# alexlawren/rag-django-project-lawrentiev architecture

> Architecture diagram of the GitHub repository alexlawren/rag-django-project-lawrentiev, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/alexlawren/rag-django-project-lawrentiev
- Repository: https://github.com/alexlawren/rag-django-project-lawrentiev

## Overview

This is a local Django document-question-answering assistant. A user uploads a document, which the web view passes through parsing, sentence-window chunking, embedding, and persistent ChromaDB indexing. The active collection is tracked in the session; a later question is embedded for vector retrieval, candidates are cross-encoder re-ranked, and their context windows are used to prompt local Qwen generation. The page displays the answer and retrieved context. README documents these capabilities; implementation details below are grounded in the sampled views and RAG module.

Main entry points: [`core/views.py`](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/views.py) (Request handler), [`core/rag_logic.py`](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py) (Vector candidate search), [`config/urls.py`](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/config/urls.py) (Django routing), [`core/urls.py`](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/urls.py) (App route)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_web["Web interaction"]
  node_django_routing["Django routing<br/>[urls.py]"]
  node_app_route["App route<br/>[urls.py]"]
  node_request_handler["Request handler<br/>[views.py]"]
  node_document_form["Upload/question form<br/>[forms.py]"]
  node_page["Assistant page<br/>[main_page.html]"]
end

subgraph group_ingest["Document ingestion"]
  node_document_parser["Document parsing<br/>[rag_logic.py]"]
  node_sentence_splitter["Sentence segmentation<br/>[rag_logic.py]"]
  node_window_chunker["Sentence windows<br/>[rag_logic.py]"]
  node_embedding["Sentence embeddings<br/>[rag_logic.py]"]
  node_index_writer["Chunk indexing<br/>[rag_logic.py]"]
end

subgraph group_retrieve["Retrieval and generation"]
  node_vector_search["Vector candidate search<br/>[rag_logic.py]"]
  node_reranker["Cross-encoder reranker<br/>[rag_logic.py]"]
  node_context_windows["Context windows<br/>[rag_logic.py]"]
  node_prompt_builder["Grounded prompt<br/>[rag_logic.py]"]
  node_qwen["Local Qwen generation<br/>[rag_logic.py]"]
end

subgraph group_state["Persistent state"]
  node_vector_store[("ChromaDB collection")]
  node_session_state["Active document session<br/>[views.py]"]
end

node_user(("User"))

node_user -->|"opens app"| node_django_routing
node_django_routing -->|"includes routes"| node_app_route
node_app_route -->|"routes request"| node_request_handler
node_request_handler -->|"validates input"| node_document_form
node_request_handler -->|"reads and updates"| node_session_state
node_request_handler -->|"processes upload"| node_document_parser
node_document_parser -->|"segments text"| node_sentence_splitter
node_sentence_splitter -->|"forms windows"| node_window_chunker
node_request_handler -->|"indexes chunks"| node_index_writer
node_window_chunker -->|"provides chunks"| node_index_writer
node_index_writer -->|"encodes sentences"| node_embedding
node_index_writer -->|"writes vectors"| node_vector_store
node_request_handler -->|"searches question"| node_vector_search
node_vector_search -->|"encodes query"| node_embedding
node_vector_search -->|"queries candidates"| node_vector_store
node_vector_search -->|"ranks candidates"| node_reranker
node_reranker -->|"selects windows"| node_context_windows
node_request_handler -->|"builds answer context"| node_prompt_builder
node_context_windows -->|"supplies context"| node_prompt_builder
node_prompt_builder -->|"generates answer"| node_qwen
node_request_handler -->|"renders answer"| node_page
node_context_windows -->|"displays retrieved context"| node_page

click node_django_routing "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/config/urls.py"
click node_app_route "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/urls.py"
click node_request_handler "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/views.py"
click node_document_form "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/forms.py"
click node_page "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/templates/core/main_page.html"
click node_document_parser "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_sentence_splitter "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_window_chunker "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_embedding "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_index_writer "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_vector_store "https://github.com/alexlawren/rag-django-project-lawrentiev/tree/main/chroma_db"
click node_session_state "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/views.py"
click node_vector_search "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_reranker "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_context_windows "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_prompt_builder "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"
click node_qwen "https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_django_routing,node_app_route,node_request_handler,node_document_form,node_page,node_user toneBlue
class node_document_parser,node_sentence_splitter,node_window_chunker,node_embedding,node_index_writer toneAmber
class node_vector_search,node_reranker,node_context_windows,node_prompt_builder,node_qwen toneMint
class node_vector_store,node_session_state toneRose
```

## Components

### Web interaction

- [Django routing](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/config/urls.py): `config/urls.py`
- [App route](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/urls.py): `core/urls.py`
- [Request handler](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/views.py): `core/views.py`
- [Upload/question form](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/forms.py): `core/forms.py`
- [Assistant page](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/templates/core/main_page.html): `core/templates/core/main_page.html`

### Document ingestion

- [Document parsing](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`
- [Sentence segmentation](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`
- [Sentence windows](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`
- [Sentence embeddings](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`
- [Chunk indexing](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`

### Retrieval and generation

- [Vector candidate search](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`
- [Cross-encoder reranker](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`
- [Context windows](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`
- [Grounded prompt](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`
- [Local Qwen generation](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/rag_logic.py): `core/rag_logic.py`

### Persistent state

- [ChromaDB collection](https://github.com/alexlawren/rag-django-project-lawrentiev/tree/main/chroma_db): `chroma_db`
- [Active document session](https://github.com/alexlawren/rag-django-project-lawrentiev/blob/main/core/views.py): `core/views.py`

### Other components

- **User**

## Connections

- User → Django routing: opens app (evidence: `README.md`)
- Django routing → App route: includes routes (evidence: `config/urls.py`)
- App route → Request handler: routes request (evidence: `core/urls.py`)
- Request handler → Upload/question form: validates input (evidence: `core/views.py`)
- Request handler → Active document session: reads and updates (evidence: `core/views.py`)
- Request handler → Document parsing: processes upload (evidence: `core/views.py`)
- Document parsing → Sentence segmentation: segments text (evidence: `core/rag_logic.py`)
- Sentence segmentation → Sentence windows: forms windows (evidence: `core/rag_logic.py`)
- Request handler → Chunk indexing: indexes chunks (evidence: `core/views.py`)
- Sentence windows → Chunk indexing: provides chunks (evidence: `core/rag_logic.py`)
- Chunk indexing → Sentence embeddings: encodes sentences (evidence: `core/rag_logic.py`)
- Chunk indexing → ChromaDB collection: writes vectors (evidence: `core/rag_logic.py`)
- Request handler → Vector candidate search: searches question (evidence: `core/views.py`)
- Vector candidate search → Sentence embeddings: encodes query (evidence: `core/rag_logic.py`)
- Vector candidate search → ChromaDB collection: queries candidates (evidence: `core/rag_logic.py`)
- Vector candidate search → Cross-encoder reranker: ranks candidates (evidence: `core/rag_logic.py`)
- Cross-encoder reranker → Context windows: selects windows (evidence: `core/rag_logic.py`)
- Request handler → Grounded prompt: builds answer context (evidence: `core/views.py`)
- Context windows → Grounded prompt: supplies context (evidence: `core/rag_logic.py`)
- Grounded prompt → Local Qwen generation: generates answer (evidence: `core/rag_logic.py`)
- Request handler → Assistant page: renders answer (evidence: `core/views.py`)
- Context windows → Assistant page: displays retrieved context (evidence: `README.md`)
