+ - **Main Content**: 2–4 slides, each presenting a core idea
61
+ - **Summary**: Key takeaways
62
+
63
+ It also incorporates relevant images extracted from the documents, organizing them into LaTeX Beamer-friendly layouts (e.g., *Core Idea 2* and *Core Idea 3* slide formats).
64
+
65
+ Installation
66
+ ============
67
+
68
+ Install poetry to manage dependencies and virtual environments:
69
+
70
+ .. code-block:: bash
71
+
72
+ pip install poetry
73
+
74
+ Then, install the project dependencies:
75
+
76
+ .. code-block:: bash
77
+
78
+ poetry install
79
+
80
+ Running
81
+ =======
82
+
83
+ Run project with:
84
+
85
+ .. code-block:: bash
86
+
87
+ poetry run python ragntex.py
88
+
89
+ Or via Docker:
90
+
91
+ .. code-block:: bash
92
+
93
+ docker run \
94
+ --env-file .env \
95
+ -e IN_DOCKER=true \
96
+ -p 7860:7860 \
97
+ -v /path/to/your/output:/app/output \
98
+ ragntex
99
+
100
+ .. toctree::
101
+ :maxdepth: 2
102
+ :caption: Contents:
103
+
104
+ src
docs/make.bat ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ @ECHO OFF
2
+
3
+ pushd %~dp0
4
+
5
+ REM Command file for Sphinx documentation
6
+
7
+ if "%SPHINXBUILD%" == "" (
8
+ set SPHINXBUILD=sphinx-build
9
+ )
10
+ set SOURCEDIR=.
11
+ set BUILDDIR=_build
12
+
13
+ %SPHINXBUILD% >NUL 2>NUL
14
+ if errorlevel 9009 (
15
+ echo.
16
+ echo.The 'sphinx-build' command was not found. Make sure you have Sphinx
17
+ echo.installed, then set the SPHINXBUILD environment variable to point
18
+ echo.to the full path of the 'sphinx-build' executable. Alternatively you
19
+ echo.may add the Sphinx directory to PATH.
20
+ echo.
21
+ echo.If you don't have Sphinx installed, grab it from
22
+ echo.https://www.sphinx-doc.org/
23
+ exit /b 1
24
+ )
25
+
26
+ if "%1" == "" goto help
27
+
28
+ %SPHINXBUILD% -M %1 %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O%
29
+ goto end
30
+
31
+ :help
32
+ %SPHINXBUILD% -M help %SOURCEDIR% %BUILDDIR% %SPHINXOPTS% %O%
33
+
34
+ :end
35
+ popd
docs/modules.rst ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ src
2
+ ===
3
+
4
+ .. toctree::
5
+ :maxdepth: 4
6
+
7
+ src
docs/source/modules.rst ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ RAGnTeX
2
+ =======
3
+
4
+ .. toctree::
5
+ :maxdepth: 4
6
+
7
+ ragntex
8
+ src
docs/source/ragntex.rst ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ ragntex module
2
+ ==============
3
+
4
+ .. automodule:: ragntex
5
+ :members:
6
+ :undoc-members:
7
+ :show-inheritance:
docs/source/src.compilation.rst ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.compilation package
2
+ =======================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.compilation.json\_to\_tex module
8
+ ------------------------------------
9
+
10
+ .. automodule:: src.compilation.json_to_tex
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.compilation.latex\_compilation module
16
+ -----------------------------------------
17
+
18
+ .. automodule:: src.compilation.latex_compilation
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ Module contents
24
+ ---------------
25
+
26
+ .. automodule:: src.compilation
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
docs/source/src.database.rst ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.database package
2
+ ====================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.database.database module
8
+ ----------------------------
9
+
10
+ .. automodule:: src.database.database
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.database.db\_manipulation module
16
+ ------------------------------------
17
+
18
+ .. automodule:: src.database.db_manipulation
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ Module contents
24
+ ---------------
25
+
26
+ .. automodule:: src.database
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
docs/source/src.interface.rst ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.interface package
2
+ =====================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.interface.download\_files module
8
+ ------------------------------------
9
+
10
+ .. automodule:: src.interface.download_files
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.interface.gradio\_interface module
16
+ --------------------------------------
17
+
18
+ .. automodule:: src.interface.gradio_interface
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ src.interface.upload\_files module
24
+ ----------------------------------
25
+
26
+ .. automodule:: src.interface.upload_files
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
30
+
31
+ Module contents
32
+ ---------------
33
+
34
+ .. automodule:: src.interface
35
+ :members:
36
+ :undoc-members:
37
+ :show-inheritance:
docs/source/src.processing.rst ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.processing package
2
+ ======================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.processing.document\_processing module
8
+ ------------------------------------------
9
+
10
+ .. automodule:: src.processing.document_processing
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.processing.images\_processing module
16
+ ----------------------------------------
17
+
18
+ .. automodule:: src.processing.images_processing
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ src.processing.output\_folder module
24
+ ------------------------------------
25
+
26
+ .. automodule:: src.processing.output_folder
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
30
+
31
+ Module contents
32
+ ---------------
33
+
34
+ .. automodule:: src.processing
35
+ :members:
36
+ :undoc-members:
37
+ :show-inheritance:
docs/source/src.rst ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src package
2
+ ===========
3
+
4
+ Subpackages
5
+ -----------
6
+
7
+ .. toctree::
8
+ :maxdepth: 4
9
+
10
+ src.compilation
11
+ src.database
12
+ src.interface
13
+ src.processing
14
+ src.services
15
+ src.telemetry
16
+
17
+ Submodules
18
+ ----------
19
+
20
+ src.generator module
21
+ --------------------
22
+
23
+ .. automodule:: src.generator
24
+ :members:
25
+ :undoc-members:
26
+ :show-inheritance:
27
+
28
+ Module contents
29
+ ---------------
30
+
31
+ .. automodule:: src
32
+ :members:
33
+ :undoc-members:
34
+ :show-inheritance:
docs/source/src.services.rst ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.services package
2
+ ====================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.services.google\_client module
8
+ ----------------------------------
9
+
10
+ .. automodule:: src.services.google_client
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.services.output\_schema module
16
+ ----------------------------------
17
+
18
+ .. automodule:: src.services.output_schema
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ src.services.prompt module
24
+ --------------------------
25
+
26
+ .. automodule:: src.services.prompt
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
30
+
31
+ Module contents
32
+ ---------------
33
+
34
+ .. automodule:: src.services
35
+ :members:
36
+ :undoc-members:
37
+ :show-inheritance:
docs/source/src.telemetry.rst ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.telemetry package
2
+ =====================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.telemetry.logging\_utils module
8
+ -----------------------------------
9
+
10
+ .. automodule:: src.telemetry.logging_utils
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.telemetry.telemetry module
16
+ ------------------------------
17
+
18
+ .. automodule:: src.telemetry.telemetry
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ Module contents
24
+ ---------------
25
+
26
+ .. automodule:: src.telemetry
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
docs/src.compilation.rst ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.compilation package
2
+ =======================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.compilation.json\_to\_tex module
8
+ ------------------------------------
9
+
10
+ .. automodule:: src.compilation.json_to_tex
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.compilation.latex\_compilation module
16
+ -----------------------------------------
17
+
18
+ .. automodule:: src.compilation.latex_compilation
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ Module contents
24
+ ---------------
25
+
26
+ .. automodule:: src.compilation
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
docs/src.database.rst ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.database package
2
+ ====================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.database.database module
8
+ ----------------------------
9
+
10
+ .. automodule:: src.database.database
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.database.db\_manipulation module
16
+ ------------------------------------
17
+
18
+ .. automodule:: src.database.db_manipulation
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ Module contents
24
+ ---------------
25
+
26
+ .. automodule:: src.database
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
docs/src.interface.rst ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.interface package
2
+ =====================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.interface.download\_files module
8
+ ------------------------------------
9
+
10
+ .. automodule:: src.interface.download_files
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.interface.gradio\_interface module
16
+ --------------------------------------
17
+
18
+ .. automodule:: src.interface.gradio_interface
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ src.interface.upload\_files module
24
+ ----------------------------------
25
+
26
+ .. automodule:: src.interface.upload_files
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
30
+
31
+ Module contents
32
+ ---------------
33
+
34
+ .. automodule:: src.interface
35
+ :members:
36
+ :undoc-members:
37
+ :show-inheritance:
docs/src.processing.rst ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.processing package
2
+ ======================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.processing.document\_processing module
8
+ ------------------------------------------
9
+
10
+ .. automodule:: src.processing.document_processing
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.processing.images\_processing module
16
+ ----------------------------------------
17
+
18
+ .. automodule:: src.processing.images_processing
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ src.processing.output\_folder module
24
+ ------------------------------------
25
+
26
+ .. automodule:: src.processing.output_folder
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
30
+
31
+ Module contents
32
+ ---------------
33
+
34
+ .. automodule:: src.processing
35
+ :members:
36
+ :undoc-members:
37
+ :show-inheritance:
docs/src.rst ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src package
2
+ ===========
3
+
4
+ Subpackages
5
+ -----------
6
+
7
+ .. toctree::
8
+ :maxdepth: 4
9
+
10
+ src.compilation
11
+ src.database
12
+ src.interface
13
+ src.processing
14
+ src.services
15
+ src.telemetry
16
+
17
+ Submodules
18
+ ----------
19
+
20
+ src.generator module
21
+ --------------------
22
+
23
+ .. automodule:: src.generator
24
+ :members:
25
+ :undoc-members:
26
+ :show-inheritance:
27
+
28
+ Module contents
29
+ ---------------
30
+
31
+ .. automodule:: src
32
+ :members:
33
+ :undoc-members:
34
+ :show-inheritance:
docs/src.services.rst ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.services package
2
+ ====================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.services.google\_client module
8
+ ----------------------------------
9
+
10
+ .. automodule:: src.services.google_client
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.services.output\_schema module
16
+ ----------------------------------
17
+
18
+ .. automodule:: src.services.output_schema
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ src.services.prompt module
24
+ --------------------------
25
+
26
+ .. automodule:: src.services.prompt
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
30
+
31
+ Module contents
32
+ ---------------
33
+
34
+ .. automodule:: src.services
35
+ :members:
36
+ :undoc-members:
37
+ :show-inheritance:
docs/src.telemetry.rst ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ src.telemetry package
2
+ =====================
3
+
4
+ Submodules
5
+ ----------
6
+
7
+ src.telemetry.logging\_utils module
8
+ -----------------------------------
9
+
10
+ .. automodule:: src.telemetry.logging_utils
11
+ :members:
12
+ :undoc-members:
13
+ :show-inheritance:
14
+
15
+ src.telemetry.telemetry module
16
+ ------------------------------
17
+
18
+ .. automodule:: src.telemetry.telemetry
19
+ :members:
20
+ :undoc-members:
21
+ :show-inheritance:
22
+
23
+ Module contents
24
+ ---------------
25
+
26
+ .. automodule:: src.telemetry
27
+ :members:
28
+ :undoc-members:
29
+ :show-inheritance:
gfx/favicon.ico ADDED
gfx/github-mark-white.png ADDED

Git LFS Details

  • SHA256: 60d03f0f1a09aafcc2df2e10b4ffd24eab44787b451f8fdb367c3baf589cf95a
  • Pointer size: 129 Bytes
  • Size of remote file: 4.84 kB
gfx/icon_logo.png ADDED

Git LFS Details

  • SHA256: 1bd9b84f065b306abd8a2780bd39e88931dc98ae0e62ce75088b5add46cf4371
  • Pointer size: 131 Bytes
  • Size of remote file: 383 kB
gfx/long_logo.png ADDED

Git LFS Details

  • SHA256: 51fdbe2713564f222d57728adcafa9950fdce794f8af4eb149caabbffdd0d93a
  • Pointer size: 132 Bytes
  • Size of remote file: 1.07 MB
mypy.ini ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ # mypy.ini
2
+
3
+ [mypy]
4
+ ignore_missing_imports = True
5
+
poetry.lock CHANGED
The diff for this file is too large to render. See raw diff
 
pylint-output.txt ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+
2
+ ------------------------------------
3
+ Your code has been rated at 10.00/10
4
+
pyproject.toml CHANGED
@@ -3,11 +3,13 @@ name = "ragntex"
3
  version = "1.0.0"
4
  description = "From Text to Visuals: Auto-Generating LaTeX Beamer Presentations with GenAI"
5
  authors = [
6
- {name = "AnnaErsh",email = "aershova096@gmail.com"}
 
7
  ]
8
  license = {text = "Apache 2.0"}
9
  readme = "README.md"
10
- requires-python = ">=3.13"
 
11
  dependencies = [
12
  "google-genai (==1.11.0)",
13
  "chromadb (==1.0.5)",
@@ -18,9 +20,26 @@ dependencies = [
18
  "rtree (>=1.4.0,<2.0.0)",
19
  "pymupdf (>=1.25.5,<2.0.0)",
20
  "gradio (>=5.27.1,<6.0.0)",
 
 
 
21
  ]
22
 
23
 
24
  [build-system]
25
  requires = ["poetry-core>=2.0.0,<3.0.0"]
26
  build-backend = "poetry.core.masonry.api"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  version = "1.0.0"
4
  description = "From Text to Visuals: Auto-Generating LaTeX Beamer Presentations with GenAI"
5
  authors = [
6
+ {name = "AnnaErsh",email = "aershova096@gmail.com"},
7
+ {name = "KNiewczas",email = "kajetan.niewczas@gmail.com"}
8
  ]
9
  license = {text = "Apache 2.0"}
10
  readme = "README.md"
11
+ packages = [{ include = "src" }]
12
+ requires-python = ">=3.10,<3.11"
13
  dependencies = [
14
  "google-genai (==1.11.0)",
15
  "chromadb (==1.0.5)",
 
20
  "rtree (>=1.4.0,<2.0.0)",
21
  "pymupdf (>=1.25.5,<2.0.0)",
22
  "gradio (>=5.27.1,<6.0.0)",
23
+ "langfuse (>=2.60.5,<3.0.0)",
24
+ "pydantic (>=2.6.0,<2.10.0)",
25
+ "rsa (>=4.7,<5.0)",
26
  ]
27
 
28
 
29
  [build-system]
30
  requires = ["poetry-core>=2.0.0,<3.0.0"]
31
  build-backend = "poetry.core.masonry.api"
32
+
33
+ [tool.poetry.group.dev.dependencies]
34
+ pylint = "^3.3.7"
35
+ black = "^25.1.0"
36
+ pytest = "^8.4.0"
37
+ mypy = "^1.16.0"
38
+ isort = "^6.0.1"
39
+ safety = "^3.5.2"
40
+ bandit = "^1.8.3"
41
+
42
+
43
+ [tool.poetry.group.docs.dependencies]
44
+ sphinx-autodoc-typehints = "1.22"
45
+ furo = "^2024.8.6"
ragntex.py CHANGED
@@ -1,135 +1,16 @@
1
- import os
2
- import re
3
- from datetime import datetime
4
- from dotenv import load_dotenv
5
-
6
- # from src.database import db, embed_fn
7
- # from src.document_processing import process_documents
8
- # from src.images_processing import save_pdf_images, save_pdf_figures
9
- # from src.latex_compilation import CompilePresentation
10
- # from src.prompt import prompt
11
- # from src.google_client import client
12
-
13
- from src import (
14
- db,
15
- embed_fn,
16
- process_documents,
17
- save_pdf_images,
18
- save_pdf_figures,
19
- prompt,
20
- CompilePresentation,
21
- client,
22
- # demo,
23
- Logger,
24
- )
25
-
26
- LOGGER = Logger.get_logger()
27
-
28
- # demo.launch()
29
- load_dotenv()
30
- MAIN_DIR = os.getenv("MAIN_DIR")
31
 
32
- # Get all the existing papers in the database
33
- all_entries = db.get(include=["metadatas"])
34
- existing_pdfs = [meta.get("source_pdf") for meta in all_entries["metadatas"]]
35
-
36
- # Dataset
37
- dataset = MAIN_DIR + "dataset/"
38
- files = os.listdir(dataset)
39
 
40
- LOGGER.info("Looking for new PDFs to append to our database...")
41
- # Look for new PDFs to append to our database
42
- pdf_files = [
43
- os.path.join(dataset, f)
44
- for f in files
45
- if f.lower().endswith(".pdf") and f not in existing_pdfs
46
- ]
47
 
48
- documents, metadatas = process_documents(pdf_files)
49
- LOGGER.info("Adding new PDFs to the database...")
50
- # Fill in the database
51
- if documents:
52
- db.add(
53
- documents=documents,
54
- ids=[str(i) for i in range(len(documents))],
55
- metadatas=metadatas,
56
  )
57
-
58
- # Prompting
59
- LOGGER.info("Retrieving relevant documents...")
60
- embed_fn.document_mode = False
61
- query = "I need an interesting presentation."
62
- query_oneline = query.replace(
63
- "\n", " "
64
- ) # (Optional) For cleaner input in case of newlines
65
- result = db.query(query_texts=[query], n_results=2)
66
- [documents] = result["documents"]
67
- [metadatas] = result["metadatas"]
68
-
69
-
70
- for passage, metas in zip(documents, metadatas):
71
- passage_oneline = passage.replace("\n", " ")
72
- images_passage = metas["images_passage"]
73
-
74
- prompt += f"PASSAGE: {passage_oneline}\n"
75
- prompt += f"IMAGES: {images_passage}\n"
76
-
77
- # Create a new subfolder
78
- LOGGER.info("💾 Creating new subfolder...")
79
- base_path = MAIN_DIR + "output/"
80
- timestamp = datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
81
- work_dir = os.path.join(base_path, timestamp)
82
- os.makedirs(work_dir, exist_ok=False)
83
- LOGGER.info("📁 Created folder: %s", work_dir)
84
-
85
- # Generate the presentation code
86
- model_name = "gemini-2.0-flash"
87
- # model_name = "gemini-2.5-flash-preview-04-17"
88
- LOGGER.info("⭐️ Generating response...")
89
- answer = client.models.generate_content(model=model_name, contents=prompt)
90
- LOGGER.info("🎲 Generated the response using: %s", model_name)
91
-
92
- # Create a subfolder for graphics
93
- graphics_dir = os.path.join(work_dir, "gfx")
94
- os.makedirs(graphics_dir, exist_ok=True)
95
- LOGGER.info("🌄 Images will be saved to: %s", graphics_dir)
96
-
97
- # Find images, which are used in the presentation
98
- pattern_img = re.compile(
99
- r"doc(?P<doc>[a-zA-Z0-9_]+)_page(?P<page>\d+)_img(?P<img>\d+)_hash(?P<hash>[a-fA-F0-9]{8})\.png"
100
- )
101
- matches_img = pattern_img.finditer(answer.text)
102
-
103
- req_imgs = []
104
- for match in matches_img:
105
- req_img = {
106
- "doc": match.group("doc"),
107
- "page": int(match.group("page")),
108
- "img": int(match.group("img")),
109
- "hash": match.group("hash"),
110
- }
111
- req_imgs.append(req_img)
112
-
113
- # Find figures, which are used in the presentation
114
- pattern_fig = re.compile(
115
- r"doc(?P<doc>[a-zA-Z0-9_]+)_page(?P<page>\d+)_fig(?P<fig>\d+)_hash(?P<hash>[a-fA-F0-9]{8})\.png"
116
- )
117
- matches_fig = pattern_fig.finditer(answer.text)
118
-
119
- req_figs = []
120
- for match in matches_fig:
121
- req_fig = {
122
- "doc": match.group("doc"),
123
- "page": int(match.group("page")),
124
- "fig": int(match.group("fig")),
125
- "hash": match.group("hash"),
126
- }
127
- req_figs.append(req_fig)
128
-
129
- # Save the required graphics
130
- for metadata in metadatas:
131
- save_pdf_images(metadata["pdf_path"], req_imgs, graphics_dir)
132
- save_pdf_figures(metadata["pdf_path"], req_figs, graphics_dir)
133
-
134
- LOGGER.info("Compiling presentation...")
135
- CompilePresentation(answer.text, work_dir)
 
1
+ """RAGnTex: A Retrieval-Augmented Generation for LaTeX Documents"""
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
 
3
+ import os
 
 
 
 
 
 
4
 
5
+ from src import demo, init_telemetry
 
 
 
 
 
 
6
 
7
+ # Setup OpenTelemetry
8
+ _ = init_telemetry()
9
+ # Run Gradio
10
+ if os.getenv("IN_DOCKER") == "true":
11
+ demo.launch(
12
+ server_name="0.0.0.0", # nosec: intentional bind all interfaces for HuggingFace Spaces
13
+ server_port=7860,
 
14
  )
15
+ else:
16
+ demo.launch()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
src/__init__.py CHANGED
@@ -1,20 +1,11 @@
1
- from .database import db, embed_fn
2
- from .processing import process_documents, save_pdf_images, save_pdf_figures, prompt
3
- from .compilation import CompilePresentation
4
- from .services import client
5
- from .utils import upload_files, Logger
6
- # from .interface import demo
7
 
8
  __all__ = [
9
- "db",
10
- "embed_fn",
11
- "process_documents",
12
- "save_pdf_images",
13
- "save_pdf_figures",
14
- "prompt",
15
- "CompilePresentation",
16
- "client",
17
- "upload_files",
18
- "Logger",
19
- # "demo",
20
  ]
 
1
+ """Initialize the package by exposing the main interface and telemetry setup."""
2
+
3
+ from .generator import generate_presentation
4
+ from .interface import demo
5
+ from .telemetry import init_telemetry
 
6
 
7
  __all__ = [
8
+ "demo",
9
+ "init_telemetry",
10
+ "generate_presentation",
 
 
 
 
 
 
 
 
11
  ]
src/compilation/__init__.py CHANGED
@@ -1,3 +1,7 @@
1
- from .latex_compilation import CompilePresentation
2
 
3
- __all__ = ["CompilePresentation"]
 
 
 
 
 
1
+ """This module initializes the latex compilation package."""
2
 
3
+ from .json_to_tex import json_to_tex
4
+ from .latex_compilation import compile_presentation
5
+ from .latex_tools import replace_unicode_greek
6
+
7
+ __all__ = ["compile_presentation", "json_to_tex", "replace_unicode_greek"]
src/compilation/json_to_tex.py ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """This module provides functionality to convert a JSON LLM output to LaTeX Beamer presentation."""
2
+
3
+ import json
4
+
5
+
6
+ def render_content(content: list[str], content_format: str) -> str:
7
+ """Render content in LaTeX format depending is it si itemized list or plain text.
8
+ Args:
9
+ content (list[str]): List of content items.
10
+ content_format (str): Format of the content, either "itemize" or "text".
11
+ Returns:
12
+ str: LaTeX formatted content.
13
+ """
14
+ # Remove blank or whitespace-only entries
15
+ filtered_content = [c.strip() for c in content if c.strip()]
16
+ if content_format == "text":
17
+ return "\n\n".join(filtered_content)
18
+ return (
19
+ "\\begin{itemize}\n"
20
+ + "\n".join(f"\\item {c}" for c in filtered_content)
21
+ + "\n\\end{itemize}"
22
+ )
23
+
24
+
25
+ def json_to_tex(
26
+ presentation_json: str, theme: str, color_theme: str, aspect_ratio: str
27
+ ) -> str:
28
+ """Convert a JSON representation of a presentation to LaTeX Beamer format.
29
+ Args:
30
+ presentation (Dict[str, Any]): JSON LLM output of the presentation.
31
+ theme (str): Beamer theme to use.
32
+ color_theme (str): Beamer color theme to use.
33
+ aspect_ratio (str): Aspect ratio for the presentation, e.g., "16:9", "4:3".
34
+ Returns:
35
+ str: LaTeX Beamer code as a string.
36
+ """
37
+ presentation = json.loads(presentation_json)
38
+ aspect_ratio_num = 169 if aspect_ratio == "16:9" else 43
39
+ header = f"""\\documentclass[aspectratio={aspect_ratio_num}]{{beamer}}
40
+ \\usepackage[utf8]{{inputenc}}
41
+ \\usetheme{{{theme}}}
42
+ \\usecolortheme{{{color_theme}}}
43
+ \\title{{{presentation["title"]}}}
44
+ \\author{{{presentation["author"]}}}
45
+ \\date{{\\today}}
46
+
47
+ \\usepackage{{graphicx}}
48
+
49
+ \\begin{{document}}
50
+
51
+ \\frame{{\\titlepage}}"""
52
+
53
+ slides = []
54
+
55
+ for slide in presentation["slides"]:
56
+ slide_type = slide["type"]
57
+ content = slide.get("content", [])
58
+ content_format = slide.get("content_format", "itemize")
59
+ body = render_content(content, content_format)
60
+
61
+ # if slide_type == "title":
62
+ # continue
63
+
64
+ if slide_type in {"introduction", "summary"}:
65
+ title = "Introduction" if slide_type == "introduction" else "Summary"
66
+ slides.append(
67
+ f"""\\begin{{frame}}
68
+ \\frametitle{{{title}}}
69
+ {body}
70
+ \\end{{frame}}"""
71
+ )
72
+
73
+ elif slide_type == "core_idea":
74
+ core_title = slide.get("title", "Core Idea")
75
+ layout = slide.get("layout", "text_only")
76
+
77
+ if layout == "text_only":
78
+ slides.append(
79
+ f"""\\begin{{frame}}
80
+ \\frametitle{{{core_title}}}
81
+ {body}
82
+ \\end{{frame}}"""
83
+ )
84
+
85
+ elif layout == "two_column":
86
+ img = slide["image"]
87
+ slides.append(
88
+ f"""\\begin{{frame}}
89
+ \\frametitle{{{core_title}}}
90
+ \\begin{{columns}}
91
+ \\begin{{column}}{{0.5\\linewidth}}
92
+ {body}
93
+ \\end{{column}}
94
+ \\begin{{column}}{{0.5\\linewidth}}
95
+ \\center{{\\includegraphics[height=1.0\\textheight, width=1.0\\textwidth, keepaspectratio]{{{img["path"]}}} \\\\ {img["caption"]}}}
96
+ \\end{{column}}
97
+ \\end{{columns}}
98
+ \\end{{frame}}"""
99
+ )
100
+
101
+ elif layout == "single_image":
102
+ img = slide["image"]
103
+ slides.append(
104
+ f"""\\begin{{frame}}
105
+ \\frametitle{{{core_title}}}
106
+ \\center{{\\includegraphics[height=0.5\\textheight, width=0.8\\textwidth, keepaspectratio]{{{img["path"]}}} \\\\ {img["caption"]} \\\\}}
107
+ {body}
108
+ \\end{{frame}}"""
109
+ )
110
+
111
+ footer = "\\end{document}"
112
+
113
+ return "\n\n".join([header] + slides + [footer])
src/compilation/latex_compilation.py CHANGED
@@ -1,11 +1,24 @@
1
- import os
2
- import subprocess
3
- from ..utils import Logger
 
 
 
 
 
 
4
 
5
  LOGGER = Logger.get_logger()
6
 
7
 
8
- def CompilePresentation(latex_code, work_dir):
 
 
 
 
 
 
 
9
  # Remove possible Markdown wrapping around the output
10
  if latex_code.startswith("```latex"):
11
  latex_code = latex_code.split("\n", 1)[1] # Remove the first line
@@ -14,40 +27,66 @@ def CompilePresentation(latex_code, work_dir):
14
 
15
  # Save LaTeX code to file
16
  tex_file = os.path.join(work_dir, "presentation.tex")
17
- with open(tex_file, "w") as f:
18
  f.write(latex_code)
19
 
20
- LOGGER.info("=" * 100)
21
- LOGGER.info("📄 Files in the directory:")
22
 
23
  # List files in the directory using Python (instead of `!ls`)
24
  for file in os.listdir(work_dir):
25
  LOGGER.info(file)
26
 
27
  # Compile with pdflatex (using subprocess instead of `!`)
28
- os.chdir(work_dir) # Change working directory
29
-
30
  try:
31
  # Run pdflatex twice to ensure proper slide enumeration
32
  subprocess.run(
33
  ["pdflatex", "-interaction=nonstopmode", "presentation.tex"],
 
34
  check=True,
35
  stdout=subprocess.PIPE,
36
  stderr=subprocess.PIPE,
37
- )
38
  subprocess.run(
39
  ["pdflatex", "-interaction=nonstopmode", "presentation.tex"],
 
40
  check=True,
41
  stdout=subprocess.PIPE,
42
  stderr=subprocess.PIPE,
43
- )
44
 
45
  except subprocess.CalledProcessError as e:
46
  # Handle error in case of failed LaTeX compilation
47
- LOGGER.error("❌ PDF generation failed. Here's the log: %s", e.stderr.decode())
48
- else:
 
49
  # Check for PDF output
50
- if not os.path.exists("presentation.pdf"):
51
- LOGGER.error("❌ PDF generation failed. No PDF file found.")
52
- else:
53
- LOGGER.info("💾 PDF generated successfully in: %s", work_dir)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """Module for compiling LaTeX presentations into PDF format."""
2
+
3
+ import os # nosec: checked usage, inputs are sanitized
4
+ import subprocess # nosec: checked usage, inputs are sanitized
5
+ from pathlib import Path
6
+
7
+ from langfuse.decorators import langfuse_context, observe
8
+
9
+ from ..telemetry import Logger
10
 
11
  LOGGER = Logger.get_logger()
12
 
13
 
14
+ @observe(name="🧱 compile_presentation")
15
+ def compile_presentation(latex_code, work_dir) -> str:
16
+ """Compile LaTeX code into a PDF presentation.
17
+ Args:
18
+ latex_code (str): LaTeX code to be compiled.
19
+ work_dir (str): Directory where the LaTeX file will be saved and compiled.
20
+ """
21
+ LOGGER.info("Compiling presentation...")
22
  # Remove possible Markdown wrapping around the output
23
  if latex_code.startswith("```latex"):
24
  latex_code = latex_code.split("\n", 1)[1] # Remove the first line
 
27
 
28
  # Save LaTeX code to file
29
  tex_file = os.path.join(work_dir, "presentation.tex")
30
+ with open(tex_file, "w", encoding="utf-8") as f:
31
  f.write(latex_code)
32
 
33
+ LOGGER.info("📄 Files in the directory: %s", work_dir)
 
34
 
35
  # List files in the directory using Python (instead of `!ls`)
36
  for file in os.listdir(work_dir):
37
  LOGGER.info(file)
38
 
39
  # Compile with pdflatex (using subprocess instead of `!`)
 
 
40
  try:
41
  # Run pdflatex twice to ensure proper slide enumeration
42
  subprocess.run(
43
  ["pdflatex", "-interaction=nonstopmode", "presentation.tex"],
44
+ cwd=work_dir,
45
  check=True,
46
  stdout=subprocess.PIPE,
47
  stderr=subprocess.PIPE,
48
+ ) # nosec B603 B607 - fixed command, no untrusted input
49
  subprocess.run(
50
  ["pdflatex", "-interaction=nonstopmode", "presentation.tex"],
51
+ cwd=work_dir,
52
  check=True,
53
  stdout=subprocess.PIPE,
54
  stderr=subprocess.PIPE,
55
+ ) # nosec B603 B607 - fixed command, no untrusted input
56
 
57
  except subprocess.CalledProcessError as e:
58
  # Handle error in case of failed LaTeX compilation
59
+ langfuse_context.update_current_observation(
60
+ output={"pdf.success": False, "pdf.error_log": e.stderr.decode()}
61
+ )
62
  # Check for PDF output
63
+ pdf_file = Path(work_dir) / "presentation.pdf"
64
+ if not pdf_file.exists():
65
+ LOGGER.error(
66
+ "❌ PDF generation failed and no PDF file found. Here's the log: %s",
67
+ e.stderr.decode(),
68
+ )
69
+ return "❌ Presentation compilation failed. Please try again later."
70
+ LOGGER.error(
71
+ "⚠️ PDF generation failed, but PDF file exists. Here's the log: %s",
72
+ e.stderr.decode(),
73
+ )
74
+ return (
75
+ "⚠️ Presentation compilation contains errors. Please cross-check the output."
76
+ )
77
+
78
+ # Success, but still check for PDF output
79
+ pdf_file = Path(work_dir) / "presentation.pdf"
80
+ if not pdf_file.exists():
81
+ langfuse_context.update_current_observation(
82
+ output={
83
+ "pdf.success": False,
84
+ "pdf.error_log": "❌ PDF generation succeded, but no PDF file found.",
85
+ }
86
+ )
87
+ LOGGER.error("❌ PDF generation failed. No PDF file found.")
88
+ return "❌ Presentation compilation failed. Please try again later."
89
+
90
+ langfuse_context.update_current_observation(output={"pdf.success": True})
91
+ LOGGER.info("✅ PDF generated successfully in: %s", work_dir)
92
+ return "⭐️ Presentation generated successfully!"
src/compilation/latex_tools.py ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """This module provides tools to work with LaTeX Beamer presentations."""
2
+
3
+
4
+ def replace_unicode_greek(text):
5
+ """Replace Greek Unicode characters with LaTeX equivalents.
6
+ Args:
7
+ text (str): Input text containing Greek characters.
8
+ Returns:
9
+ str: Text with Greek characters replaced by LaTeX commands.
10
+ """
11
+ greek_map = {
12
+ 'α': r'$\alpha$', 'β': r'$\beta$', 'γ': r'$\gamma$',
13
+ 'δ': r'$\delta$', 'ε': r'$\epsilon$', 'ζ': r'$\zeta$',
14
+ 'η': r'$\eta$', 'θ': r'$\theta$', 'ι': r'$\iota$',
15
+ 'κ': r'$\kappa$', 'λ': r'$\lambda$', 'μ': r'$\mu$',
16
+ 'ν': r'$\nu$', 'ξ': r'$\xi$', 'ο': 'o', 'π': r'$\pi$',
17
+ 'ρ': r'$\rho$', 'σ': r'$\sigma$', 'ς': r'$\varsigma$',
18
+ 'τ': r'$\tau$', 'υ': r'$\upsilon$', 'φ': r'$\phi$',
19
+ 'χ': r'$\chi$', 'ψ': r'$\psi$', 'ω': r'$\omega$',
20
+ 'Α': 'A', 'Β': 'B', 'Γ': r'$\Gamma$', 'Δ': r'$\Delta$',
21
+ 'Ε': 'E', 'Ζ': 'Z', 'Η': 'H', 'Θ': r'$\Theta$',
22
+ 'Ι': 'I', 'Κ': 'K', 'Λ': r'$\Lambda$', 'Μ': 'M',
23
+ 'Ν': 'N', 'Ξ': r'$\Xi$', 'Ο': 'O', 'Π': r'$\Pi$',
24
+ 'Ρ': 'P', 'Σ': r'$\Sigma$', 'Τ': 'T', 'Υ': r'$\Upsilon$',
25
+ 'Φ': r'$\Phi$', 'Χ': 'X', 'Ψ': r'$\Psi$', 'Ω': r'$\Omega$',
26
+ 'ϕ': r'$\varphi$', 'ϑ': r'$\vartheta$', 'ϰ': r'$\varkappa$', 'ϱ': r'$\varrho$',
27
+ }
28
+
29
+ for char, replacement in greek_map.items():
30
+ text = text.replace(char, replacement)
31
+
32
+ return text
src/database/__init__.py CHANGED
@@ -1,3 +1,7 @@
1
- from .database import db, embed_fn
2
 
3
- __all__ = ["db", "embed_fn"]
 
 
 
 
 
1
+ """This module initializes the database package."""
2
 
3
+ from .database import chroma_client, embed_fn
4
+ from .db_manipulation import (clean_db, ingest_files_to_db,
5
+ retrive_files_from_db)
6
+
7
+ __all__ = ["chroma_client", "embed_fn", "ingest_files_to_db", "retrive_files_from_db", "clean_db"]
src/database/database.py CHANGED
@@ -1,8 +1,10 @@
 
 
 
 
 
1
  from google.api_core import retry
2
  from google.genai import types
3
- from google import genai
4
- from chromadb import Documents, EmbeddingFunction, Embeddings
5
- import chromadb
6
 
7
  from ..services import client
8
 
@@ -11,11 +13,13 @@ is_retriable = lambda e: (isinstance(e, genai.errors.APIError) and e.code in {42
11
 
12
 
13
  class GeminiEmbeddingFunction(EmbeddingFunction):
 
 
14
  def __init__(self):
15
  self.document_mode = True
16
 
17
  @retry.Retry(predicate=is_retriable)
18
- def __call__(self, input: Documents) -> Embeddings:
19
  if self.document_mode:
20
  embedding_task = "retrieval_document"
21
  else:
@@ -23,7 +27,7 @@ class GeminiEmbeddingFunction(EmbeddingFunction):
23
 
24
  response = client.models.embed_content(
25
  model="models/text-embedding-004",
26
- contents=input,
27
  config=types.EmbedContentConfig(
28
  task_type=embedding_task,
29
  ),
@@ -31,11 +35,7 @@ class GeminiEmbeddingFunction(EmbeddingFunction):
31
  return [e.values for e in response.embeddings]
32
 
33
 
34
- DB_NAME = "ragntex"
35
-
36
  embed_fn = GeminiEmbeddingFunction()
37
  embed_fn.document_mode = True
38
 
39
  chroma_client = chromadb.Client()
40
- # chroma_client.delete_collection(DB_NAME)
41
- db = chroma_client.get_or_create_collection(name=DB_NAME, embedding_function=embed_fn)
 
1
+ """Database module for user-uploaded files."""
2
+
3
+ import chromadb
4
+ from chromadb import Documents, EmbeddingFunction, Embeddings
5
+ from google import genai
6
  from google.api_core import retry
7
  from google.genai import types
 
 
 
8
 
9
  from ..services import client
10
 
 
13
 
14
 
15
  class GeminiEmbeddingFunction(EmbeddingFunction):
16
+ """A class to handle retrieval embedding functions using Google Gemini API."""
17
+
18
  def __init__(self):
19
  self.document_mode = True
20
 
21
  @retry.Retry(predicate=is_retriable)
22
+ def __call__(self, inputs: Documents) -> Embeddings:
23
  if self.document_mode:
24
  embedding_task = "retrieval_document"
25
  else:
 
27
 
28
  response = client.models.embed_content(
29
  model="models/text-embedding-004",
30
+ contents=inputs,
31
  config=types.EmbedContentConfig(
32
  task_type=embedding_task,
33
  ),
 
35
  return [e.values for e in response.embeddings]
36
 
37
 
 
 
38
  embed_fn = GeminiEmbeddingFunction()
39
  embed_fn.document_mode = True
40
 
41
  chroma_client = chromadb.Client()
 
 
src/database/db_manipulation.py ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """Module for adding and getting PDF files to/from the database."""
2
+
3
+ from typing import List, Mapping, Union
4
+
5
+ from langfuse.decorators import langfuse_context, observe
6
+
7
+ from ..processing import process_documents
8
+ from ..telemetry import Logger
9
+ from .database import chroma_client, embed_fn
10
+
11
+ LOGGER = Logger.get_logger()
12
+
13
+
14
+ @observe(name="𝌊 ingest_files_to_db")
15
+ def ingest_files_to_db(pdf_files, session_id) -> list[str]:
16
+ """Add new files to the database if they are not yet there.
17
+ Args:
18
+ pdf_files (list): List of PDF files to be ingested.
19
+ session_id (str): Unique identifier for the current session.
20
+ """
21
+ # Get or create the database collection
22
+ db = chroma_client.get_or_create_collection(
23
+ name=session_id, embedding_function=embed_fn
24
+ )
25
+
26
+ # Get all the existing entries and extract the known filenames
27
+ all_entries = db.get(include=["metadatas"]) or []
28
+
29
+ if not all_entries:
30
+ LOGGER.info("📂 No existing entries in the database. Ingesting all PDFs.")
31
+
32
+ existing_pdfs = set(
33
+ meta.get("pdf_path")
34
+ for meta in all_entries["metadatas"] # type: ignore[union-attr]
35
+ if meta.get("pdf_path")
36
+ )
37
+
38
+ # Filter out already ingested PDFs
39
+ new_pdfs = [file for file in pdf_files if file not in existing_pdfs]
40
+
41
+ if not new_pdfs:
42
+ LOGGER.info("📂 No new PDFs to ingest. Skipping ingestion.")
43
+ return []
44
+
45
+ # Process new files only
46
+ documents, metadatas, failed = process_documents(new_pdfs)
47
+
48
+ LOGGER.info("➕ Adding %d new PDFs to the database...", len(new_pdfs))
49
+ if documents:
50
+ db.add(
51
+ documents=documents,
52
+ ids=[str(i) for i in range(len(documents))],
53
+ metadatas=metadatas,
54
+ )
55
+ LOGGER.info("✅ Documents ingested successfully.")
56
+
57
+ return failed
58
+
59
+
60
+ Metadata = Mapping[str, Union[str, int, float, bool]]
61
+ DocumentsBatch = List[List[str]]
62
+ MetadataBatch = List[List[Metadata]]
63
+
64
+
65
+ @observe(name="🔍 retrieve_files_from_db")
66
+ def retrive_files_from_db(
67
+ topic: str, session_id: str
68
+ ) -> tuple[DocumentsBatch, MetadataBatch]:
69
+ """Retrieve relevant documents from the database based on the provided topic.
70
+ Args:
71
+ topic (str): The topic for which to retrieve documents.
72
+ session_id (str): Unique identifier for the current session.
73
+ Returns:
74
+ tuple: A 2-element tuple:
75
+ - list[str]: List of retrieved documents.
76
+ - list[dict]: List of metadata associated with the documents.
77
+ """
78
+ # Get or create the database collection
79
+ db = chroma_client.get_or_create_collection(
80
+ name=session_id, embedding_function=embed_fn
81
+ )
82
+ if not db:
83
+ LOGGER.error("❌ Database collection not found or could not be created.")
84
+ return [], []
85
+
86
+ LOGGER.info("🔍 Retrieving relevant documents...")
87
+ embed_fn.document_mode = False
88
+ query_oneline = topic.replace("\n", " ")
89
+
90
+ n_yelded_docs = 2
91
+ result = db.query(query_texts=[query_oneline], n_results=n_yelded_docs)
92
+ langfuse_context.update_current_observation(
93
+ output={
94
+ "input.query_oneline": query_oneline,
95
+ "output.n_results": n_yelded_docs,
96
+ "output.metadatas": result["metadatas"],
97
+ }
98
+ )
99
+ [documents] = result.get("documents") or [[]]
100
+ [metadatas] = result.get("metadatas") or [[]]
101
+
102
+ return [documents], [metadatas]
103
+
104
+
105
+ def clean_db(session_id: str) -> None:
106
+ """Clean the database collection for the given session ID.
107
+ Args:
108
+ session_id (str): Unique identifier for the current session.
109
+ """
110
+ db = chroma_client.get_or_create_collection(
111
+ name=session_id, embedding_function=embed_fn
112
+ )
113
+ if db:
114
+ chroma_client.delete_collection(name=session_id)
115
+ LOGGER.info("🗑️ Cleaning database for session ID: %s", session_id)
116
+ else:
117
+ LOGGER.warning(
118
+ "❌ Database collection not found for session ID: %s", session_id
119
+ )
src/generator.py ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """Module to put it all together and manage the generation of a presentation."""
2
+
3
+ from dataclasses import dataclass
4
+ from typing import Optional
5
+
6
+ from langfuse.decorators import langfuse_context, observe
7
+
8
+ from src.compilation import (compile_presentation, json_to_tex,
9
+ replace_unicode_greek)
10
+ # from datetime import datetime
11
+ from src.database import retrive_files_from_db
12
+ from src.processing import create_output_folder, find_used_gfx
13
+ from src.services import build_prompt, client, generate_with_retry
14
+ from src.telemetry import Logger
15
+
16
+ LOGGER = Logger.get_logger()
17
+
18
+
19
+ @dataclass
20
+ class PresentationConfig:
21
+ """Configuration for the presentation generation."""
22
+
23
+ model_name: str = "gemini-2.0-flash"
24
+ aspect_ratio: str = "16:9"
25
+ theme: str = "default"
26
+ color_theme: str = "default"
27
+ topic: str = ""
28
+
29
+
30
+ @observe(name="🧑‍🎨 generate_presentation")
31
+ def generate_presentation(
32
+ config_dict, session_id
33
+ ) -> tuple[str, Optional[str], Optional[str]]:
34
+ """Main function to generate a presentation based on the provided theme, color, and topic.
35
+
36
+ Args:
37
+ config_dict (dict): Configuration dictionary containing:
38
+ - model_name (str): Name of the model to use for generation.
39
+ - aspect_ratio (str): Aspect ratio for the presentation, e.g., "16:9", "4:3".
40
+ - theme (str): Beamer theme to use.
41
+ - color_theme (str): Beamer color theme to use.
42
+ - topic (str): Topic of the presentation.
43
+ session_id (str): Unique identifier for the current session.
44
+
45
+ Returns:
46
+ tuple:
47
+ A 3-element tuple containing:
48
+ - str: Status message (success or error).
49
+ - Optional[str]: Trace ID for telemetry purposes, or None if unavailable.
50
+ - Optional[str]: Path to the presentation folder, or None if unavailable.
51
+ """
52
+ # Cast dict to dataclass
53
+ try:
54
+ config = PresentationConfig(**config_dict)
55
+ except TypeError as e:
56
+ # Handle missing or extra keys gracefully
57
+ return f"Config error: {str(e)}", None, None
58
+
59
+ trace_id = langfuse_context.get_current_trace_id()
60
+
61
+ [documents], [metadatas] = retrive_files_from_db(config.topic, session_id)
62
+
63
+ prompt = build_prompt(documents, metadatas, config.aspect_ratio)
64
+
65
+ work_dir = create_output_folder(session_id)
66
+ # Generate the presentation code
67
+ model_name = config.model_name
68
+ LOGGER.info("⭐️ Generating response...")
69
+ try:
70
+ answer = generate_with_retry(client, model_name, prompt)
71
+ except Exception as e:
72
+ LOGGER.error("❌ Gemini is not reponding: %s", str(e))
73
+ return f"Gemini error: {str(e)}", trace_id, work_dir
74
+
75
+ langfuse_context.update_current_observation(
76
+ output={
77
+ "model_name": model_name,
78
+ "session_id": session_id,
79
+ "aspect_ratio": config.aspect_ratio,
80
+ "prompt_length": len(prompt),
81
+ "output.response": answer.text,
82
+ }
83
+ )
84
+ LOGGER.info("🎲 Generated the response using: %s", model_name)
85
+
86
+ find_used_gfx(answer, work_dir, metadatas)
87
+ latex_code = json_to_tex(
88
+ answer.text, config.theme, config.color_theme, config.aspect_ratio
89
+ )
90
+ latex_code = replace_unicode_greek(latex_code)
91
+ compilation_status = compile_presentation(latex_code, work_dir)
92
+
93
+ if compilation_status:
94
+ return compilation_status, trace_id, work_dir
95
+ return "❌ Unknown critial error. Please try again later.", trace_id, work_dir
src/interface/__init__.py CHANGED
@@ -1,3 +1,16 @@
 
 
1
  from .gradio_interface import demo
 
 
 
2
 
3
- __all__ = ["demo"]
 
 
 
 
 
 
 
 
 
1
+ """This module initializes the UI package."""
2
+
3
  from .gradio_interface import demo
4
+ from .manage_files import delete_files, download_files, upload_files
5
+ from .session_manager import (check_session_status, create_session,
6
+ with_update_session)
7
 
8
+ __all__ = [
9
+ "demo",
10
+ "upload_files",
11
+ "download_files",
12
+ "delete_files",
13
+ "create_session",
14
+ "with_update_session",
15
+ "check_session_status",
16
+ ]
src/interface/gradio_interface.py ADDED
@@ -0,0 +1,506 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # pylint: disable=no-member, line-too-long
2
+ """Gradio UI for the RAG'n'TeX LaTeX Presentation Generator."""
3
+
4
+ import base64
5
+
6
+ import gradio as gr
7
+
8
+ from ..generator import generate_presentation
9
+ from ..telemetry import submit_feedback
10
+ from .manage_files import download_files, upload_files
11
+ from .session_manager import (check_session_status, create_session,
12
+ with_update_session)
13
+
14
+ SESSION_TIMEOUT = 900
15
+
16
+
17
+ def update_config(key, value, config) -> dict:
18
+ """Helper function to update configuration settings.
19
+ These settings will be later used for generate_presentation function.
20
+
21
+ Args:
22
+ key (str): Configuration key to update.
23
+ value (str): New value for the configuration key.
24
+ config (dict): Current configuration dictionary.
25
+
26
+ Returns:
27
+ dict: Updated configuration dictionary with the new key-value pair.
28
+ """
29
+ config[key] = value
30
+ return config
31
+
32
+
33
+ def upload_and_update_list(
34
+ files: list, uploaded_list, session_id
35
+ ) -> tuple[str, list[str]]:
36
+ """Helper function to handle uploaded documents and update the list of uploaded files.
37
+ Args:
38
+ files (list): List of file-like objects to be uploaded.
39
+ uploaded_list (list): Current list of uploaded file paths.
40
+ session_id (str): Unique identifier for the current session.
41
+ Returns:
42
+ tuple: A 2-element tuple:
43
+ - str: Status message indicating the result of the upload operation.
44
+ - list[str]: Updated list of uploaded file paths."""
45
+
46
+ status, new_paths = upload_files(files, session_id)
47
+ updated_list = uploaded_list + [p for p in new_paths if p not in uploaded_list]
48
+
49
+ return status, updated_list
50
+
51
+
52
+ def generate_iframe(folder_path) -> str:
53
+ """Generate an HTML iframe to display the PDF presentation.
54
+ Args:
55
+ folder_path (str): Path to the folder containing the generated PDF presentation.
56
+ Returns:
57
+ str: HTML string containing the iframe to display the PDF.
58
+ """
59
+
60
+ file_path = f"{folder_path}/presentation.pdf"
61
+ if not file_path:
62
+ return ""
63
+ with open(file_path, "rb") as f:
64
+ base64_pdf = base64.b64encode(f.read()).decode("utf-8")
65
+ pdf_display = f"""
66
+ <div style="position:relative; width:100%; padding-top:76.5%;">
67
+ <iframe src="data:application/pdf;base64,{base64_pdf}#view=FitH"
68
+ style="position:absolute; top:0; left:0; width:100%; height:100%; border:none;">
69
+ </iframe></div>
70
+ """
71
+
72
+ return pdf_display
73
+
74
+
75
+ def encode_image(image_path) -> str:
76
+ """Encode an image file to a base64 data URL.
77
+ Args:
78
+ image_path (str): Path to the image file.
79
+ Returns:
80
+ str: Base64 data URL of the image.
81
+ """
82
+ with open(image_path, "rb") as f:
83
+ data = f.read()
84
+ return f"data:image/png;base64,{base64.b64encode(data).decode()}"
85
+
86
+
87
+ banner_base64 = encode_image("gfx/long_logo.png")
88
+ logo_base64 = encode_image("gfx/icon_logo.png")
89
+ github_base64 = encode_image("gfx/github-mark-white.png")
90
+ favicon_base64 = encode_image("gfx/favicon.ico")
91
+
92
+
93
+ # Custom Gradio theme
94
+ theme = gr.themes.Monochrome(
95
+ neutral_hue=gr.themes.Color(
96
+ c50="#f8f4fe",
97
+ c100="#ede9fe",
98
+ c200="#ddd6fe",
99
+ c300="#c4b5fd",
100
+ c400="#a78bfa",
101
+ c500="#8b5cf6",
102
+ c600="#7c3aed",
103
+ c700="#6d28d9",
104
+ c800="#5b21b6",
105
+ c900="#4c1d95",
106
+ c950="#181e44",
107
+ ),
108
+ ).set(body_background_fill="*neutral_50", body_text_color="*neutral_950")
109
+
110
+
111
+ # Additional JavaScript to handle session management and UI updates
112
+ JS_FUNC = r"""
113
+ () => {
114
+ // Force light theme
115
+ const url = new URL(window.location);
116
+ if (url.searchParams.get('__theme') !== 'light') {
117
+ url.searchParams.set('__theme', 'light');
118
+ window.location.href = url.href;
119
+ }
120
+
121
+ const overlayDiv = document.createElement("div");
122
+ overlayDiv.id = "session-overlay";
123
+ overlayDiv.style = `
124
+ display: flex;
125
+ position: fixed;
126
+ top: 0; left: 0; width: 100vw; height: 100vh;
127
+ justify-content: center;
128
+ align-items: center;
129
+ background-color: rgba(0, 0, 0, 0.7);
130
+ color: white;
131
+ font-size: 1.5rem;
132
+ text-align: center;
133
+ z-index: 999999;
134
+ `;
135
+ overlayDiv.innerHTML = `
136
+ ❌ Session expired.<br/>🔄 Please refresh the page to start over.
137
+ `;
138
+
139
+ const checker = setInterval(() => {
140
+ const checkBtn = document.querySelector('#check-session-button');
141
+ if (checkBtn) checkBtn.click();
142
+
143
+ const statusBox = document.querySelector('#session-status textarea');
144
+ if (!statusBox) return;
145
+
146
+ const value = statusBox.value;
147
+
148
+ if (value === "expired") {
149
+ document.body.appendChild(overlayDiv);
150
+ console.log("Session expired, please refresh the page.");
151
+
152
+ clearInterval(checker);
153
+ }
154
+ }, 5000);
155
+
156
+ }
157
+ """
158
+
159
+
160
+ with gr.Blocks(theme=theme, js=JS_FUNC) as demo:
161
+ # Session management
162
+ session_id_state = gr.State()
163
+ session_timeout = gr.State(SESSION_TIMEOUT)
164
+ status_output = gr.Textbox(visible=False, elem_id="session-status")
165
+ check_btn = gr.Button(visible=False, elem_id="check-session-button")
166
+ check_btn.click(
167
+ fn=check_session_status,
168
+ inputs=[session_id_state],
169
+ outputs=[status_output],
170
+ )
171
+ ###
172
+
173
+ uploaded_files_state = gr.State([])
174
+ presentation_folder_state = gr.State("")
175
+
176
+ # Page title and favicon (doesn't work? Gradio is really agressive)
177
+ gr.HTML(
178
+ f"""
179
+ <script>
180
+ document.title = "RAG'n'TeX Presentation Generator";
181
+
182
+ const links = document.querySelectorAll("link[rel*='icon']");
183
+ links.forEach(link => link.parentNode.removeChild(link));
184
+
185
+ const newFavicon = document.createElement('link');
186
+ newFavicon.rel = 'icon';
187
+ newFavicon.type = 'image/png';
188
+ newFavicon.href = '{favicon_base64}';
189
+ document.head.appendChild(newFavicon);
190
+ </script>
191
+ """,
192
+ elem_id="custom-head-injection",
193
+ visible=False,
194
+ )
195
+
196
+ # Mess with the download-box
197
+ gr.HTML(
198
+ """
199
+ <style>
200
+ #download-box svg.feather-file {
201
+ display: none !important;
202
+ }
203
+ </style>
204
+ """,
205
+ elem_id="custom-style-injection",
206
+ visible=False,
207
+ )
208
+
209
+ gr.Markdown(
210
+ '<h1 style="text-align:center;">AI-powered LaTeX Presentation Generator</h1>'
211
+ )
212
+ with gr.Row():
213
+ with gr.Column(scale=1):
214
+ gr.HTML(
215
+ f"""
216
+ <div style="text-align:center;">
217
+ <img src="{banner_base64}" style="max-width: 100%; height: auto;" />
218
+ </div>
219
+ """
220
+ )
221
+ with gr.Column(scale=1):
222
+ gr.Markdown(
223
+ '<div style="text-align:justify; margin-top: 15px;">Turn your PDFs and ideas into beautiful LaTeX '
224
+ "presentations with AI. Built on top of the <a href=\"https://ragntex.github.io/lore/\">RAG'n'TeX</a>"
225
+ " engine, this tool automatically selects relevant content and compiles well-structured"
226
+ " slides. Ideal for students, researchers, and educators looking to save time and "
227
+ "create topic-focused presentations with minimal effort.</div>"
228
+ )
229
+ gr.Markdown(
230
+ '<div style="text-align:justify; margin-top: 0px;">Just upload your PDFs, click <b>Upload Files</b>, '
231
+ "choose a theme, enter your topic, and hit <b>Generate Presentation</b> — it's that easy!</div>"
232
+ )
233
+ gr.HTML(
234
+ f"""
235
+ <div style="display: flex; flex-direction: row; align-items: center; justify-content: center; gap: 20px; margin-bottom: 12px;">
236
+ <p style="font-size: 16px; line-height: 1; margin: 0;">Please ⭐️ us on GitHub!</p>
237
+ <a href="https://github.com/RAGnTeX/RAGnTeX" target="_blank" style="
238
+ display: inline-flex;
239
+ align-items: center;
240
+ justify-content: center;
241
+ padding: 12px 16px;
242
+ background-color: #4c1d95;
243
+ color: white;
244
+ font-weight: 600;
245
+ font-size: 16px;
246
+ line-height: 1;
247
+ border-radius: 0px;
248
+ text-decoration: none;
249
+ gap: 8px;
250
+ transition: background-color 0.2s ease-in-out;
251
+ border: 2px solid transparent;
252
+ "
253
+ onmouseover="this.style.backgroundColor='#6d28d9'; this.style.border='2px solid #686868';"
254
+ onmouseout="this.style.backgroundColor='#4c1d95'; this.style.border='2px solid transparent';"
255
+ >
256
+ <img src="{github_base64}" alt="GitHub" style="height: 20px;">
257
+ RAG'n'TeX
258
+ </a>
259
+ </div>
260
+ """
261
+ )
262
+
263
+ with gr.Row():
264
+ with gr.Column(scale=1):
265
+ gr.Markdown("## 📂 Step 1: Upload PDF Documents")
266
+ file_input = gr.File(
267
+ file_types=[".pdf"], file_count="multiple", label="Select PDF Files"
268
+ )
269
+ upload_button = gr.Button("Upload Files", variant="primary")
270
+
271
+ gr.Markdown("## 🎨 Step 2: Choose the parameters")
272
+ config_state = gr.State({})
273
+ model_state = gr.State("default")
274
+ model_radio = gr.Radio(
275
+ ["gemini-2.0-flash", "gemini-2.5-flash-preview-05-20"],
276
+ label="LLM",
277
+ info="Choose Gemini version",
278
+ value="gemini-2.0-flash",
279
+ )
280
+ model_radio.change(
281
+ fn=lambda value, config: update_config("model_name", value, config),
282
+ inputs=[model_radio, config_state], # only real components here
283
+ outputs=config_state,
284
+ )
285
+
286
+ aspect_ratio_radio = gr.Radio(
287
+ ["16:9", "4:3"],
288
+ label="Aspect Ratio",
289
+ info="Choose presentation aspect ratio",
290
+ value="16:9",
291
+ )
292
+
293
+ aspect_ratio_radio.change(
294
+ fn=lambda value, config: update_config("aspect_ratio", value, config),
295
+ inputs=[aspect_ratio_radio, config_state], # only real components here
296
+ outputs=config_state,
297
+ )
298
+ presentation_theme_state = gr.State("default")
299
+ theme_dropdown = gr.Dropdown(
300
+ choices=[
301
+ "default",
302
+ "Darmstadt",
303
+ "Malmoe",
304
+ "AnnArbor",
305
+ "Dresden",
306
+ "Marburg",
307
+ "Antibes",
308
+ "Frankfurt",
309
+ "Montpellier",
310
+ "Bergen",
311
+ "Goettingen",
312
+ "PaloAlto",
313
+ "Berkeley",
314
+ "Hannover",
315
+ "Pittsburgh",
316
+ "Berlin",
317
+ "Ilmenau",
318
+ "Rochester",
319
+ "Boadilla",
320
+ "JuanLesPins",
321
+ "Singapore",
322
+ "CambridgeUS",
323
+ "Luebeck",
324
+ "Szeged",
325
+ "Copenhagen",
326
+ "Madrid",
327
+ "Warsaw",
328
+ ],
329
+ value="default", # default option
330
+ label="Choose a presentation theme",
331
+ )
332
+
333
+ theme_dropdown.change(
334
+ fn=lambda value, config: update_config("theme", value, config),
335
+ inputs=[theme_dropdown, config_state], # only real components here
336
+ outputs=config_state,
337
+ )
338
+ color_theme_state = gr.State("default")
339
+ color_dropdown = gr.Dropdown(
340
+ choices=[
341
+ "default",
342
+ "albatross",
343
+ "beetle",
344
+ "crane",
345
+ "dolphin",
346
+ "fly",
347
+ "lily",
348
+ "orchid",
349
+ "rose",
350
+ "seagull",
351
+ "seahorse",
352
+ "sidebartab",
353
+ "structure",
354
+ "whale",
355
+ "wolverine",
356
+ ],
357
+ value="default", # default option
358
+ label="Choose a color theme",
359
+ )
360
+
361
+ color_dropdown.change(
362
+ fn=lambda value, config: update_config("color_theme", value, config),
363
+ inputs=[color_dropdown, config_state], # only real components here
364
+ outputs=config_state,
365
+ )
366
+
367
+ gr.Markdown("## 🧠 Step 3: Enter Presentation Topic")
368
+ topic_input = gr.Textbox(
369
+ label="Presentation Topic",
370
+ placeholder="e.g., Introduction to Quantum Computing",
371
+ lines=1,
372
+ )
373
+ topic_input.change(
374
+ fn=lambda value, config: update_config("topic", value, config),
375
+ inputs=[topic_input, config_state], # only real components here
376
+ outputs=config_state,
377
+ )
378
+ submit_topic_button = gr.Button("Generate Presentation", variant="primary")
379
+
380
+ with gr.Column(scale=1):
381
+ gr.Markdown("## 📄 Uploaded Files")
382
+ upload_output = gr.Textbox(
383
+ label="Upload Status", lines=3, interactive=False
384
+ )
385
+
386
+ gr.Markdown("## 📝 Presentation Processing")
387
+ compilation_status = gr.Textbox(
388
+ label="Compilation Status", lines=1, interactive=False
389
+ )
390
+
391
+ gr.Markdown("## 🎉 Final Presentation")
392
+ pdf_output = gr.File(
393
+ label="Download/View Presentation",
394
+ interactive=False,
395
+ elem_id="download-box",
396
+ )
397
+ pdf_output_viewer = gr.HTML(label="Presentation Preview")
398
+ trace_id_state = gr.State("")
399
+
400
+ gr.Markdown("<br>")
401
+
402
+ with gr.Row():
403
+ with gr.Column(scale=1):
404
+ gr.Markdown("## ⭐ Rate Us")
405
+ rating = gr.Radio(
406
+ choices=["⭐️", "⭐️⭐️", "⭐️⭐️⭐️", "⭐️⭐️⭐️⭐️", "⭐️⭐️⭐️⭐️⭐️"],
407
+ label="How would you rate this app?",
408
+ interactive=True,
409
+ type="value",
410
+ value=None,
411
+ show_label=True,
412
+ container=True,
413
+ )
414
+ feedback_comment = gr.Textbox(
415
+ label="Give us a comment (optional)",
416
+ placeholder="What did you like or what can be improved?",
417
+ lines=3,
418
+ max_lines=5,
419
+ interactive=True,
420
+ )
421
+ submit_feedback_button = gr.Button("Submit Feedback", variant="secondary")
422
+ feedback_output = gr.Textbox(
423
+ label="Feedback Status", interactive=False, lines=2
424
+ )
425
+
426
+ with gr.Column(scale=1):
427
+ gr.Markdown("## 🙌 Credits")
428
+ gr.HTML(
429
+ f"""
430
+ <div style="display: flex; flex-direction: row; gap: 24px; margin-top: 0; align-items: flex-start;">
431
+
432
+ <!-- Left Column -->
433
+ <div style="flex: 1;">
434
+ <div style="text-align:justify; margin-top: 0px;">
435
+ <p> RAG'n'TeX is a collaborative effort by a team of developers and researchers driven
436
+ to make the everyday work of scientists and professionals easier. By combining LaTeX with AI,
437
+ we create smart tools that help you build polished, accurate presentations quickly and smoothly.
438
+ Our goal is to make the process simple and stress-free, so you can spend more time on what really matters.</p>
439
+ <br>
440
+ <p>📄 Licensed under the <a href="https://opensource.org/licenses/MIT" target="_blank" rel="noopener noreferrer">MIT License</a></p>
441
+ <br>
442
+ <h4>Authors:</h4>
443
+ <ul style="padding-left: 20px; margin-top: 4px;">
444
+ <li>👩🏻‍💻 <a href="https://github.com/AnnaErsh" target="_blank">Anna Ershova</a></li>
445
+ <li>👨🏼‍🔬 <a href="https://github.com/KajetanNiewczas" target="_blank">Kajetan Niewczas</a></li>
446
+ </ul>
447
+ </div>
448
+ </div>
449
+
450
+ <!-- Right Column -->
451
+ <div style="flex: 1; display: flex; justify-content: flex-end; align-items: center;">
452
+ <img src="{logo_base64}" alt="Illustration" style="max-width: 100%; border-radius: 8px;">
453
+ </div>
454
+ </div>
455
+ """
456
+ )
457
+
458
+ # Event handlers
459
+
460
+ demo.load(
461
+ fn=create_session,
462
+ inputs=session_timeout,
463
+ outputs=session_id_state,
464
+ )
465
+
466
+ upload_button.click(
467
+ fn=with_update_session(upload_and_update_list),
468
+ inputs=[file_input, uploaded_files_state, session_id_state],
469
+ outputs=[upload_output, uploaded_files_state],
470
+ )
471
+
472
+ trace_id_state = gr.State("")
473
+ submit_topic_button.click(
474
+ fn=with_update_session(generate_presentation),
475
+ inputs=[
476
+ config_state,
477
+ session_id_state,
478
+ ],
479
+ outputs=[compilation_status, trace_id_state, presentation_folder_state],
480
+ )
481
+
482
+ compilation_status.change(
483
+ fn=download_files,
484
+ inputs=[
485
+ presentation_folder_state,
486
+ session_id_state,
487
+ ],
488
+ outputs=pdf_output,
489
+ )
490
+
491
+ pdf_output.change(
492
+ fn=generate_iframe,
493
+ inputs=presentation_folder_state,
494
+ outputs=pdf_output_viewer,
495
+ )
496
+
497
+ submit_feedback_button.click(
498
+ fn=with_update_session(submit_feedback),
499
+ inputs=[
500
+ rating,
501
+ feedback_comment,
502
+ trace_id_state,
503
+ session_id_state,
504
+ ],
505
+ outputs=feedback_output,
506
+ )
src/interface/manage_files.py ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """Module to handle file uploads and downloads via UI."""
2
+
3
+ import re
4
+ import shutil
5
+ import zipfile
6
+ from pathlib import Path
7
+
8
+ from ..database import ingest_files_to_db
9
+ from ..telemetry.logging_utils import Logger
10
+
11
+ LOGGER = Logger.get_logger()
12
+
13
+
14
+ def upload_files(files: list, session_id: str) -> tuple[str, list[str]]:
15
+ """Upload files and save them in the temporary directory.
16
+ Args:
17
+ files (list): List of file-like objects to be uploaded.
18
+ session_id (str): Unique identifier for the current session.
19
+ Returns:
20
+ str: Message indicating the result of the upload operation.
21
+ list[str]: List of saved file paths.
22
+ """
23
+ if not files:
24
+ return "❌ No new files selected", []
25
+
26
+ upload_dir = Path.cwd() / "tmp" / session_id
27
+ upload_dir.mkdir(parents=True, exist_ok=True)
28
+
29
+ messages = []
30
+ saved_paths = []
31
+
32
+ for file in files:
33
+ try:
34
+ temp_path = Path(file.name)
35
+ # Prepare the safe file name
36
+ safe_name = re.sub(r"[^a-zA-Z0-9_]", "_", temp_path.stem)
37
+ safe_name = re.sub(r"_+", "_", safe_name)
38
+ file_name = f"{safe_name}{temp_path.suffix}"
39
+ target_path = upload_dir / file_name
40
+
41
+ if target_path.exists():
42
+ messages.append(f"⚠️ File already exists: {file_name}")
43
+ continue
44
+
45
+ if not temp_path.exists():
46
+ messages.append(f"❌ File not found: {file_name}")
47
+ continue
48
+
49
+ shutil.move(str(temp_path), str(target_path))
50
+ messages.append(f"✅ Uploaded new file: {file_name}")
51
+ LOGGER.info("📤 Uploaded file %s to %s", file_name, upload_dir)
52
+
53
+ saved_paths.append(str(target_path))
54
+
55
+ except (FileNotFoundError, PermissionError, shutil.Error, OSError) as e:
56
+ LOGGER.error("❌ Error uploading %s", file_name, exc_info=e)
57
+ messages.append(f"❌ Error uploading {file_name}: {e}")
58
+
59
+ if saved_paths:
60
+ failed = ingest_files_to_db(saved_paths, session_id)
61
+ if failed:
62
+ messages.append(
63
+ f"⚠️ Failed to process files: {', '.join(Path(f).name for f in failed)}"
64
+ )
65
+ LOGGER.warning(
66
+ "⚠️ Failed to process files: %s", ", ".join(Path(f).name for f in failed)
67
+ )
68
+ else:
69
+ messages.append("🛠️ All files processed successfully!")
70
+ LOGGER.info("🛠️ All files processed successfully!")
71
+
72
+ return "\n".join(messages), saved_paths
73
+
74
+
75
+ def download_files(folder_path, session_id) -> str:
76
+ """Prepare downloadable zip archive from the specified folder path."""
77
+
78
+ if not folder_path:
79
+ return "No folder path provided."
80
+
81
+ folder = Path(folder_path)
82
+ zip_path = Path.cwd() / "tmp" / session_id / "presentation.zip"
83
+
84
+ with zipfile.ZipFile(zip_path, "w", zipfile.ZIP_DEFLATED) as zipf:
85
+ for file in folder.rglob("*"):
86
+ if file.is_file() and file != zip_path:
87
+ arcname = Path("presentation") / file.relative_to(folder)
88
+ zipf.write(file, arcname=arcname)
89
+ LOGGER.info("📦 Created zip archive at %s", zip_path)
90
+
91
+ return str(zip_path)
92
+
93
+
94
+ def delete_files(session_id: str) -> None:
95
+ """Delete all files in the specified session's temporary directory."""
96
+
97
+ upload_dir = Path.cwd() / "tmp" / session_id
98
+
99
+ if not upload_dir.exists():
100
+ return
101
+
102
+ try:
103
+ shutil.rmtree(upload_dir)
104
+ LOGGER.info("🗑️ Deleted folder: %s", upload_dir)
105
+ return
106
+ except Exception as e:
107
+ LOGGER.error("❌ Error deleting: %s", upload_dir, exc_info=e)
108
+ return
src/interface/session_manager.py ADDED
@@ -0,0 +1,123 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """Session Manager Module"""
2
+
3
+ import threading
4
+ import time
5
+ import uuid
6
+ from functools import wraps
7
+ from typing import Any, Callable
8
+
9
+ from ..database import clean_db
10
+ from ..telemetry.logging_utils import Logger
11
+ from .manage_files import delete_files
12
+
13
+ LOGGER = Logger.get_logger()
14
+
15
+
16
+ # Global session manager
17
+ session_data = {}
18
+
19
+
20
+ def create_session(timeout=600) -> str:
21
+ """Create a new session and return its ID.
22
+ This function generates a unique session ID, initializes the session data
23
+ with the current timestamp, and starts a background thread to monitor
24
+ the session for inactivity. If the session is inactive for a specified
25
+ timeout period, it will be automatically cleaned up.
26
+
27
+ Returns:
28
+ str: The unique session ID.
29
+ """
30
+
31
+ # Create new session
32
+ session_id = str(uuid.uuid4())
33
+ session_data[session_id] = {
34
+ "last_active": time.time(),
35
+ }
36
+ LOGGER.info("🪪 Created new session with ID: %s", session_id)
37
+
38
+ # Start clean-up watcher
39
+ threading.Thread(
40
+ target=watch_session, args=(session_id, timeout), daemon=True
41
+ ).start()
42
+
43
+ return session_id
44
+
45
+
46
+ def watch_session(session_id, timeout=600) -> None:
47
+ """Clean up if no activity for `timeout` seconds.
48
+ This function runs in a separate thread and checks if the session
49
+ is still active. If the session has not been active for the specified
50
+ timeout period, it deletes the session data.
51
+
52
+ Args:
53
+ session_id (str): The ID of the session to watch.
54
+ timeout (int): The time in seconds after which the session is considered inactive.
55
+ """
56
+
57
+ while True:
58
+ time.sleep(30)
59
+ session = session_data.get(session_id)
60
+ if not session:
61
+ break
62
+ if time.time() - session["last_active"] > timeout:
63
+ LOGGER.info("🕒 Session expired, ID: %s", session_id)
64
+ del session_data[session_id]
65
+ time.sleep(10)
66
+ delete_files(session_id)
67
+ clean_db(session_id)
68
+ LOGGER.info("🧹 Deleted session with ID: %s", session_id)
69
+ break
70
+
71
+
72
+ def update_session(session_id) -> None:
73
+ """Update the last active time for a session.
74
+ This function checks if the session ID exists in the session data
75
+ and updates its last active timestamp.
76
+
77
+ Args:
78
+ session_id (str): The ID of the session to update.
79
+ """
80
+
81
+ if session_id in session_data:
82
+ session_data[session_id]["last_active"] = time.time()
83
+
84
+
85
+ def with_update_session(fn: Callable[..., Any]) -> Callable[..., Any]:
86
+ """Decorator to update session activity.
87
+ This decorator checks if the session ID is provided in the arguments
88
+ and updates the session's last active time.
89
+
90
+ Args:
91
+ fn (callable): The function to wrap.
92
+
93
+ Returns:
94
+ callable: The wrapped function that updates the session activity.
95
+ """
96
+
97
+ @wraps(fn)
98
+ def wrapper(*args, **kwargs):
99
+ try:
100
+ session_id = kwargs.get("session_id")
101
+ if session_id is None and args:
102
+ session_id = args[-1]
103
+ if session_id in session_data:
104
+ update_session(session_id)
105
+ except Exception as e:
106
+ LOGGER.error("❌ Failed to upadate session ID: %s", session_id, exc_info=e)
107
+ return fn(*args, **kwargs)
108
+
109
+ return wrapper
110
+
111
+
112
+ def check_session_status(session_id: str) -> str:
113
+ """Check if a session is active or expired.
114
+ This function checks if the provided session ID exists in the session data.
115
+ If it exists, the session is considered active; otherwise, it is expired.
116
+
117
+ Args:
118
+ session_id (str): The ID of the session to check.
119
+
120
+ Returns:
121
+ str: "active" if the session is active, "expired" if it is not.
122
+ """
123
+ return "active" if session_id in session_data else "expired"