discourse-ai/assets/javascripts/discourse/admin/models/ai-persona.js
Sam 5e80f93e4c
FEATURE: PDF support for rag pipeline (#1118)
This PR introduces several enhancements and refactorings to the AI Persona and RAG (Retrieval-Augmented Generation) functionalities within the discourse-ai plugin. Here's a breakdown of the changes:

**1. LLM Model Association for RAG and Personas:**

-   **New Database Columns:** Adds `rag_llm_model_id` to both `ai_personas` and `ai_tools` tables. This allows specifying a dedicated LLM for RAG indexing, separate from the persona's primary LLM.  Adds `default_llm_id` and `question_consolidator_llm_id` to `ai_personas`.
-   **Migration:**  Includes a migration (`20250210032345_migrate_persona_to_llm_model_id.rb`) to populate the new `default_llm_id` and `question_consolidator_llm_id` columns in `ai_personas` based on the existing `default_llm` and `question_consolidator_llm` string columns, and a post migration to remove the latter.
-   **Model Changes:**  The `AiPersona` and `AiTool` models now `belong_to` an `LlmModel` via `rag_llm_model_id`. The `LlmModel.proxy` method now accepts an `LlmModel` instance instead of just an identifier.  `AiPersona` now has `default_llm_id` and `question_consolidator_llm_id` attributes.
-   **UI Updates:**  The AI Persona and AI Tool editors in the admin panel now allow selecting an LLM for RAG indexing (if PDF/image support is enabled).  The RAG options component displays an LLM selector.
-   **Serialization:** The serializers (`AiCustomToolSerializer`, `AiCustomToolListSerializer`, `LocalizedAiPersonaSerializer`) have been updated to include the new `rag_llm_model_id`, `default_llm_id` and `question_consolidator_llm_id` attributes.

**2. PDF and Image Support for RAG:**

-   **Site Setting:** Introduces a new hidden site setting, `ai_rag_pdf_images_enabled`, to control whether PDF and image files can be indexed for RAG. This defaults to `false`.
-   **File Upload Validation:** The `RagDocumentFragmentsController` now checks the `ai_rag_pdf_images_enabled` setting and allows PDF, PNG, JPG, and JPEG files if enabled.  Error handling is included for cases where PDF/image indexing is attempted with the setting disabled.
-   **PDF Processing:** Adds a new utility class, `DiscourseAi::Utils::PdfToImages`, which uses ImageMagick (`magick`) to convert PDF pages into individual PNG images. A maximum PDF size and conversion timeout are enforced.
-   **Image Processing:** A new utility class, `DiscourseAi::Utils::ImageToText`, is included to handle OCR for the images and PDFs.
-   **RAG Digestion Job:** The `DigestRagUpload` job now handles PDF and image uploads. It uses `PdfToImages` and `ImageToText` to extract text and create document fragments.
-   **UI Updates:**  The RAG uploader component now accepts PDF and image file types if `ai_rag_pdf_images_enabled` is true. The UI text is adjusted to indicate supported file types.

**3. Refactoring and Improvements:**

-   **LLM Enumeration:** The `DiscourseAi::Configuration::LlmEnumerator` now provides a `values_for_serialization` method, which returns a simplified array of LLM data (id, name, vision_enabled) suitable for use in serializers. This avoids exposing unnecessary details to the frontend.
-   **AI Helper:** The `AiHelper::Assistant` now takes optional `helper_llm` and `image_caption_llm` parameters in its constructor, allowing for greater flexibility.
-   **Bot and Persona Updates:** Several updates were made across the codebase, changing the string based association to a LLM to the new model based.
-   **Audit Logs:** The `DiscourseAi::Completions::Endpoints::Base` now formats raw request payloads as pretty JSON for easier auditing.
- **Eval Script:** An evaluation script is included.

**4. Testing:**

-    The PR introduces a new eval system for LLMs, this allows us to test how functionality works across various LLM providers. This lives in `/evals`
2025-02-14 12:15:07 +11:00

171 lines
4.2 KiB
JavaScript

import { tracked } from "@glimmer/tracking";
import { ajax } from "discourse/lib/ajax";
import RestModel from "discourse/models/rest";
const CREATE_ATTRIBUTES = [
"id",
"name",
"description",
"tools",
"system_prompt",
"allowed_group_ids",
"enabled",
"system",
"priority",
"top_p",
"temperature",
"user_id",
"default_llm_id",
"force_default_llm",
"user",
"max_context_posts",
"vision_enabled",
"vision_max_pixels",
"rag_uploads",
"rag_chunk_tokens",
"rag_chunk_overlap_tokens",
"rag_conversation_chunks",
"rag_llm_model_id",
"question_consolidator_llm_id",
"allow_chat",
"tool_details",
"forced_tool_count",
"allow_personal_messages",
"allow_topic_mentions",
"allow_chat_channel_mentions",
"allow_chat_direct_messages",
];
const SYSTEM_ATTRIBUTES = [
"id",
"allowed_group_ids",
"enabled",
"system",
"priority",
"tools",
"user_id",
"default_llm_id",
"force_default_llm",
"user",
"max_context_posts",
"vision_enabled",
"vision_max_pixels",
"rag_uploads",
"rag_chunk_tokens",
"rag_chunk_overlap_tokens",
"rag_conversation_chunks",
"rag_llm_model_id",
"question_consolidator_llm_id",
"tool_details",
"allow_personal_messages",
"allow_topic_mentions",
"allow_chat_channel_mentions",
"allow_chat_direct_messages",
];
class ToolOption {
@tracked value = null;
}
export default class AiPersona extends RestModel {
// this code is here to convert the wire schema to easier to work with object
// on the wire we pass in/out tools as an Array.
// [[ToolName, {option1: value, option2: value}, force], ToolName2, ToolName3]
// So we rework this into a "tools" property and nested toolOptions
init(properties) {
this.forcedTools = [];
if (properties.tools) {
properties.tools = properties.tools.map((tool) => {
if (typeof tool === "string") {
return tool;
} else {
let [toolId, options, force] = tool;
for (let optionId in options) {
if (!options.hasOwnProperty(optionId)) {
continue;
}
this.getToolOption(toolId, optionId).value = options[optionId];
}
if (force) {
this.forcedTools.push(toolId);
}
return toolId;
}
});
}
super.init(properties);
this.tools = properties.tools;
}
async createUser() {
const result = await ajax(
`/admin/plugins/discourse-ai/ai-personas/${this.id}/create-user.json`,
{
type: "POST",
}
);
this.user = result.user;
this.user_id = this.user.id;
return this.user;
}
getToolOption(toolId, optionId) {
this.toolOptions ||= {};
this.toolOptions[toolId] ||= {};
return (this.toolOptions[toolId][optionId] ||= new ToolOption());
}
populateToolOptions(attrs) {
if (!attrs.tools) {
return;
}
let toolsWithOptions = [];
attrs.tools.forEach((toolId) => {
if (typeof toolId !== "string") {
toolId = toolId[0];
}
let force = this.forcedTools.includes(toolId);
if (this.toolOptions && this.toolOptions[toolId]) {
let options = this.toolOptions[toolId];
let optionsWithValues = {};
for (let optionId in options) {
if (!options.hasOwnProperty(optionId)) {
continue;
}
let option = options[optionId];
optionsWithValues[optionId] = option.value;
}
toolsWithOptions.push([toolId, optionsWithValues, force]);
} else {
toolsWithOptions.push([toolId, {}, force]);
}
});
attrs.tools = toolsWithOptions;
}
updateProperties() {
let attrs = this.system
? this.getProperties(SYSTEM_ATTRIBUTES)
: this.getProperties(CREATE_ATTRIBUTES);
attrs.id = this.id;
this.populateToolOptions(attrs);
return attrs;
}
createProperties() {
let attrs = this.getProperties(CREATE_ATTRIBUTES);
this.populateToolOptions(attrs);
return attrs;
}
workingCopy() {
let attrs = this.getProperties(CREATE_ATTRIBUTES);
this.populateToolOptions(attrs);
const persona = AiPersona.create(attrs);
persona.forcedTools = (this.forcedTools || []).slice();
persona.forced_tool_count = this.forced_tool_count || -1;
return persona;
}
}